Water Distribution System Calibration Report Security Deliverables/HSD... · Utility Operational...

Water Distribution System Calibration Report

Developed by The University of Kentucky and KYPIPE LLC

Prepared for the National Institute of Hometown Security

368 N. Hwy 27 Somerset, KY 42503

November 19, 2012

This research was funded through funds provided by the Department of Homeland Security, administered by the National Institute for Hometown Security Kentucky Critical Infrastructure

Protection program, under OTA # HSHQDC-07-3-00005, Subcontract # 02-10-UK.

Studying Distribution System Hydraulics and Flow Dynamics to Improve Water Utility Operational Decision Making Revision Date: 01 September 12 Water Distribution System Calibration Report

ii

Water Distribution System Calibration Report Project Title: Studying Distribution System Hydraulics and Flow Dynamics to Improve Water

Utility Operational Decision Making

Water Distribution System: Nicholasville, Kentucky Project No.: 02-10-UK Grant No.: HSHQDC-07-3-00005 Organization: University of Kentucky Principal Investigator: Lindell Ormsbee _______________________ _____________

Signature Date

Field Support L. Sebastian Bryson _______________________ _____________

Signature Date

Scott Yost _______________________ _____________

Signature Date

City of Nicholasville Water Tom Calkins Water Utility Director _______________________ _____________

Signature Date


iii

Table of Contents Table of Contents ........................................................................................................................... iii List of Tables ................................................................................................................................. vi List of Figures ............................................................................................................................... vii List of Abbreviations ................................................................................................................... viii 1.0 Introduction ............................................................................................................................... 1

1.1 Project Background ............................................................................................................... 1

1.2 Purpose of Project ................................................................................................................. 2

1.3 Project Scope ........................................................................................................................ 3

1.4 Project Management ............................................................................................................. 4

1.4.1 Distribution List ............................................................................................................. 4

1.4.2 Project Organization ...................................................................................................... 6

2.0 Existing System ........................................................................................................................ 8

2.1 General .................................................................................................................................. 8

2.2 Supply Source and Storage ................................................................................................... 9

2.3 Water Treatment Plant ........................................................................................................ 10

2.4 Distribution System Piping ................................................................................................. 12

3.0 Data Collection ....................................................................................................................... 19

3.1 C-Factor Test ...................................................................................................................... 19

3.1.1 C-Factor Sites............................................................................................................... 19

3.1.2 C-Factor Test Procedure .............................................................................................. 23

3.1.3 C-Factor Results........................................................................................................... 23

3.1.4 C-Factor Sensitivity Analysis ...................................................................................... 24

3.2 Fire Flow Test ..................................................................................................................... 25

3.2.1 Fire Flow Sites ............................................................................................................. 26

3.2.2 Fire Flow Test Procedure ............................................................................................. 28

3.2.3 Fire Flow Results ......................................................................................................... 29

3.3 Boundary Conditions Collection......................................................................................... 30

3.4 Data Collection Equipment ................................................................................................. 31

3.5 Data Collection Safety Procedures ..................................................................................... 31

3.5.1 Communication and Contingencies ............................................................................. 31

3.5.2 Health and Safety Issues .............................................................................................. 32


iv

3.6 Documentation and Records ............................................................................................... 32

3.7 Quality Control for C-Factor Testing ................................................................................. 33

3.7.1 Review of Construction Records ............................................................................ 33

3.7.2 Pressure Gage Calibration and Validation .............................................................. 33

3.7.3 Duplicate Pressure Observations ............................................................................ 33

3.7.4 Adequate Hydrant Discharge .................................................................................. 34

3.8 Quality Control for Fire Flow Tests.................................................................................... 34

3.8.1 Adequate Hydrant Discharge .................................................................................. 34

3.8.2 Discharge Measurement.......................................................................................... 34

3.8.3 Fire Flow Test Validation ....................................................................................... 35

4.0 Distribution System Model ..................................................................................................... 36

4.1 General ................................................................................................................................ 36

4.2 Development of System Schematic .................................................................................... 36

4.2.1 General Procedure ........................................................................................................ 36

4.2.2 Elevation Data .............................................................................................................. 36

4.2.3 Facilities Data .............................................................................................................. 37

4.2.4 Connectivity Errors ...................................................................................................... 38

4.3 Development of Demand .................................................................................................... 38

4.3.1 Demand Allocation ...................................................................................................... 38

4.3.2 System Demand ........................................................................................................... 40

4.3.3 Diurnal Demand Pattern .............................................................................................. 41

4.3.4 System Losses .............................................................................................................. 43

4.4 Pipe Friction Losses ............................................................................................................ 44

4.4.1 Hazen-Williams Equation ............................................................................................ 44

4.4.2 Minor Losses ................................................................................................................ 45

5.0 Model Calibration ................................................................................................................... 46

5.1 General ................................................................................................................................ 46

5.2 Calibration Methods............................................................................................................ 47

5.2.1 Calibration Setup ......................................................................................................... 47

5.2.2 Pump Curve Calibration .............................................................................................. 47

5.2.3 Macro Level Calibration .............................................................................................. 48

5.2.4 Model Sensitivity Analysis .......................................................................................... 49

5.2.5 Micro-Level Calibration .............................................................................................. 49


v

5.3 Calibration Results .............................................................................................................. 50

5.3.1 Final Calibrated C-Factors ........................................................................................... 50

5.3.2 Comparison of Pressures between Model and Field Tests .......................................... 50

5.4 Model Validation ................................................................................................................ 52

5.4.1 24 hour-EPS Simulation .............................................................................................. 52

6.0 Summary ................................................................................................................................. 55

7.0 Works Cited ............................................................................................................................ 56

Appendix A: Pump Curves ........................................................................................................... 57

Appendix B: Surveying Procedures and Data .............................................................................. 60

B.1 C-Factor Surveying ............................................................................................................ 60

B.1.1 C-Factor Survey Procedure ......................................................................................... 60

B.1.2 C-Factor Surveying Results ........................................................................................ 60

B.2 Fire Flow Surveying ........................................................................................................... 61

B.2.1 Fire Flow Survey Procedure ........................................................................................ 61

Appendix C: Tank Information ..................................................................................................... 63

Appendix D: Data Collection........................................................................................................ 66

D.1 C-Factor Test...................................................................................................................... 66

D.1.1 C-Factor Test Procedure ............................................................................................. 66

D.1.2 C-Factor Calculations ................................................................................................. 67

D.1.3 C-Factor Sensitivity Analysis ..................................................................................... 69

D.2 Fire Flow Test .................................................................................................................... 72

D.2.1 Fire Flow Test Procedure ............................................................................................ 72

D.2.2 Fire Flow Data ............................................................................................................ 73

D.2.3 Fire Flow Calculations ................................................................................................ 74

D.3 Boundary Conditions Collection ........................................................................................ 75

Appendix E: Model Validation ..................................................................................................... 81

Appendix F: Demand Factor Development .................................................................................. 84

Appendix G: Data Collection Logs............................................................................................... 90


vi

List of Tables Table 1 Summary of Project Tasks ................................................................................................. 2 Table 2 Elevated Storage Tank Specifications ............................................................................. 10 Table 3 Customer Base for Nicholasville ..................................................................................... 10 Table 4 Additional Customers for Nicholasville Water ............................................................... 11 Table 5 High Service Pump Information ...................................................................................... 12 Table 6 Distribution of Pipe Diameters in System ....................................................................... 13 Table 7 Distribution of Pipe Materials in System ......................................................................... 16 Table 8 Calibration Groups ........................................................................................................... 19 Table 9 C-Factor Testing Locations ............................................................................................. 22 Table 10 C-Factor Results ............................................................................................................ 24 Table 11 C-Factor Uncertainty Results ......................................................................................... 25 Table 12 Fire Flow Testing Locations .......................................................................................... 28 Table 13 Fire Flow Test Results ................................................................................................... 29 Table 14 Data Collection Methods ............................................................................................... 30 Table 15 Data Collection Equipment List .................................................................................... 31 Table 16 Diurnal Factor Summary ............................................................................................... 43 Table 17 Typical Hazen Williams C-Factor Coefficients............................................................. 45 Table 18 Finalized Pump Curves .................................................................................................. 48 Table 19 C-Factor Calibration Results ......................................................................................... 50 Table 20 Fire Flow Calibration Results ........................................................................................ 51 Table 21 Percent Difference between Model and Field Data ....................................................... 52 Table 22 C-Factor Surveying Results ........................................................................................... 61 Table 23 Lake Street Tank Data ................................................................................................... 63 Table 24 Capital Court Tank Data ................................................................................................ 64 Table 25 Stephens Drive Tank Data ............................................................................................. 65 Table 26 C-Factor Uncertainty Data ............................................................................................. 71 Table 27 C-Factor Sensitivity Results .......................................................................................... 72 Table 28 Data Collected from Fire Flow Tests ............................................................................. 74 Table 29 EPS vs. SCADA Data (Tank Levels) 10/10/2011 ......................................................... 81 Table 30 EPS vs. SCADA Data (Tank Levels) 10/11/2011 ......................................................... 82 Table 31 EPS vs. SCADA Data (Tank Levels) 10/13/2011 ......................................................... 83 Table 32 Demand Factor Data (10/10/2011) ................................................................................ 85 Table 33 Demand Factor Data (10/11/2011) ................................................................................ 86 Table 34 Demand Factor Data 10/12/2011 ................................................................................... 87 Table 35 Demand Factor Data 10/13/2011 ................................................................................... 88


vii

List of Figures Figure 1 Project Organization Chart ............................................................................................... 7 Figure 2 Distribution System Schematic ........................................................................................ 9 Figure 3 Nicholasville Water Treatment Plant ............................................................................. 11 Figure 4 WTP Plant Setup in WTP ............................................................................................... 12 Figure 5 Schematic of Pipe Diameter Distribution ....................................................................... 14 Figure 6 Components of System by Diameter .............................................................................. 15 Figure 7 Schematic of Pipe Material Distribution ........................................................................ 17 Figure 8 Components of System by Pipe Material ....................................................................... 18 Figure 9 C-Factor Testing Sites .................................................................................................... 21 Figure 10 C-Factor Test Setup ...................................................................................................... 23 Figure 11 Fire Flow Testing Locations ......................................................................................... 27 Figure 12 Fire Flow Test Setup .................................................................................................... 29 Figure 13 Example SCADA System Output ................................................................................ 30 Figure 14 Thiessen Polygons ........................................................................................................ 39 Figure 15 Demand Factors for 10/10/2012-10/13/2012 ............................................................... 42 Figure 16 EPS vs. SCADA Data for 10/10/2011 .......................................................................... 53 Figure 17 EPS vs. SCADA Data (Tank Levels) 10/11/2011 ........................................................ 54 Figure 18 EPS vs. SCADA Data 10/13/2011 ............................................................................... 54 Figure 19 Manufacturer’s High Service Pump #1 Curve ............................................................. 57 Figure 20 Manufacturer’s High Service Pump #2 Curve ............................................................. 58 Figure 21 Manufacturer’s High Service Pump #3 Curve ............................................................. 58 Figure 22 Manufacturer’s High Service #4 Pump Curve ............................................................. 59 Figure 23 Manufacturer’s High Service #5 Pump Curve ............................................................. 59 Figure 24 Hydrant Nozzle Discharge Coefficients ....................................................................... 69 Figure 25 SCADA Data (Tank Levels) 10/9/2011 ....................................................................... 75 Figure 26 SCADA Data (Pump Flow) 10/9/2011......................................................................... 76 Figure 27 SCADA Data (Tank Levels) 10/11/2011 ..................................................................... 76 Figure 28 SCADA Data (Pump Flow) 10/11/2011....................................................................... 77 Figure 29 SCADA Data (Tank Levels) 10/12/2011 ..................................................................... 77 Figure 30 SCADA Data (Pump Flows) 10/12/2011 ..................................................................... 78 Figure 31 SCADA Data (Tank Levels) 10/13/2011 ..................................................................... 79 Figure 32 SCADA Data 10/14/2011 ............................................................................................. 80 Figure 33 SCADA Data (Tank Levels) 10/14/2011 ..................................................................... 80 Figure 34 Demand Factors (10/10/2011) ...................................................................................... 84 Figure 35 Demand Factors (10/11/2011) ...................................................................................... 86 Figure 36 Demand Factors 10/12/2011......................................................................................... 87 Figure 37 Demand Factors 10/13/2011......................................................................................... 88


viii

List of Abbreviations ATSDR- Agency for Toxic Substance and Disease Registry DHS- Department of Homeland Security DVD – Digital Versatile Disk Ft- Feet GIS – Geographical Information System GPM – Gallons per Minute ID- Identification In- Inches KYPIPE – Hydraulic Modeling Software MCL – Maximum Contaminant Level MG/L – milligrams per liter MGD – Million gallons per day NPT- National Pipe Thread PRV- Pressure Reducing Valve PSI – Pounds Per Square Inch QAPP- Quality Assurance Project Plan QA/QC – Quality Assurance/ Quality Control RPD – Relative Percent Difference SCADA- Supervisory Control and Data Acquisition (SCADA) system SDG – Sample Delivery Group SOP- Standard Operating Procedure USEPA – United States Environmental Protection Agency WDS- Water Distribution System WTP – Water Treatment Plant


1

1.0 Introduction

1.1 Project Background

The United States Department of Homeland Security (DHS) has established 18 sectors of infrastructure and resource areas that comprise a network of critical physical, cyber, and human assets. One of these sectors is the Water Sector. The Water Sector Research and Development Working Group has stated that water utilities would benefit from a clearer and more consistent understanding of their system flow dynamics. Understanding flow dynamics is important to interpreting water quality measurements and to inform basic operational decision making of the water utility. Such capabilities are critical for utilities to be able to identify when a possible attack has occurred as well as knowing how to respond in the event of such an attack. This research will seek to better understand the impact of water distribution system flow dynamics in addressing such issues. In particular this project will: (1) test the efficiency and resiliency of the real-time hydraulic/water quality model using recorded data for system boundaries in order to understand the potential accuracy of such models, and understand the relationship between observed water quality changes and network flow dynamics, and (2) develop a toolkit for use by water utilities to select the appropriate level of operational tools in support of their operation needs. The toolkit is expected to have the following functionality: (a) a graphical flow dynamic model, (b) guidance with regard to hydraulic sensor placement, and (c) guidance with regard to the appropriate level of technology needed to support their operational needs. Primary objectives of this project include:

1. Develop an improved understanding about the impact of flow dynamics changes on distribution system water quality, and the potential benefits of using real-time network models to improve operational decisions – including detection and response to potential contamination events.

2. Develop an operational guidance toolkit for use by utilities in selecting the appropriate level of operational tools needed to support of their operational needs.

3. Develop a flow distribution model that will allow small utilities to build a basic graphical schematic of their water distribution system from existing GIS datasets and to evaluate the distribution of flows across the network in response to basic operational decisions.

This project has been broken down into 12 different project tasks as shown in Table 1. This Water Distribution System Calibration Report addresses Task 6 of the project which is defined as “Develop and Calibrate Hydraulic Computer Models.”


2

Table 1 Summary of Project Tasks

Task # Project Task 1 Establishment of an Advisory Group 2 Select Water Utility Partner 3 Survey and Evaluate SCADA Data 4 Build Laboratory Scale Hydraulic Model of Selected Water Distribution System 5 Develop Graphical Flow Distribution Model 6 Develop and Calibrate Hydraulic and Water Quality Computer Model 7 Quantify Flow and Water Quality Dynamics through Real-Time Modeling 8 Develop Sensor Placement Guidance 9 Develop Toolkit 10 Test and Evaluate Toolkit 11 Validate Toolkit 12 Write Report

1.2 Purpose of Project

One of the primary objectives of this project is to gain an understanding of the benefits of using distribution system models to improve operational decisions. Because these models will be used in decisions that involve significant investment and potential impact to the community, it is important that the model be an accurate representation of the actual conditions in the system. The validity of these models depends largely on the accuracy of input data and the assumptions made in developing the model. Certain model parameters exist that are either not readily available or difficult to obtain. These parameters typically include pipe roughness factors, constituent decay parameters, and the spatial and temporal distribution of water demands. As a result of the difficulty of obtaining economic and reliable measurements of these parameters, final model values are normally determined through the process of model calibration (Ormsbee, Lingireddy, 1997).

A successful calibration provides several benefits for the utility. When using the model to make decisions regarding the operation or improvement of the network, the utility will have confidence in the model to predict system behavior. The calibration process also greatly increases understanding of the flow dynamics in the system and the overall behavior and performance of the system. Calibration also helps to uncover missing or incorrect data in the system, such as incorrect pipe diameters or closed valves (Walski, et al.).

In order to successfully calibrate a model, hydraulic tests are used to obtain information about the model. This project will utilize C-factor tests along with fire flow tests, procedure to be discussed, to gather information about the system. Values such as pressure, flow rate, storage tank level, head and flowrate values associated with pump curves, etc. are gathered during hydraulic field testing. These field results are compared to model predictions, and then


3

parameters in the model are altered until the model simulates field conditions. Data within the model is adjusted until behavior predicted by the model reasonably agrees with measured system performance over a range of operating conditions (EPA, 2005).

Calibration involves making changes to system demands, roughness of pipes, pump operating conditions, and other model attributes (Walski, et al.). However, it is important that the data is adjusted only within reasonable limits. For example, changing C-factor values of pipes outside of reasonable values based on the pipe material and age would seem like apparent calibration for certain conditions, but would probably result in unlikely results for a new range of conditions. The calibration process can also reveal closed valves, severely tuberculated pipes, missing pipes, and other issues that can be resolved to improve operation of the system (EPA, 2005).

Once the model accurately predicts field measurements under a wide range of conditions, the model is considered to be calibrated. A new set of hydraulic testing data should be collected in order to verify the calibration. If the new testing results closely match the calibrated model, the calibration is successful. The model is calibrated under the assumption of steady state conditions, simply utilizing known boundary conditions at the time the tests were performed. However, an extended period simulation (EPS) can also be utilized to verify calibration, simulating model behavior over a certain time period.

In general, a network model calibration effort should encompass seven basic steps: (1) Identification of intended use of the model (2) Identification of calibration model parameters and their initial estimates (3) Model studies to determine the calibration data sources (4) Data collection (5) Macro calibration (6) Sensitivity analysis (7) Micro calibration. Details and procedures pertaining to these seven basic steps can be found in Calibration of Hydraulic Network Models by Ormsbee and Lingireddy (1997). By using data from hydraulic testing in the calibration process, confidence in the model greatly increases. The model will be an accurate tool to aid in planning, design, and daily operation of the water distribution system (AWWA, 2005).

1.3 Project Scope

Successful calibration of the water distribution system in Nicholasville, KY, involves several major tasks that comprise the overall scope of the project.

• Data Collection: Gather and review all available information on the Nicholasville water distribution system in order to develop the computer model. This includes Autocad/GIS files showing all pipes, demand nodes, hydrants, and valves in the system. Customer usage bills will also be appropriate in order to gather accurate demand data. Specifications for the storage tanks, pumps, and the water treatment plant are also collected.

• Computer Model Development: Create a model of the system using KYPIPE, including all pipes, hydrants, nodes, junctions, demand nodes, elevated storage tanks, and pumps. Descriptive parameters that are known for each component should be entered appropriately.


4

• Field Testing: Develop and execute a field testing protocol. These tests will include C-factor tests and Fire Flow tests (procedure to be discussed). All data should be recorded appropriately, including all boundary conditions during test periods.

• Model Calibration: Results from field testing are compared to model behavior, and data in the model is adjusted until it reasonably agrees with measured system performance. System demands, roughness of pipes, pump operating conditions, and other attributes are altered in the model to match field conditions.

• Model Calibration Verification: To ensure the model calibration is an accurate representation of the system, a new set of field data is collected for verification purposes. An extended period simulation (EPS) is executed on the calibrated model and compared to results from field data over an extended time, such as water levels in elevated storage tanks. If the new test results closely match model behavior, the calibration is verified.

1.4 Project Management

1.4.1 Distribution List

Lindell Ormsbee, PhD, P.E. Kentucky Water Resources Research Institute University of Kentucky 233 Mining and Minerals Building Lexington, KY 40506-0107 (859) 257-6329 Scott Yost, PhD, P.E. Department of Civil Engineering University of Kentucky 354C O. H. Raymond Building Lexington, Kentucky 40506-0281 Phone: 859-257-4816

L. Sebastian Bryson, PhD, P.E. Department of Civil Engineering University of Kentucky 254C O. H. Raymond Building Lexington, Kentucky 40506-0281 Phone: 859-257-3247 Mr. Tom Calkins Public Utilities Director Nicholasville Water Department Nicholasville, Kentucky 40356 (859) 885-9473


5

Mr. Danny Johnson Water Distribution Superintendent Nicholasville Water Department Nicholasville, Kentucky 40356 (859) 983-8734 Mr. Ragan Cobb Assistant Water Distribution Superintendent Nicholasville Water Department Nicholasville, Kentucky 40356 (859) 983-9000 Mr. Jim McDaniel Nicholasville WTP Shift 1 Operator 595 Water Works Road Nicholasville Water Department Nicholasville, KY 40356-9690 (859) 885-6974 Mr. John Taylor National Institute for Hometown Security, Inc. 368 N. Hwy 27 Somerset, KY 42503 (859) 451-3440 Samuel G. Varnado, PhD Senior Program Advisor National Institute of Homeland Security 368 N. Hwy, 27, Suite One Somerset, KY, 42503 606-451-3450 Mr. Morris Maslia Research Environmental Engineers Agency for Toxic Substances and Disease Registry (ATSDR) National Center for Environmental Health 4770 Buford Highway Mail Stop F-59, Room 02-004 Atlanta, Georgia 30341-3717 (770) 488-3842


6

1.4.2 Project Organization

The roles and responsibilities of project participants are listed below. Refer to Figure 1 on page 7 for the project organization chart. Lindell Ormsbee, Director Kentucky Water Resources Research Institute University of Kentucky Role: Project Manager Responsibilities: Oversee data, Project Manager Scott Yost, Associate Professor Department of Civil Engineering University of Kentucky Role: Field Manager Responsibilities: Manage data collection activities; ensure data collection conducted consistent with QAPP Tom Calkins, Public Utilities Director Nicholasville Water Department City of Nicholasville Role: Primary Contact for the Nicholasville Water Department Responsibilities: Provide assistance in obtaining data for the Nicholasville System. Serve as liaison for Nicholasville personnel Danny Johnson, Water Distribution Superintendent (WDS) Nicholasville Water Department City of Nicholasville Role: Assist field crews and oversee field testing activities Responsibilities: Provide personnel for field testing, oversee training of field crew Jim McDaniel, Operator of Water Treatment Plant Nicholasville Water Department City of Nicholasville Role: WTP Shift 1 Operator Responsibilities: Help coordinate and collect real time data from the WTP during field testing (i.e. pump discharges, tank water levels). David Scott, Operator of Water Treatment Plant Nicholasville Water Department City of Nicholasville Role: WTP Shift 2 Operator Responsibilities: Help coordinate and collect real time data from the WTP during field testing (i.e. pump discharges, tank water levels)


7

Mr. Morris Maslia Research Environmental Engineers Agency for Toxic Substances and Disease Registry (ATSDR) National Center for Environmental Health Role: Tracer Analysis Consultant Responsibilities: Provide guidance on conducting tracer study Joe Goodin Graduate Research Assistant(s) Department of Civil Engineering University of Kentucky Role: Data acquisition oversight Responsibilities: Collect field data from hydrant testing; troubleshoot field equipment; undertake corrective measures as needed to develop and calibrate hydraulic model of the water distribution system.

Figure 1 Project Organization Chart


8

2.0 Existing System

2.1 General

The City of Nicholasville is located in Jessamine County, Kentucky, southwest of the City of Lexington. The population was reported as 28,015 for the 2010 census, making it the 12th largest city in the state. The topography of the area varies from a maximum elevation of approximately 1042 feet to a minimum elevation of around 560 feet. The city has a total area of 8.5 square miles according to the U.S. census bureau, all of which is serviced by the Nicholasville Water Treatment Plant (WTP).

The Nicholasville water distribution system consists of an intake pumping facility, a water treatment plant, a high service pumping facility, and transmission and distribution systems. The treatment facility is a conventional turbidity removal plant that utilizes chemical coagulation, flocculation, settling and filtration to remove suspended particles from the raw water. The treated water transmission and distribution system consists of a grid of mains ranging from 2 to 24 inches in diameter and has a total elevated storage of 3 million gallons (3 Tanks) (Nicholasville, 2009-2011). The transmission mains from the water plant to town consists of two 10 inch lines for the entire route and a 16 inch line installed approximately 2/3 of the distance from the water plant to town. The 16 inch line becomes a 24 inch line which continues directly into town. In town, distribution consists of 4 through 12 inch water mains. The majority of the system is looped with a combination of 10, 12 and 20 inch pipes. A schematic of the distribution system is shown below in Figure 2.


9

Figure 2 Nicholasville Distribution System Schematic

2.2 Supply Source and Storage

The Nicholasville Water Treatment plant is supplied by surface water from Pool 8 of the Kentucky River. The location of the Kentucky River relative to the city of Nicholasville and the WTP can be observed in Figure 2 above.

The distribution system contains three elevated storage tanks, providing a total of 3 million gallons of storage for the system. Both the Lake Street and Stephens Drive tanks are ovaloid shaped and each has a capacity of 750,000 gallons. The largest tank, located at Capital Court, is composite shaped with a capacity of 1,500,000 gallons. The Lake Street tank was constructed in 1965, Stephens Drive in 1974, and Capital Court in 2005. Specifications of all storage tanks are summarized in Table 2. One ovaloid tank is located off Stephens Drive in the northwest part of town, and the other ovaloid tank is off Lake Street between Beechmont Drive and Hillcrest Drive in the northeast portion of town. The composite elevated tank is located off Capital Court near the county fairgrounds in the southeast part of Nicholasville.


10

Table 2 Elevated Storage Tank Specifications

Elevated Storage Tank Identification and Elevations* Name Lake St Capital Court Stephens Drive Size (gallons) 750,000 1,500,000 750,000 Elevation of bottom of the tank (ft) 1025.75 952.5 966.5 Minimum Level (ft) 1105.75 1111.5 1109.5 Maximum Level (ft) 1143.75 1151 1148 Shape Ovaloid Composite Ovaloid Inside Diameter (ft) 60 86 68 * Data from Nicholasville Water Utility Department

2.3 Water Treatment Plant

The water treatment plant (WTP) is located at an elevation of approximately 870 feet msl. At the Nicholasville WTP, raw surface water is pumped from Pool 8 of the Kentucky River into a chemical mix basin. Once it has passed through the chemical mix basin it continues through a series of flocculation basins to the settling basins. After the treatment process of coagulation and sedimentation, the clarified water flows into dual media filter beds to remove any remaining solids. After filtration, fluoride is added to the treated water to help improve dental hygiene. Prior to pumping the water into the distribution system, the water is disinfected with chloramines. The water distribution plant has a capacity of 9 million gallons per day (MGD). In 2010, the average day demand was approximately 4.4 MGD. The treatment plant serves approximately 13,000 retail customers and two wholesale customers. According to the Kentucky Infrastructure Authority website, Nicholasville directly serves the following customer base as shown in Table 3.

Table 3 Customer Base for Nicholasville

Wholesale Customers 2 Wholesale Usage (MG) 235.8 Residential Customers 12521 Residential Usage (MG) 803.5 Commercial Customers 619 Commercial Usage (MG) 11.1 Industrial Customers 25 Industrial Usage (MG) 95 Total Customers 13167 Total Usage (MG) 1145.4

The wholesale customers include Jessamine County Water District #1 and Jessamine South Elkhorn Water District. Statistics associated with both of these systems are provided in Table 4.


11

Table 4 Additional Customers for Nicholasville Water

Purchaser Annual Volume (MG)

Connection Meters

Serviceable Population

Jessamine County Water District #1 207.7 6 3994 Jessamine South Elkhorn Water District 28.1 1 7693 Total 235.8 7 11687

Plant operations are monitored and controlled by a computer based Supervisory Control and Data Acquisition (SCADA) system. The SCADA system monitors and controls pumps, chemical feeds, treatment equipment, flow rates, water levels, etc. The SCADA system also provides real time data for pumping operations as well as tank levels, pump flows and pump pressures. This data was obtained during field testing to help calibrate the hydraulic model. A photo of the WTP is shown below in Figure 3.

Figure 3 Nicholasville Water Treatment Plant

When demand causes water levels in the elevated storage tanks to drop below a minimum water-level mark, high service pumps are turned on at the Nicholasville WTP. Typical pressure for high service turbine pumps ranges between 135 and 165 psi. Instruments to measure pressure are located on each pump, and the flow from the pump station can be recorded and stored at any


12

time. The capacity and horsepower of each high service pump is shown below in Table 5. The pump curves for all 5 high service pumps are shown in Appendix A: Pump Curves.. The layout of the high service pumps in the WTP (shown in the KYPIPE model) is shown in Figure 4.

Table 5 High Service Pump Information

Capacity (GPM) Horsepower High Service #1 1500 200 High Service #2 2100 200 High Service #3 2100 250 High Service #3 3200 300 High Service #5 3200 300

Figure 4 WTP Plant Setup in WTP

2.4 Distribution System Piping

The treated water transmission and distribution system consists of a grid of mains with a total length in the system over 282 miles (1,490,617.2 feet). The system contains pipe ranging from 1 to 24 inches in diameter, with the most widely represented size being 6 inch diameter pipes at approximately 43% of the system. Four inch and 8 inch pipes are the most common following 6


13

inch pipe at 15% and 14%, respectively. 13.5% (38.2 miles) of the total water lines in the system are distribution mains with diameters between 10 and 18 inches.

Table 6 shows the total length and percentages of varying pipe diameters in the system, and Figure 5 shows the distribution of varying pipe diameters throughout the system. The blue lines display the pipes with 6 inch diameters. Figure 6 shows a zoomed in schematic of the system, highlighting the location of the elevated storage tanks.

Table 6 Distribution of Pipe Diameters in System

Pipe Diameter

(in) Length (ft) Length (miles)

Percentage of Total Length

1 1227.3 0.23 0.1% 2 154189.1 29.20 10.3% 3 49929.8 9.46 3.3% 4 225137.2 42.64 15.1% 6 646894.8 122.52 43.4% 8 211586.7 40.07 14.2% 10 119503.9 22.63 8.0% 12 24296.2 4.60 1.6% 16 19472.7 3.69 1.3% 20 33852.8 6.41 2.3% 24 4526.8 0.86 0.3%


14

Figure 5 Schematic of Pipe Diameter Distribution


15

Figure 6 Components of System by Diameter


16

The majority of water lines throughout the distribution network consist of PVC pipes; they encompass approximately 66% of the network. Asbestos cement, cast iron, and ductile iron are the most prominent pipe materials after PVC, with percentages of 15%, 10%, and 6%, respectively. Galvanized steel, iron, polyethylene, and copper are also present in the system, but in fairly small amounts. The older pipes (55+ years) in the system are predominately cast iron and ductile iron, while the newer pipes are mostly PVC with small amounts of ductile iron as well. Generally the asbestos cement pipes are around 40 years old. Table 7 shows the quantities of varying pipe materials in the system, while Figure 7 and Figure 8 show a schematic of the different pipe materials present in the system.

Table 7 Distribution of Pipe Materials in System

Pipe Material Length (ft) Length (miles)

Percentage of Total Length

PVC 985,244 186.6 66.1% Asbestos Cement 218,514 41.4 14.7%

Cast Iron 151,736 28.7 10.2% Ductile Iron 904,54 17.1 6.1%

Galvanized Steel 17,563 3.3 1.2% Other 16,854 3.2 1.1% Iron 9,678 1.8 0.7%

Polyethylene 553 0.1 0.04%


17

Figure 7 Schematic of Pipe Material Distribution


18

Figure 8 Components of System by Pipe Material


19

3.0 Data Collection

3.1 C-Factor Test

All pipes were categorized into different calibration groups based upon material, size, and age. The breakdown of calibration groups, along with the percentage of the distribution system that each calibration group encompasses, is shown below in Table 8. The pipes are first divided by pipe material classification and then further broken down by pipe diameter within each pipe material group. The age classification of each calibration group is also included, along with the total length of pipe in each group and the percentage each encompasses of the entire system.

Table 8 Calibration Groups

Group Pipe Material Diameter (in)

Average Age (yr)

Age Low End (yr)

Age High End (yr)

Total Length (ft)

% of System

1 Asbestos Cement 4, 6 40 N/A 40 183,550 12.6% 2 PVC 2,3 15 2 30 182,943 12.6% 3 PVC 4,6 15 2 30 594,692 40.9% 4 PVC 8,10,12 5 1 10 207,608 14.3% 5 Ductile Iron 10,12,16 30 20 45 40,677 2.8% 6 Ductile Iron 6, 16,20,24 15 15 20 57,850 4.0% 7 Asbestos Cement 8,10 40 N/A 40 34,963 2.4% 8 Cast Iron 4,6 40 40 60 82,071 5.6% 9 Cast Iron 8,10,12 40 40 60 69,664 4.8%

These pipes were then assigned an initial roughness value, found from C-factors tests, to be placed in the uncalibrated hydraulic model. The results of the C-factor tests can be used to assign C-factors to other pipes in the system with similar characteristics (EPA, 2005). The goal of the sampling locations was to try and perform a C-factor test for each of the calibration groups. This was not possible due to accessibility of hydrants and lack of available, suitable locations for a given pipe material. Ten C-factor tests were executed in Nicholasville, representing pipes from four different calibration groups. Additional background on the C-factor is described in section 4.4 Pipe Friction Losses on page 44.

3.1.1 C-Factor Sites

When selecting site for C-factor tests, many factors had to be considered in order to gather data that would be useful in the calibration process. These factors include:


20

• Age of the Pipe: Pipes of different ages were selected to help obtain a representative sample of all the pipes.

• Material of the Pipe: When possible, sampling sites contained different material to help obtain a better Hazen Williams coefficient.

• Accessibility of the hydrant: Some hydrant locations were not accessible due to location in a congested area, near hospitals/schools, etc.

• Diameter of the Pipe: Pipes of different sizes were selected to help obtain a representative sample of all the pipes.

• Amount of Flow in the pipe: In order to obtain a good sample, enough flow should be produced to drop the residual pressure at least 15 psi (McEnroe, 1989).

It is also important that testing locations were not located near boundary conditions, such as pumps or elevated storage tanks. If data is collected near boundary nodes, the difference between model and field results may be minimal because of the short distance. However, the difference in slopes of the hydraulic grade lines will be significant (Walski, et al.). In order to calculate accurate C-factor vales, a homogeneous section of pipe between 400 and 1,200 feet should be selected. Selecting pipes in this length range will likely result in an adequate pressure drop (EPA, 2005). It is also necessary to close valves near the flow hydrant to force flow through the pipe section being measured (AWWA, 2005), so it is important that the required valve is accessible.

Each C-Factor sampling location has been given a test site ID. Each test site corresponds to 3 hydrants and associated valve(s) to be closed. One hydrant has been designated the flow hydrant where the other two hydrants will be used to collect pressure drops. Each individual hydrant has previously been assigned an ID by the city of Nicholasville and each hydrant has also been given an ID for this project (assigned in the KYPIPE Model). Figure 9 contains a map of the testing locations, and Table 9 shows descriptions of all sites used for C-factor tests.


21

Figure 9 C-Factor Testing Sites


22

Table 9 C-Factor Testing Locations

Test Site ID

Residual Hydrant #1 Nicholasville Hydrant ID Location

Pipe Diameter

(in)

Target Calibration

Group Residual Hydrant #2

Flow Hydrant

C-1 P1 1145 Squires Way between

Bennett Drive and end of Squires Way

8 4 P21 1146 8

F1 1147 8

C-2 P2 786 John C. Watt Drive

between Lancaster Road and Delta Drive

6 8 P22 157 6

F2 788 6

C-3 P3 337 Shun Pike between

Alta Drive and W. Brown Street

10 5 P23 268 10

F3 698 10

C-4 P4 101 S Central Avenue

between Royalty Court and Kingway Drive

6 3 P24 100 6

F4 99 6

C-5 P5 790

Wilmore Road next to schools

12 4 P25 836 12

F5 1009 12

C-6 P62 1158 Harlan Drive between

Stanley Drive and Cannonball Drive

8 4 P26 989 8

F6 988 8

C-7 P7 403 Bell Place between

Hillbrook Drive and Cloverdale Drive

8 4 P27 717 8

F6 390 8

C-8 P8 1266 Bernie Trail near the

intersection with Lebeau Drive

8 4 P28 1195 8

F9 1194 8

C-9 P9 406 Hawthrone Drive near

the intersection of Old KY-29

6 3 P29 670 6

F9 671 6

C-10 P10 239 Weil Lane between

Linden Lane and Beacon Hill

6 3 P30 240 6

F10 287 6


23

3.1.2 C-Factor Test Procedure

C-factor tests calculate the friction coefficient used in the Hazen Williams equation (further explanation in section 4.4 Pipe Friction Losses on page 44) by measuring flow and head loss in the field. The two-gage method was used in Nicholasville to execute the C-factor tests. During the test, pressure was read using a static pressure gage (equipment specifications to be discussed in section 3.4 Data Collection Equipment on page 31) at hydrants (residual) at the upstream and downstream ends of the section while a hydrant downstream of the section was opened to force flow and a sufficient pressure drop. The elevation difference between the residual hydrants is then used to calculate head loss in the section. It is ideal to select residual hydrants that are spaced far enough apart to induce a pressure drop of at least 15 psi. It is also necessary to close valves downstream of the flow hydrant to force flow through the pipe section being measured. A pitot gage was attached to the flow hydrant to measure the flow rate needed for calculations (EPA, 2005). All calculations used to determine C-factors are shown in Appendix D.1.2 C-Factor Calculations on page 67. A schematic illustrating the C-factor test setup is shown below in Figure 10.

Figure 10 C-Factor Test Setup

C-factor testing procedures were performed according to the American Water Works Association M32-Computer Modeling of Water Distributions Systems. A step by step procedure for conducting the C-factor test is shown in Appendix D.1.1 C-Factor Test Procedure on page 66. All data collected during C-factor tests were recorded on specified data sheets, and these data sheets are shown in Appendix G: Data Collection Logs on page 90.

3.1.3 C-Factor Results

A summary of the results of the C-factor tests performed on the system are shown below in Table 10.


24

Table 10 C-Factor Results

Site ID Pipe Material

Pipe Diameter

(in)

Calibration Group

Head Loss (ft) C-Factor

C-11 PVC 8 4 5.81 149 C-11 PVC 8 4 15.05 224 C-2 Cast Iron 6 8 55.00 143 C-3 Ductile Iron 10 5 14.94 105 C-41 PVC 6 3 26.74 233 C-41 PVC 6 3 29.05 211 C-5 PVC 12 4 12.86 127 C-5 PVC 12 4 27.88 144 C-6 PVC 8 4 9.25 132 C-71 PVC 8 4 3.50 130 C-81 PVC 8 4 5.05 145

C-91&3 PVC 6 3 20.23 180 C-102 PVC 6 3 119.63 75 C-102 PVC 6 3 117.32 78

Note 1: After C-Factor Testing was complete and the gauges were recalibrated it was discovered that Gauge 1 was found to be approximately 1 to 2psi lower than the other gauges. The following results reflect this adjustment. Note 2: C-Factor Site resulted in extreme headloss. Note 3: Located upstream from a broken valve.

Several of the C-factor sites were tested multiple times at different flows because the head loss was not high enough to collect a sound measurement. Site C-9 was later to be found to have a broken valve directly downstream of the hydrant which would have affected the C-Factor rating. Some C-Factors appear to be much higher than what is possible for a particular material pipe. Thus a sensitivity analysis was performed to check to see what the possible errors were in the calculations.

3.1.4 C-Factor Sensitivity Analysis

The calculated C-factors found for each C-factor test performed are subject to error based on measurements used in the calculations. A sensitivity analysis was performed to investigate how the uncertainty in the C-factor value can be attributed to the uncertainty of variables used in the calculations. This investigation provides information about which specific variables are more


25

influential in the total uncertainty of the C-factor. An uncertainty analysis was also performed to quantify the uncertainty in the calculated C-factor. The level of precision in instruments used in the data collection process, along with the expected errors in reading the instruments, are taken into account to find a range of possible values associated with each C-factor.

The sensitivity analysis provides information about which variables have the greatest contribution to the total uncertainty in the calculated C-factor. The uncertainty in the diameter of the pipe is the most influential in the uncertainty of the C-factor, followed by the uncertainty in the coefficient of discharge of the hydrant and diameter of the hydrant opening. The uncertainty in the discharge pressure, length of pipe, and change in pressure and elevation between hydrants is not as influential in the uncertainty of the C-factor as the variables already mentioned.

The uncertainty for each C-factor test was calculated, and the results of the C-factor uncertainty analysis are shown in Table 11. All calculations used to find these uncertainties along with the sensitivity of the C-factor in relationship to the uncertainty in each variable are shown in Appendix D.1.3 C-Factor Sensitivity Analysis on page 69.

Table 11 C-Factor Uncertainty Results

Site C-Factor with Uncertainty

C-1 149 ± 48.6 C-1 224 ± 73.1 C-2 143 ± 51.9 C-3 105 ± 31.0 C-4 233 ± 84.5 C-4 211 ± 76.5 C-5 127 ± 35.2 C-5 144 ± 40.2 C-6 132 ± 43.7 C-7 130 ± 42.9 C-8 145 ± 47.2 C-9 180 ± 65.5 C-10 75 ± 65.5 C-10 78 ± 28.7

3.2 Fire Flow Test

Fire flow tests are useful for collecting both discharge and pressure data for use in calibrating hydraulic network models. Opening a hydrant to full flow puts stress on the system, resulting in significant head loss in adjacent pipes. Such tests are normally conducted using a static pressure


26

gauge (for measuring both static and dynamic heads) and a pitot gauge (for use in calculating discharge from the flow hydrant). In performing a fire flow test, at least two separate hydrants were first selected. One hydrant was identified as the pressure or residual hydrant, and the remaining hydrant was identified as the flow hydrant.

In order to obtain sufficient data for an adequate model calibration, it is important that data from several fire flow tests be collected. Before conducting each test, it is also important that the associated system boundary condition data be collected, which includes information on tank levels, pump status, etc. The values for flow and pressure recorded during a fire flow test are used along with data about the state of the system including pump operation, tank water levels, and general system demand. The system model is run under the observed conditions and adjustments are made to the roughness coefficients or other parameters until the model represents the field data (EPA, 2005).

3.2.1 Fire Flow Sites

Fire flow testing should occur during peak flow conditions to ensure that adequate pressure drops are created. If sampling occurs during low flow conditions, the velocities may not be high enough to produce enough head loss for a good calibration. In order to determine Fire flow sampling locations several factors had to be taken into account. These factors include:

• Distance from Boundary Conditions: It is suggested that the testing site take place far

away from boundary conditions such as tanks, WTP, PRV to increase the head loss in the system (Walski, et al.).

• Accessibility of the hydrant: Some hydrant locations were not accessible due to location in a congested area, near hospitals/schools, etc.

• Expected head loss: Walski suggests a head loss at least five times as large as the error in the head loss measuring device (Walski T. , 2000).

• Amount of Flow in the pipe: In order to obtain a good sample enough flow should be produced to drop the residual pressure at least 10 psi (AWWA, 1999).

Each fire flow sampling location has been given a Test Site ID. Each test site contains 2 hydrants. One hydrant is the designated flow hydrant and the other hydrant is the residual hydrant. Each individual hydrant has been given an ID by the city of Nicholasville and each hydrant has also been given an ID for this project (assigned in KYPIPE Model). Figure 11 shows the location for each fire flow site, and Table 12 shows descriptions of all sites used for fire flow tests.


27

Figure 11 Fire Flow Testing Locations


28

Table 12 Fire Flow Testing Locations

Test Site ID

Residual Hydrant Nicholasville Hydrant ID Location Pipe

Diameter (in) Flow Hydrant

FF-1 R1 805 Juniper Drive between Arbee

Drive 8

FH101 799 8

FF-2 R2 236 Kimberly Heights Drive near

intersection of Shreveport Drive 8

FH102 478 8

FF-3 R3 1129 Between 144 Brome Drive and

124 Brome Drive 8

FH103 1130 8

FF-4 R4 435 South Creek Drive near

intersection of Bridge Side Drive 6

FH104 564 6

FF-5 R5 462

Lindsey Drive 6

FH105 463 6

FF-6 R6 192 South 5th Street between

Broadway St and West Maple St 8

FH106 214 8

FF-7 R7 681 Christopher Drive between Kevin

Drive and Quinn Drive 8

FH107 514 8

FF-8 R8 476 Intersection of Bell Lawn and

Hillbrook Drive 10

FH108 489 6

FF-9 R9 1184 Dawson Pass between 120

Dawson Pass and Curtis Ford Trail 8

FH109 1183 8

FF-10 R10 45 East Oak Street between Scott

Alley and N York Street 8

FH110 44 8

3.2.2 Fire Flow Test Procedure

The AWWA M17 guide- Installation, Field Testing, and Maintenance of Fire Hydrants was used to develop the standard operating procedures for the fire flow test. A schematic illustrating the setup of a fire flow test is shown in Figure 12. A step by step procedure for conducting the fire flow test is also outlined in Appendix D.2.1 Fire Flow Test Procedure on page 72.


29

Figure 12 Fire Flow Test Setup

Fire flow tests are useful for collecting both discharge and pressure data for calibrating hydraulic network models. Calculations are performed to determine the the maximum capacity of a hydrant if it is pumped down to a 20 psi residual pressure. These calculations are shown in Appendix D.2.3 Fire Flow Calculations on page 74.

3.2.3 Fire Flow Results

A summary of the results from fire flow tests executed in Nicholasville are shown below in Table 13. The static and residual pressures recorded at the residual hydrant are shown (including the pressure drop), along with the static pressure and flow rate at the flow hydrant.

Table 13 Fire Flow Test Results

Residual Hydrant Flow Hydrant

Site Static Pressure (psi)

Residual Pressure (psi)

Pressure Drop (psi)

Static Pressure (psi)

Flow (gpm)

FF-1 99 85 14 102 1507 FF-2 88 82 6 92 1488 FF-3 92 64 28 94 2428

FF-4** 94 79 15 92 1425 FF-5 56 44 12 46 950 FF-6 72 64 8 73 2279 FF-7 101 91 10 98 2760 FF-8 69 62 7 75 1300 FF-9 83 78 5 73 1354 FF-10 84 77 7 82 2480

** Demand was adjusted in the area to better reflect the Industrial's Park Demand Pattern


30

3.3 Boundary Conditions Collection

Boundary conditions information was collected from Nicholasville’s SCADA data. Table 14 shows the parameters collected that are necessary for model calibration along with the methods of collection.

Table 14 Data Collection Methods

Parameter Number/Frequency Collection Method

Pressure Data Collected during each test as specified

Hydrant Flow Meters and Hydrant Static Pressure

Gages

Tank Water Levels 1 hour intervals for Extended Period Simulation (EPS), 5-minute interval

during field testing SCADA system records

Flow from High Service Pumps

1 hour intervals for Extended Period Simulation (EPS), 5-minute interval

during field testing SCADA system records

On/Off cycling of Pumps

Every Pump on/off cycle of pumps collected SCADA system records

Figure 13 shows an example of SCADA system ouput, showing a tank level and pump flowrate on October 12, 2011. Boundary conditions from the SCADA system collected during testing are shown in Appendix D: Data Collection.

Figure 13 Example of SCADA System Output from Nicholasville Control Center


31

3.4 Data Collection Equipment

Table 15 below shows all equipment used during field testing. Descriptions of various equipment items are also described below.

Table 15 Data Collection Equipment List

Field Testing Supplies Supplier Item #

Hydrant Flow Meter Pollard P669LF Hydrant Static Pressure Gage Pollard P67022LF

Pressure Snubbers Pollard P605 Hydrant Wrenches Pollard P66602

• Hydrant Flow Gage: Hydrant flow data was gathered using the Pollard hydrant flow gage for both the fire flow test and C-factor test. During the test, hydrant flow data was recorded by viewing the flow gage and recording the results. These observations were confirmed by a second field person before recording.

• Hydrant Static Pressure Gage: Static and residual pressures will be recorded using a Pollard hydrant static pressure gage. The fire hydrant gage comes with a bleeder valve allowing the user to vent air and water from the hydrant before taking readings. Once installed on the hydrant, the hydrant static pressure gage can be used to record the residual pressures by visually recording the gage data.

3.5 Data Collection Safety Procedures A Quality Assurance Project Plan was distributed to all project personnel via email prior to data collection. All personnel were informed of the standard operating procedures and associated Quality Assurance/ Quality Control (QA/QC) protocols. Prior to any field data collection, all graduate research students viewed a short video produced by American Water Works Association entitled Field Guide: Hydrant Flow Tests. The video covers the basic protocol and necessary steps for proper fire flow testing.

3.5.1 Communication and Contingencies Possible problems associated with hydrant testing such as downstream flooding, mechanical problems with the distribution system, poor instrumentation and inaccurate record keeping have been documented along with actions to remedy and prevent such problems. Each member of the field testing crew had in their possession at all times during testing a cellular telephone or two-way radio communication device. Each member also had a complete list of all cellular telephone numbers and radio frequencies. This will allow for immediate communication during the testing and will allow for a quick response in the event of an emergency.


32

The health and safety of the public is extremely important in conducting the field testing procedures. A Nicholasville WTP staff operator was present while the hydrants were flowed. The WTP staff operator was responsible for frequently monitoring the flow in the system and if anything unusual was observed, the operator would take the proper course of action to remedy the situation.

3.5.2 Health and Safety Issues Prior to testing, all relevant local authorities (i.e. Nicholasville city government, Nicholasville police department, Nicholasville fire department) were notified of the location, time and extent of field sampling. Emergency contact numbers for the field team were provided to all relevant authorities. The team also had in their possession emergency contact numbers for all relevant agencies. All traffic regulations, procedures, and laws were strictly observed by teams when driving vehicles from site to site. Because of the duration of time that field teams were possibly exposed to the sun, sun block was provided. All field personnel were required to wear reflective vests during the tests as well as proper clothing and shoes to protect against injury. Field testing teams were also required to carry proper identification on them at all times in the field in case a situation arose where a field member needed to be identified by local residents.

Before any tests that involved the opening and flushing of hydrants were executed, the location of the tests were approved by local water utility and fire department officials to ensure that system pressures were not lowered below a level that could induce cross contamination of the system by sucking contaminants into the distribution system. The field team also had the option to survey the area to determine the direction of flow and ultimate disposition of any discharges so as to prevent any safety issues or loss or damage of private property. Where warranted, a hydrant diffuser or a 4 x 8 piece of plywood was available to avoid damage to green space as a result of the discharging jet of water from the fire hydrant.

Prior to opening any hydrant nozzle, the field crew confirmed that the hydrant valve was closed. As an added precaution, the nozzle cap was removed with a hydrant wrench with the field personnel standing to the side so as to prevent injury from a hydrant cap shooting off in the event the hydrant valve was actually open. In opening any hydrant, care was observed to open the hydrant slowly and in incremental steps so as to minimize any transient pressure issues in the distribution system. Prior to installing any instruments (i.e. flow/pressure gage) on the discharge nozzle of the hydrant, the hydrant was first opened and flowed until the water flowed clear to remove any particles or rust that may have accumulated in the hydrant service line and barrel. Once this was performed, the hydrant valve was closed and the instruments installed prior to opening the hydrant a second time for use in data collection.

3.6 Documentation and Records Raw data collected in the field was recorded on paper forms (in ink) that were developed for this purpose (shown in Appendix G: Data Collection Logs). Once completed, the forms were scanned into an adobe pdf for subsequent electronic archival. The data was also transcribed


33

into an Excel spreadsheet. The data manager reviewed all data for consistency and compliance with all sampling QA/QC protocols prior to recording. Any apparently anomalous values were verified with the field personnel prior. This information was conveyed to the field supervisor for possible review and revision of the current data collection protocols. Electronic data backup was performed after each entry session on a peripheral hard drive. A hardcopy of all project logs, forms, records, and reports were archived by the data manager. Hardcopies of all logs, forms, records, and reports are made available upon request and pending approval of the data manager.

3.7 Quality Control for C-Factor Testing The quality of the data collected as part of the C-factor testing was controlled through the procedures described in the following sections.

3.7.1 Review of Construction Records Prior to conducting any C-factor tests, recent construction records were reviewed to identify those parts of the system where valves could have been left closed or partially closed. These valves will be checked in the field to verify that they were in the open position.

3.7.2 Pressure Gage Calibration and Validation Pressure gauges were recently purchased and had previously been calibrated. Following the field tests, the gages were then rechecked against the known pressure source to confirm the gages are still within the calibration limits (i.e. + - 2 psi). After initiating the field tests, the gages were again checked. At this time, each of the gages were found to have slightly changed from their original calibration valuves. One pressure gauge was found to consistently record pressures by approximately 1 to 2 psi lower than the other two pressure gauges. The pressure results from this particular gauge were adjusted by 1 psi to more clearly reflect actual pressures.

3.7.3 Duplicate Pressure Observations All pressure gage readings were performed independently by two separate observers. These readings should be confirmed prior to recording a single value. In the event the observed values remained consistently apart, the mean of the readings was recorded.

In performing any C-factor tests, two pressure gages were used. Prior to flowing the discharge hydrant, the static pressures at each of the residual hydrants was measured and recorded. In order to minimize any potential gage error, the static pressures at each hydrant were measured twice during certain tests, with the gages switched between measurements. The observed pressures should have remained consistent within the specified pressure tolerance (i.e. + - 2 psi). In the event the gage readings were not consistent then the difference was noted on the data


34

collection form prior to their use. This test was performed during the first test and the last test of the day to confirm that the gages did not lose their calibration over the course of the tests.

After the static differences was confirmed, the C-factor test was performed twice, with the gages switched between tests. The observed pressures should have remained consistent within the specified pressure tolerance (i.e. + -2 psi). In the event the gage readings were not consistent then the difference was noted on the data collection form prior to their use. This check was performed during the first test and the last test of the day to confirm that the gages did not lose their calibration over the course of the tests.

3.7.4 Adequate Hydrant Discharge In order to insure that sufficient head loss is generated during the C-factor test to allow the accurate calculation of the C-factor, the pressure drop between the two residual hydrants should have been at least 15 psi. If such a pressure drop was not obtained, it was necessary to open additional hydrants, creating more flow, so as to generate a sufficient pressure drop. In the event that a smallpressure drop was observed with an associated unexpected low discharge from the hydrant it was concluded that that there was likely a closed or partially closed valve upstream of the test area. If this occurs, the upstream valves should have been rechecked to make sure that they are opened prior to repeating the test.

3.8 Quality Control for Fire Flow Tests In conducting a fire flow test for the purpose of hydraulic model calibration, a minimum of two hydrants were employed. One hydrant (flow hydrant) was used to discharge flows to the environment while another upstream hydrant (residual hydrant) was used to measure the pressure drop.

3.8.1 Adequate Hydrant Discharge The magnitude of the discharge from the hydrant should be sufficient to ensure a pressure drop in the residual hydrant of at least 10 psi. In the event that such a drop was not achieved, then a second downstream hydrant may need to be flowed simultaneously with the first. In this case, both discharge hydrants needed to be instrumented with flow/pressure meters. If a smallpressure drop was associated with an unexpected low discharge from the hydrant it is possible that there is a closed or partially closed valve upstream of the test area. When this occurred, the upstream valves were checked to make sure that they were opened prior to repeating the test.

3.8.2 Discharge Measurement Most hydrant flow/pressure gages come with two scales, one for discharge and one for pressure. The discharge scale is only applicable for certain types of hydrant nozzles. As a result, the


35

discharge scale should not be used. Instead, the discharge pressure was measured and then converted into discharge using the discharge equation (shown in Appendix D.2.3 Fire Flow Calculations on page 74).

In some cases, the accuracy of the results cannot be determined on site due to the time needed to input the collected data into KYPIPE. Once the data is entered into KYPIPE, there may be additional errors with the data that were not readily identified in the field. An example would be if the computer model produced a low Hazen Williams Coefficient such as 40 or below. This would indicate that there may have been a valve closed in the system or that there was error in the C-factor test data. These errors were reviewed by the principal investigator and a course of action was determined based upon the complexity of the situation.

3.8.3 Fire Flow Test Validation The City of Nicholasville has previously run fire flow tests on many of their existing hydrants. In the event that one of the hydrants used in this study corresponded to one of these previously tested hydrants, the previous fire flow results were obtained and compared with the results from the new fire flow test. Prior testing information usually contains the available fire flow at a 20 psi residual. In the event that these results were significantly different (e.g. significantly lower), the field crew checked to ensure that there were no closed or partially closed valves upstream of the test area. In the event that such errors were identified, then the fire-flow tests were re-run. In the event that no such valves could be located, the field team noted the discrepancy and attempted to develop a hypothesis for the difference.

While every attempt was made to ensure that the system geometry of the computer model was correct and that there were no closed or partially closed valves upstream of the test area, such errors may not be readily apparent until after the collected data are entered into the computer and the model used to predict the observed pressures and flows. When such an analysis required a roughness coefficient excessively lower than those observed during the C-factor test, the most likely reason is due to errors in the system geometry or the existence of closed or partially closed valves. In the event that such errors are determined, then the fire flow tests were repeated.


36

4.0 Distribution System Model

4.1 General

A hydraulic model representative of the current water distribution system in Nicholasville was created using the KYPIPE Program (Pipe 2010). KYPIPE was developed by civil engineering professors at the University of Kentucky. The program allows users to create a model of a system comprised of pipe links, internal nodes, and end nodes. The point where pipes intersect is represented as a junction, and locations where a demand occurs are shown as nodes. Background maps and drawings can be input in vector and raster formats.

The model can be created to precisely match the conditions present in the system. The program can be used to simulate numerous different scenarios in the system, analyzing the network through an iterative process utilizing the mass balance concept. The process provides results for pressures, velocities, hydraulic grade lines, etc. in pipes and nodes throughout the system. The program can be utilized to analyze both steady state and extended period simulations.

4.2 Development of System Schematic

4.2.1 General Procedure

A system schematic illustrates the system of pipes and other components in a water distribution network. Various categories of data regarding the system are required in order to create an accurate model. These categories are classified as geographical information, facilities data, operational data, and demand data (AWWA, 2005).

Geographical data is used to establish the physical location of the model, including aspects like jurisdictional boundaries and street centerlines. Facilities data includes all the attributes of the pipes, pumps, tanks, and reservoirs in the system. For example, parameters describing the pipes that are required for an analysis include diameter, length, and pipe roughness. This data is the core component of a hydraulic model. Operating data included attributes of the system that are subject to change, such as flow rates, valve/pump controls, valve/pump status, and fixed pressures that create boundary conditions in the system. Demand data is the amount of water consumption assigned to all demand nodes throughout the system (AWWA, 2005).

4.2.2 Elevation Data

Modeling software has the ability to graphically present results of a system analysis, so it is important that input data contains the geographical (x-y) coordinates and elevation (cartesian z-coordinates) for each node. An ArcGIS (Geographic Information System) file of the system used by the utility can be used in the modeling software, and the coordinate system from ArcGIS will ensure spatial compatibility. It is also important for elevations to be accurate to ensure that various calculations for pressures, C-factors, and other attributes are accurate.


37

Two types of elevations are used in a hydraulic analysis, control elevations and ground elevations. Control elevations are located at critical components of the system, such as a pump. The pressures at these locations are critical in model calibration, so it is important that these elevations are accurate. The elevation should be measured exactly where the pressure gage is located, and surveying is commonly used to gather accurate elevation data for these points. Control elevations in this system were considered to be pumps, storage tanks, the water treatment plant, and all hydrants used in both C-factor and fire flow tests.

Because it is crucial to have accurate elevations of hydrant for fire flow and C-factor calculations, surveying was used to determine the elevations of the hydrants used in hydraulic testing. The surveying procedures for C-factor tests are shown in Appendix B.1.1, and the surveying procedure for fire flow tests are outlined in Appendix B.2.1. Telemetry data was utilized to input accurate elevation data for the elevated storage tanks, and a plan set of the water treatment plant was used to gather information for the high service pumps.

Ground elevations include the elevations of the remaining nodes throughout the system, such as typical demand nodes. These elevations are not as critical, but it is recommended that elevations be accurate within 1.5 feet (AWWA, 2005). They are used for calculating available delivery pressures in the system.

Elevations were also established for these remaining nodes in the system, those not considered control elevations. Data was extracted from digital elevation models (DEMS), obtained from kymartian.ky.gov, and input to ARCGIS. These digital elevations models were extrapolated to obtain elevations for all remaining nodes in the system.

4.2.3 Facilities Data

Facilities data usually remains fairly constant in an analysis. When entering data for each pipeline in the model, the pipe diameter, length, and initial roughness coefficient (estimated based off C-factor tests) are needed. Manufacturers of pipes will provide typical roughness coefficients for new pipes of a certain material. These coefficients will generally remain accurate for several years, until the effects of corrosion, encrustation, and biofilm buildup will cause the roughness inside of the pipe to change (AWWA, 2005). This buildup with time that occurs inside of pipes will also cause the diameter of the inside of the pipe to decrease, causing a slight difference between the nominal diameter and the actual diameter. However, the actual diameter of the pipe is difficult to measure, so the nominal diameter should be used as input in the model. The calculations for the C-factor found for each pipe segment tested will account for these changes in pipe diameter (AWWA, 2005).

Pumps in the distribution system are important for filling storage tanks and also providing adequate pressure in the system. A pump characteristic curve shows the relationship between discharge pressure and flow for a particular pump. A pump manufacturer provides the utilities with this curve upon installation, though they are of questionable accuracy to actual pump performance. This concept along with wear and stress on the impeller over time will negatively affect pump performance. The pump curve is commonly altered during the calibration process (AWWA, 2005).


38

Storage facilities within the system are needed for water supply and they also supply pressure to the system. The overall capacity of the tanks includes fire and emergency storage, so the total storage will not be available for peak flow periods. The tank geometry, total capacity, freeboard constraints, and minimum/maximum water levels are important parameters to enter into the model (AWWA, 2005). The water levels of storage tanks are monitored throughout testing. In order to gather information about the total water storage in a tank at a certain time, it is important to understand the shape of the tank and the volume of water stored versus water levels. Tank manufacturers develop depth to total depth ratios and volume to total volume ratios based on water depth, and these are used as input data to the model. These ratios for each of the three storage tanks are shown in Appendix C: Tank Information.

4.2.4 Connectivity Errors

When data is imported from ArcGIS into KYPIPE, several of the pipes experienced connectivity issues. Pipes would appear to be connected but upon further investigation it was shown that the pipes would have a separation of a few inches. The majority of these connectivity errors were found and corrected. Additional connectivity issues were found throughout the calibration process and corrected.

4.3 Development of Demand

4.3.1 Demand Allocation

Demand data is input in the model after the layout and facilities data has been accurately set up. In order to input accurate demand data into the system model, metering data for the month of July 2010 was used. Metering data provided the total water usage in gallons for each customer. The metering data was then averaged and converted gallons per minute (gpm). The following calculation to find the average demand in gpm can be seen in Equation 1.

𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨 𝑫𝑫𝑨𝑨𝑫𝑫𝑨𝑨𝑫𝑫𝑫𝑫 (𝑨𝑨𝒈𝒈𝑫𝑫) = 𝑫𝑫𝑨𝑨𝑫𝑫𝑨𝑨𝑫𝑫𝑫𝑫(𝑨𝑨𝑨𝑨𝒈𝒈)𝑴𝑴𝑴𝑴𝑫𝑫𝑴𝑴𝑴𝑴

× 𝟏𝟏 𝑫𝑫𝑴𝑴𝑫𝑫𝑴𝑴𝑴𝑴𝟑𝟑𝟏𝟏 𝑫𝑫𝑨𝑨𝒅𝒅𝒅𝒅

× 𝟏𝟏 𝑫𝑫𝑨𝑨𝒅𝒅𝟐𝟐𝟐𝟐 𝑴𝑴𝑨𝑨𝒅𝒅

× 𝟏𝟏 𝑴𝑴𝑨𝑨𝟔𝟔𝟔𝟔 𝑫𝑫𝒎𝒎𝑫𝑫

(1) ArcGIS was utilized to allocate the metered data. The metered data was compiled into a table that listed the address, city and state of the user as well as the amount of gallon used per month. This table was then imported into ArcGIS. The ArcGIS geocoding tool was then used to match the address shown in the table to their spatial location. Junction nodes, which are used to represent demand, were exported from the KYPIPE model into ArcGIS. Since the number of metered data outnumbered the number of junction nodes, demand was distributed to the nearest junction.

To determine the closest junction, Thiessen polygons were created from the junction nodes using ArcGIS, and a demand node was assigned to a junction if it resided inside that Thiessen polygon.


39

A Thiessen polygon (shown in Figure 14) is a shape determined by the proximity of nodes whereas it creates a region of influence for a particular node without overlapping another node.

Figure 14 Thiessen Polygons

All the demand nodes that were assigned to a junction were summed to yield the total demand at that junction. This grouping process resulted in each junction node in the model accounting for the demand of several households in the nearby vicinity and is all accomplished with the Spatial Join tool in the ESRI Spatial Analyst extension of ArcGIS.

Quality assurance steps were taken to determine if the demand was placed accordingly. Occasionally some of the metered data would not be correctly placed using the geocoding tool and it was necessary to place the demand by hand. The top water users were allocated by hand to ensure proper placement. It was also necessary to check the allocation of demand to each individual junction node. The Thiessen polygon method had its limitations. In some cases demand was allocated to the nearest junction node which would have been located on a very large transmission main as opposed to the actual pipeline that runs parallel to the transmission main. Effort was taken to correct some of these inconsistencies in the demand allocation. Additionally a Jessamine County land use zoning map was used to double check demand allocation. For example if a low residential area had higher than expected metered demand one month it could be an anomaly and further metered records would need to be checked.


40

4.3.2 System Demand

In order to find the total demand in the system, the concept of mass balance was utilized by looking at the inflow, outflow, and changes in storage in the system. Specifically, the outflow from the system subtracted from the inflow to the system equals the change in total storage. Inflow includes water pumped into the system from the supply source, outflow encompasses all water demand throughout the system, and the change in storage refers to the change in tank levels.

The change in storage (water levels of storage tanks) is found using the SCADA system previously discussed. This data is used with the known pump flow inflow to the system recorded at the WTP to calculate the total outflow of water out of the system (system demand).

The Law of Conservation of mass states that the rate of change in storage (S) is equal to the difference in inflow (I) and outflow (O).

𝑫𝑫𝒅𝒅𝑫𝑫𝑴𝑴

= 𝑰𝑰 − 𝑶𝑶 (2)

𝑑𝑑𝑑𝑑 = (𝐼𝐼 − 𝑂𝑂)𝑑𝑑𝑑𝑑

� 𝑑𝑑𝑑𝑑 = � (𝐼𝐼 − 𝑂𝑂)𝑑𝑑𝑑𝑑𝑑𝑑1

𝑑𝑑𝑜𝑜

𝑑𝑑1

𝑑𝑑𝑜𝑜

∆𝑑𝑑 = � 𝐼𝐼 𝑑𝑑𝑑𝑑 − � 𝑂𝑂 𝑑𝑑𝑑𝑑𝑑𝑑

0

𝑑𝑑

0

Since the pump flows were measured (inflow into the system) at specified times those measurements can be used to estimate the total inflow into the system of the time period between the two measurements by numerical integration; the trapezoidal rule was used for this calculation.

Trapezoidal Rule of Integration:

� 𝐼𝐼(𝑑𝑑)𝑑𝑑𝑑𝑑 ≈ ∆𝑑𝑑 (𝐼𝐼𝑑𝑑𝑖𝑖 + 𝐼𝐼𝑑𝑑𝑖𝑖+1 )

2

𝑑𝑑𝑖𝑖+1

𝑑𝑑𝑖𝑖

The mean value theorem of integration states that there is an average value of a function (in this case demand as a function of time) that represents the function over that period. This was used to approximate the average demand over a certain time period.

Mean Value Theorem:

� 𝑂𝑂(𝑑𝑑)𝑑𝑑𝑑𝑑 = 𝑂𝑂�𝑑𝑑𝑖𝑖+1

𝑑𝑑𝑖𝑖(𝑑𝑑𝑖𝑖+1 − 𝑑𝑑𝑖𝑖)

𝑂𝑂� = 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑜𝑜𝑜𝑜𝑑𝑑𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑜𝑜𝑎𝑎𝑎𝑎𝑎𝑎 𝑝𝑝𝑎𝑎𝑎𝑎𝑖𝑖𝑜𝑜𝑑𝑑 𝑑𝑑𝑖𝑖 𝑑𝑑𝑜𝑜 𝑑𝑑𝑖𝑖+1


41

This was used in the equation for conservation of mass to yield our expression for average demand over a given time interval.

𝑂𝑂� =∆𝑑𝑑

�𝐼𝐼𝑑𝑑𝑖𝑖 + 𝐼𝐼𝑑𝑑𝑖𝑖+1�2 − ∆𝑑𝑑

(𝑑𝑑𝑖𝑖+1 − 𝑑𝑑𝑖𝑖)

𝑶𝑶� =𝑰𝑰𝑴𝑴𝒎𝒎+𝑰𝑰𝑴𝑴𝒎𝒎+𝟏𝟏

𝟐𝟐− ∆𝒅𝒅

∆𝑴𝑴 (3)

Example:

Tank Depth (ft) Pump Flows (gpm)

Tank Diameter (ft) Time Lake St Capital Ct Stephens Dr Lake St 60 11:00 33.0 26.2 25.8 4200 Capital Ct 86 12:00 31.0 24.8 24.0 4100 Stephens Dr 68

∆𝑑𝑑 = 60𝑚𝑚𝑖𝑖𝑚𝑚 ∆𝑑𝑑 = �∆ℎ𝑗𝑗

𝜋𝜋4

𝑗𝑗

𝐷𝐷𝑗𝑗2; 𝑗𝑗 = 𝑑𝑑𝑎𝑎𝑚𝑚𝑡𝑡 1, 𝑑𝑑𝑎𝑎𝑚𝑚𝑡𝑡 2, 𝑑𝑑𝑎𝑎𝑚𝑚𝑡𝑡3

∆𝑑𝑑 = (31.0 − 33.0) 𝜋𝜋4

602 + (24.8 − 26.2) 𝜋𝜋4

862 + (24.0 − 25.8) 𝜋𝜋4

682 = −20324.2 𝑎𝑎𝑎𝑎l

𝑂𝑂� =4200 𝑎𝑎𝑝𝑝𝑚𝑚 + 4100 𝑎𝑎𝑝𝑝𝑚𝑚

2+

20324.2 𝑎𝑎𝑎𝑎𝑜𝑜60 𝑚𝑚𝑖𝑖𝑚𝑚

= 4489 𝑎𝑎𝑝𝑝𝑚𝑚

4.3.3 Diurnal Demand Pattern

The KYPIPE computer modeling program allows the user to input varying water demand patterns throughout the day. This feature allows the user to input a particular water demand at any moment in time or it can be placed in intervals if a 24 hour simulation is performed. For example, the majority of the C-factor testing occurred between 9:30 a.m. and 4:30 p.m, whereas the fire flow testing occurred between 6:00 p.m. to 8:00 p.m. For a typical weekday in Nicholasville demand fluctuated ± 30% from the average daily demand depending on the time of day.

Developing a demand pattern can be time consuming and will vary based upon several factors such as seasonal changes, rainfall, temperature, economic conditions, etc. The demand pattern (an average over a three day period) is shown in Figure 15 below. Table 16 shows the calculated demand factor for each day along with the calculated three day average.


42

Figure 15 Demand Factors for 10/10/2012-10/13/2012 The demand patterns displayed in Error! Reference source not found. above shows demand patterns for three weekdays in October that occurred during the same week, with approximately the same average temperature and the same amount of rainfall. This demonstrates the obstacles faced when trying to develop an “average” demand curve for the hydraulic model.

In order to accurately reflect the demand pattern associated with each C-factor test, Fire Flow test and each 24 hour simulation, an excel spreadsheet was created that used a mass balance approach to calculate the demand for a given time. This spreadsheet allows the user of the model to input the tank levels at a given interval and the inflow from the pump which is then used to calculate the diurnal demand pattern. This spreadsheet also goes one step further and normalizes the global demand factor of the diurnal curve so that it can be inserted in the KYPIPE Hydraulic model. The user then has the option to insert the demand pattern without having to go into the model and reallocate all of the demands placed on individual nodes. The data used to calculate the demand factor for each day, including graphs of these results for each of the three days, are shown in Appendix F: Demand Factor Development on page 84.

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

0:00 6:00 12:00 18:00 0:00

Diur

nal D

eman

d Fa

ctor

Time (hr)

Avereage Demand Factors (10/10/2011 -10/13/2011)


43

Table 16 Diurnal Factor Summary

Time (hr)

Diurnal Curve (Oct 13 Thurs)

Diurnal Curve (Oct 11 - Tuesday)

Diurnal Curve (Oct 10 - Monday)

3 Day Average

1:00 0.85 0.90 0.74 0.83 2:00 0.75 0.73 0.75 0.75 3:00 0.80 1.05 0.78 0.88 4:00 0.84 0.57 1.01 0.81 5:00 0.91 0.59 0.57 0.69 6:00 0.91 0.72 0.50 0.71 7:00 1.17 0.92 0.75 0.95 8:00 1.19 1.02 1.45 1.22 9:00 1.27 1.00 0.92 1.06 10:00 1.27 1.05 0.87 1.06 11:00 1.35 1.23 1.22 1.27 12:00 1.03 1.03 1.26 1.11 13:00 1.19 1.02 1.28 1.17 14:00 1.03 1.01 1.21 1.08 15:00 1.05 1.10 1.21 1.12 16:00 0.74 1.03 1.09 0.96 17:00 0.84 1.26 1.06 1.05 18:00 1.11 1.24 1.16 1.17 19:00 1.02 1.06 1.37 1.15 20:00 1.09 1.27 1.29 1.22 21:00 1.04 1.09 1.05 1.06 22:00 1.05 1.09 1.10 1.08 23:00 0.74 1.08 0.72 0.85 0:00 0.75 0.92 0.64 0.77

Therefore for each individual case the correct demand pattern was calculated as seen in the above section.

4.3.4 System Losses

The process of gathering demand data also had to take into account water loss in the system. Real loss is the physical loss of water from the system, usually in the form of leaks. Apparent losses encompass meter inaccuracy, fire hydrant flushing, water plant use, and other unauthorized use. At the time of data collection, Nicholasville was experiencing a water loss that


44

accounted for approximately 13% of their total water usage. This loss was distributed evenly throughout the distribution system, i.e.. every demand node was increased by 13%.

4.4 Pipe Friction Losses

4.4.1 Hazen-Williams Equation

Various equations have been developed to determine the head losses in a pipe due to friction forces. The Hazen-Williams equation is widely used to relate the physical properties and flow parameters of a pipe to the resulting head loss or pressure drop that will occur. A widely used version of the equation in English units is shown below (Mays, 2011).

𝑴𝑴𝑳𝑳 = 𝟐𝟐.𝟕𝟕𝟑𝟑∗𝑳𝑳∗𝑸𝑸𝟏𝟏.𝟖𝟖𝟖𝟖

𝑪𝑪𝟏𝟏.𝟖𝟖𝟖𝟖∗𝑫𝑫𝟐𝟐.𝟖𝟖𝟕𝟕 (4) Where, hL = head loss (ft) L = length of pipe (ft) Q = flow rate (cfs) C = Hazen Williams C-Factor D = diameter of pipe (ft) The C-factor used in the Hazen Williams equation varies for pipes based on pipe material and age of the pipe. Different pipe materials will result in varying C-factors because pipe roughness is dependent on pipe material. Steel and PVC pipes tend to be smoother and result in less friction loss than cast iron pipes (AWWA, 2005).

The C-factor is also dependent on the age of the pipe. New pipes are typically very smooth and have not yet undergone a great deal of corrosion and deposition, resulting in minimal head loss. After time, the pipes will accumulate deposits and experience tuberculation on the interior of the pipe. This reduces the actual inside diameter of the pipe, causing the actual inside diameter to be less than the expected nominal diameter, which allows less water than expected to flow through the pipe. The accumulation of deposits also causes greater frictional head loss from the increased roughness in the pipe. When the C-factor is determined through field measurements, the C-factor can compensate for the change in diameter based on build up in the pipes (Walski, et al.).

In terms of the C-factor coefficient used in the Hazen Williams equation, the frictional head loss experienced in the pipe will increase as the C-factor decreases. Therefore, pipes made out of smoother material, such as PVC, will have higher C-factors than materials with greater roughness vales like cast iron. Similarly, older pipes of the same material that have experienced significant corrosion and deposition will have lower C-factors than new pipes of the same material (AWWA, 2005). If the flow rate remains constant, a smaller C-factor will result in a larger pressure drop in a segment of pipe. Table 17 shows typical C-factors for pipes based on material and age.


45

Table 17 Typical Hazen Williams C-Factor Coefficients

The purpose of the C-factor test is to measure all factors in the Hazen Williams equation during hydraulic testing and then solve for the unknown C-factor. The flow rate is measured in the field, along with parameters to find the corresponding head loss, in order to calculate the unknown C-factor (EPA, 2005).

4.4.2 Minor Losses

The majority of the total head loss through a specific segment of pipe can be attributed to the frictional head loss. However, a portion of head loss through pipes is caused by minor losses. These losses occur because of changes in the geometry of the pipes such as bends, valves, and other fittings. Losses at these fittings are typically minimal for normal velocities in the system (AWWA, 2005). The C-factor calculated through field testing will account for minor losses (EPA, 2005). Even though the effects are minimal, the minor losses will increase the overall head loss measured in the field. Because the C-factor is calculated using field data, the coefficient will encompass both friction losses and minor losses. The added effects of minor losses will cause the C-factor values to decrease.


46

5.0 Model Calibration

5.1 General

The hydraulic model can be very beneficial to the utility in planning and everyday operation of the water distribution system. Like every type of model, the objective is to predict future conditions based on known system data; and the quality of a model is its ability to accurately predict those conditions. Calibration of a model involves the post diction of known system conditions based on simultaneously collected system data; all of which are gathered during the fire flow tests, C-factor tests, and use of 24 hour SCADA data. The system data (which are used as inputs) are tank levels, pump settings (on or off), valve closures and water demand. The model outputs are system pressures, pipe flows and pump operating conditions; namely discharge pressure and flow.

Calibration involves adjusting system demand distribution, pipe roughness, pump curve and other model attributes (Walski, et al.). However, it is still important that the data is adjusted only within reasonable limits. For example, changing a C-factor value of a pipe outside of reasonable values based on the pipe material and age might seem like appropriate calibration in a particular circumstance, but would probably result in unlikely results for a new range of conditions. The process of calibration can also reveal undocumented changes to the pipe system such as additional pipe connections, closed valves, severely tuberculated pipes, missing pipes and other issues that can be resolved to improve operation of the system (EPA, 2005).

Once the model accurately predicts field measurements under a wide range of conditions, the model is considered to be calibrated. It was desired to achieve relative convergence between values measured and those predicted in the model, specifically for static and residual pressures, pump discharge pressure, pump outflows, etc. However, this was to be achieved without producing unrealistic system conditions, such as C-factors for pipes out of the reasonable range based on pipe material and age. To further verify the calibrated model, an extended period simulation of a typical day, using measured demands and change patterns (valve closures and turning the pumps on and off), is compared with the actual tank levels.

The primary activities of the hydraulic calibration are pipe roughness adjustments, demand distribution and pump calibration. These activities are only effective as “fine tuning” measures for a much more general procedure. If the broader system details such as tank elevations and total system demand are not accurate, the pipe roughness will have little impact on converging model results and measured values. A thorough investigation of system attributes and operating conditions prior to calibration is paramount in achieving timely and acceptable results.

However, working with an un-calibrated model can lead the modeler to questionable data sources, inconsistent testing procedures, or omitted operations procedures (such as active valve and pump schedules). Upon finding input data errors, such as previously undocumented pipe changes or erroneous instrument measurements, the modeler should re-measure all possible values with redundancy (repeating tests if necessary) to verify the error and any new measurements.


47

5.2 Calibration Methods

5.2.1 Calibration Setup Hydraulic calibration is a long process that involves a lot of changes and, so far, is not well automated by programming (AWWA, 2005). The calibration started by using the results of the C-factor tests as input for pipes in the model to be used as a starting point for the actual calibration. As mentioned earlier each pipe was assigned an individual pipe calibration group 0-9. The results of the C-Factor test were used to assign a C-factor to each pipe calibration group before additional calibration was performed.

On this step was complete The Fire Flow data was input into the system by setting up ten cases in KYPIPE, one for each fire flow test, so as to apply the appropriate boundary condition and demand patterns. The boundary conditions were setup as change patterns, which override the setting at specified nodes or pipes with a new value that is applied to that case. For example FF-9 was modeled as case 9, and the HGL for all of the tanks are “changed” to the HGL’s recorded for that test for KYPIPE to use those in the hydraulic analysis. If not performing extended period simulations, KYPIPE uses these demand patterns and change patterns as a series of steady state simulations.

To model the fire flows, junction nodes were added at the locations in the model of the hydrants used in testing. In the change pattern, the newly assigned junction nodes were given a demand equivalent to flow produced from the flowing hydrant during the Fire Flow test In order to analyze both static and residual pressures, two sets of change data were used that were exactly the same except one changed the demand at these junctions to zero.

Using this procedure, KYPIPE will report the results of all the simulation runs (all the fire flow tests) simultaneously. This setup allows the modeler to easily see the effect of changes on all tests; which is advantageous since changing pump or pipe attributes will affect all other simulation runs, if only slightly.

5.2.2 Pump Curve Calibration

In order to calibrate the high service pumps at the water treatment plant, the manufacturer’s pump curves were obtained. The manufacturer’s pump curve for all high service pumps are shown in Appendix A: Pump Curves on page 57. The pump curve is subject to change due to wear and stress on the impeller over time. If pumps are not tested periodically to update the pump curve, the actual pump curve could vary dramatically from the curve provided by the manufacturer. This attribute is commonly altered during the calibration process (AWWA, 2005). Nicholasville WTP had recently undergone improvement within the past 5 years to two of their high service pumps and therefore the pumps were reasonably close to the curve the manufacture had provided. Data was collected from the SCADA system that measured both the discharge pressure of the pump and the flow of the pump to help fine tune the pump curves. The suction pressure of the pump was calculated from the hydraulic grade line of the clear wells. Nicholasville had 5 high service pumps. One of the pumps was only used as a backwash pump


48

and was not used for the hydraulic model. The results of the finalized pump curves can be shown below in Table 18.

Table 18 Finalized Pump Curves .

Pump- 1 and 2 Pump- 3 Pump- 4 Head (ft)

Flow (GPM)

Efficiency (%)

Head (ft)

Flow (GPM)

Efficiency (%)

Head (ft)

Flow (GPM)

Efficiency (%)

600 200 0 550 800 40 500 1120 50 500 1600 78 500 1360 55 470 1480 60 490 1700 81 470 1760 65 450 1760 65 470 1800 82 450 2100 72 430 2160 75 460 1900 82.5 430 2600 80 410 2600 80 440 2000 83 410 2840 83 390 2880 83 420 2100 83 380 3200 83.5 360 3200 84 400 2200 82 340 3500 83.5 320 3520 83 380 2300 81 320 3640 82.5 300 3700 82 350 2400 78 290 3900 81 270 3900 80 300 2600 73 240 4200 73 240 4100 77

190 4400 65 210 4320 71

5.2.3 Macro Level Calibration

The calibration of the Nicholasville model involved incremental changes in pipe roughness. The pipes were changed as groups, and these groups are classified by diameter, age and material as previously shown. This aids calibration, as pipes of similar size and material should have similar roughness. However, pipes with similar attributes will wear differently over the years depending on their location in the system, causing their roughness values to diverge. The initial model used the results from the C-factor tests for each calibration group so as to develop a baseline from which to make adjustments.

Once the C-factors from the C-factor testing were input and the pump curve calibrated, a macro calibration was performed on the system. The macro calibration consists of entering all 10 fire flow test and measuring such parameters as the static pressures and the corresponding residual pressures for the given fire flow. “If any of the measured state variables are different from the modeled valves by 30% or more, then it is likely that the cause for the difference may extend beyond errors in the estimates for pipe roughness or nodal demand. Possible causes for such differences are many but may include: 1) closed or partially closed valves, 2) inaccurate pump curves or tank telemetry data, 3) incorrect pipe diameter, 4) incorrect pipe length, 5) incorrect geometry ” (Lindell Ormsbee, 1997).It was discovered through the macro calibration that several of the modeled pressures were excessive in comparison to the measured pressures.


49

The data in the system was reviewed and a wide range of fixes were applied to the model. Some of the fixes include but are not limited to:

• Connectivity errors- pipes appear to be connected but were not • Incorrect shape data for tank- input error • Demand allocation- demand had been placed on the wrong pipe • Demand pattern- global demand factor was incorrect • Incorrect pump operations- wrong pump was turned on • Incorrect flow entered – input error • Broken valve was found • Incorrect hydrant elevation • Incorrect placement of hydrant

After the macro calibration was performed, the modeled pressures were much closer to the measured pressures.

5.2.4 Model Sensitivity Analysis

A sensitivity analysis was also conducted on the model to help identify the most likely source of model error. This analysis was accomplished by varying different model parameters by different amounts and measuring the associated effect. The KYPIPE program has a build in algorithm that performs a global adjustment factor to say a pipes roughness or nodal demand value. The calibration routine performed and optimization on the calibration trying to adjust a given C-factor for a pipe to ultimately create a smaller source of error between the measured pressure data and the modeled pressures. The calibration routine also allowed the ability to place a tolerance on the value of the fire flow and the demand of the system.

The results of this analysis showed that system was more susceptible to the sensitivity of the demand compared to anything else. It was also discovered that the C-factors didn’t play as significant an impact as previously thought. As a result of this analysis, greater efforts were applied to accurately depicting the demands of the system.

5.2.5 Micro-Level Calibration

The micro calibration was performed into two different steps. The first step of the micro calibration was a steady state analysis and the second was a 24 hour extended period simulation.

The steady state analysis consisted of fine tuning the C-factors of individual pipes through a trial and error approach. Once the C-factors were assigned to individual pipes a 24 hour extended period simulation (EPS) was performed to determine how well the model parameters predicted the flows and tank levels. The EPS allowed the modeler to determine if tanks were filling at the appropriate level. In some cases the tanks would all be over filling which indicates an incorrect system demand. Another scenario is when one tank is filling faster than another tank which would indicate incorrect C-factors or incorrect demand pattern in a particular region of the system.


50

The Micro level calibration consisted of an iterative process until the model reasonably predicted pressures in the steady state analysis as well as tank levels in the extended period simulation.

5.3 Calibration Results

5.3.1 Final Calibrated C-Factors

The calibration process resulted in alterations to the C-factors for pipes in the system model. These new C-factors, developed for each calibration group, were assigned to pipes in the model and varied slightly from the original C-Factors calculated from the C-Factor field tests. These results are shown below in Table 19.

Table 19 C-Factor Calibration Results

Group Pipe Material Diameter (in)

Average Age (yr)

Age Low End (yr)

Age High End (yr)

AssignedC-Factor

1 Asbestos Cement 4, 6 40 N/A 40 123.5 2 PVC 2,3 15 2 30 111 3 PVC 4,6 15 2 30 131 4 PVC 8,10,12 5 1 10 139 5 Ductile Iron 10,12,16 30 20 45 98 6 Ductile Iron 6, 16,20,24 15 15 20 112.6 7 Asbestos Cement 8,10 40 N/A 40 115.81 8 Cast Iron 4,6 40 40 60 99.4 9 Cast Iron 8,10,12 40 40 60 117.4

5.3.2 Comparison of Pressures between Model and Field Tests

Results from each fire flow test performed were also found using the model and compared to real field results. The values for pressure (for residual and flow hydrants) were observed for both model and field results, and the difference between the measured and predicted values were calculated. The results are shown below in Table 20.


51

Table 20 Fire Flow Calibration Results

Site Data Source

Residual Hydrant Flow Hydrant Static

Pressure (psi)

Difference (psi)

Residual Pressure

(psi)

Difference (psi)

Static Pressure

(psi)

Difference (psi)

FF-1 Field 99 0.3 85 4.5 102 3.6 Model 98.7 80.5 98.4

FF-2 Field 88 -1.9 82 3.2 92 3.3 Model 89.9 78.8 88.7

FF-3 Field 92 -2.3 64 4.5 94 3.1 Model 94.3 59.5 90.9

FF-4¹ Field 94 0 79 3 92 2.4 Model 94 76 89.6

FF-5 Field 56 -1 44 -1.4 46 -0.3 Model 57 45.4 46.3

FF-6 Field 72 -1.5 64 -0.3 73 1.9 Model 73.5 64.3 71.1

FF-7 Field 101 1.5 91 2.1 98 -1.8 Model 99.5 88.9 99.8

FF-8 Field 69 -0.8 62 1 75 1.9 Model 69.8 61 73.1

FF-9 Field 83 -0.9 78 -0.2 73 1.7 Model 83.9 78.2 71.3

FF-10 Field 84 1.3 77 0.2 82 0.9 Model 82.7 76.8 81.1

Note 1: FF-4 was located near the water systems largest demand users. During Fire Flow testing, alarms were set off and the factories demand was decreased significantly. Thus the demand data of the factory was decreased in the model to represent the demand conditions during this particular fire flow event.

The percent difference between model and field results for pressure measurements was also calculated, and the results are shown below in Table 21. It was desired to have relative convergence between model and field results, and Table 21 shows the percent difference were reasonable. All data collected during fire flow tests (including model results) used to calculate the pressure differences shown below can be viewed in D.2.2 Fire Flow Data on page 73.


52

Table 21 Percent Difference between Model and Field Data

Residual Hydrant Flow Hydrant

Site Percent

Difference (Static Pressure)

Percent Difference (Residual Pressure)

Percent Difference (Static

Pressure) FF-1 0.3% 5.3% 3.5% FF-2 2.2% 3.9% 3.6% FF-3 2.5% 7.0% 3.3% FF-4 0.0% 3.8% 2.6% FF-5 1.8% 3.2% 0.7% FF-6 2.1% 0.5% 2.6% FF-7 2.1% 2.3% 1.8% FF-8 1.2% 1.6% 2.5% FF-9 1.1% 0.3% 2.3% FF-10 1.5% 0.3% 1.1%

5.4 Model Validation

5.4.1 24 hour-EPS Simulation

In order to validate the model calibration process, an Extended Period Simulation (EPS) was performed on the calibrated model. A 24 hour period was examined in 1 hour intervals. Specifically, the water levels at all three storage tanks, and pump flow rates were examined. SCADA data was collected for a few days, and the data for tank levels and flow rate was compared to the EPS performed on the system model. This comparison can be seen graphically; the water levels in each tank measured by the SCADA system and the levels predicted in the model EPS for October 10, 2011 are shown in Figure 16. The same comparison for October 11 along with October 13 is shown in Figure 17 and Figure 18.

The goal of the EPS is to show that all values predicted by the model are reasonably close to measurements taken during field testing. The exact values for water levels of the tanks under the EPS simulation and recorded by the SCADA system, along with the calculated differences in water level, are shown in Appendix E: Model Validation on page 81 for each of the three days.


53

Figure 16 EPS vs. SCADA Data for 10/10/2011

18

20

22

24

26

28

30

32

34

36

38

40

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Tank

Ele

vatio

n (ft

)

Time (hr)

EPS vs. SCADA Data (Tank Levels) 10/10/2011

Stephens Drive-SCADALake Street-SCADACapital Ct- SCADA

EPS-Stephens DriveEPS Lake Street

EPS-Capital Ct


54

Figure 17 EPS vs. SCADA Data (Tank Levels) 10/11/2011

Figure 18 EPS vs. SCADA Data 10/13/2011

10121416182022242628303234363840

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Tank

Ele

vatio

n (ft

)

Time (hr)


Stephens Drive-SCADALake Street-SCADACapital Ct-SCADA

EPS-Stephens

EPS-Lake Street

EPS-Capital Ct

18

20

22

24

26

28

30

32

34

36

38

40

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Tank

Ele

vatio

n (ft

)

Time (hr)


Stephens Drive-SCADALake Street-SCADACapital Ct-SCADAEPS-Stephens DrEPS- Lake StreetEPS- Capital Ct


55

6.0 Summary Water utilities would greatly improve their ability to make operational decisions regarding their distribution system with a solid understanding of their system flow dynamics. Decisions regarding everyday operation along with system improvements can have significant impact on the community and require substantial investment as well. A distribution system model is a helpful tool for simulating the behavior of a system under various conditions, but it is important that the model be an accurate representation of the actual conditions in the system.

The calibration process will ensure that the model is able to accurately predict system behavior. Once hydraulic tests are executed to gather information about the system, these field results are compared to behavior predicted by the developed hydraulic model. The model developed prior to calibration encompasses all known information about the system, which would provide a fairly reasonable representation of system behavior. During the calibration process, data within the model is adjusted until behavior predicted by the model reasonably agrees with measured system performance over a range of operating conditions. This causes the model to include parameters of the system that are unknown or altered over time, such as closed valves, weakened pump performance, increased roughness in pipes over time, etc. These adjustments made to the model allow the utility to observe flow dynamics and behavior of the system accurately through model simulations. Because the model will closely match true conditions of the system, the model will be a critical tool for the utility.


56

7.0 Works Cited Aquacraft, Inc. . (1998). Comparison of Demand Patterns Among Residential and CI Customers

in Westminster, Colorado. Westminster: City of Westminster Department of Water Resources.

AWWA. (1989). Installation, Field Testing, and Maintenance of Fire Hydrants. Denver: American Water Works Association.

AWWA (Director). (1999). Field Guide: Hydrant Flow Tests [Motion Picture]. AWWA. (2005). Computer Modeling of Water Distribution Systems. Manual of Water Supply

Practices- M32. Denver: American Water Works Association. ECAC. (1999). Calibration Guidelines for Water Distribution System Modelling. Proc. AWWA

1999 Imtech Conference. Engineering Computer Applications Committe. EPA. (2005). Water Distribution System Analysis: Field Studies, Modeling, and Management.

Cincinnati: U.S. Environmental Protection Agency. Johnson, R. P., Blackschleger, V., Boccelli, D. L., & Lee, Y. (2006). Water Security Initiative

Field Study: Improving Confidence in a Distribution System Model. Cinncinatio, Ohio: CH2M HILL.

Kentucky Administrative Regulations. (n.d.). Surface water standards. 401 KAR 10:031. Frankfort: Kentucky Administrative Regulations.

Lindell Ormsbee, S. L. (1997). Calibration of Hydraulic Network Models. American Water Works Association(89), 42-50.

Lingireddy, S. O. (2005, July 15). Calibration of Hydraulic Network Models. Water Encyclopedia, pp. 313-320.

M. L. Maslia, J. B. (2005). Use of Continuous Recording Water-Quality Monitoring Equipment for Conducting Water-Distribution System Tracer Tests: The Good, the Bad, and the Ugly. ASCE/EWRI Congress. Anchorage.

Maslia, M. (2004). Field Data Collection Activities for Water Distribution System Serving Marine Corps Base, Camp Lejeune, North Carolina. Atlanta: Agency for Toxic Substances and Disease Registry.

Mays, L. W. (2011). Water Resources Engineering. John Wiley & Sons, Inc. McEnroe, B. C. (1989). Field testing water mains to determine carrying capacity. Vicksburg:

Environmental Laboratory of the Army Corps of Engineers Waterways Experiment Station.

Ormsbee, L. E. (2005, July 15). Calibration of Hydraulic Network Models. Water Encyclopedia, pp. 313-320.

Walski, T. (2000). Model Calibration Data: The Good, the Bad and the useless. Journal of American Water Works Association, 94.

Walski, T. M., Chase, D. V., Savic, D. A., Grayman, W., Beckwith, S., & Koelle, E. (n.d.). Advanced Water Distribution Modeling and Management. Bentley Institute Press.


57

Appendix A: Pump Curves The pump curves provided by the manufacturer for each of the five high service pumps are shown below in Figure 19 through Figure 23.

Figure 19 Manufacturer’s High Service Pump #1 Curve


58




59

Figure 22 Manufacturer’s High Service #4 Pump Curve

Figure 23 Manufacturer’s High Service #5 Pump Curve


60

Appendix B: Surveying Procedures and Data

B.1 C-Factor Surveying

B.1.1 C-Factor Survey Procedure

During the C-Factor Testing prescribed for this project, it was necessary to determine the difference in elevation between hydrants. The following provides an outline of the appropriate procedure for performing a C-Factor Hydrant Elevation Survey. NOTE: Prior to any surveying activities, the proper care and operation of the total station and its accompanying equipment should be studied and reviewed.

1) Identify the hydrants that are designated as the flow and the residual and position a Leica TC400NL Total Station and its tripod so that the machine can have a clear line of sight to both hydrants.

2) Level the total station and measure the instrument’s height using a tape measure, yard stick or similar device. Duplicate this height on the prism rods. In situations where it is impractical or undesirable for the instruments and rods to have the same height, record each individual height for use in future calculation.

3) At this point, the total station is turned on and the rods are placed at their respective hydrants. Place the rods on top of the nut located in the center of the desired flow nozzle. This will approximate the elevation at the center of the hydrant’s flow.

4) Once the rods are placed and steady, the total station operator can take a measurement by sighting the center of the prism and pressing the “DISP” button on the instrument. After a few moments, the total station will display the slope distance, horizontal distance, vertical angle and the vertical distance between the prism and the instrument’s sight. The vertical distance should be recorded for this hydrant on the C-Factor Surveying Data Log.

5) Step 4 should be repeated, leaving the total station in place and simply turning it towards the second hydrant.

6) Once the vertical distances for both hydrants have been measured and recorded, elevations will be assigned to each residual hydrant to differentiate which hydrant is located at a higher elevation (with the lower elevation being assigned a 0 ft elevation). The difference between elevations of the residual hydrants will be calculated and recorded on the same C-Factor Surveying Data Log.

B.1.2 C-Factor Surveying Results

Table 22 shows the elevation difference and distances between residual hydrants needed for C-factor calculations.


61

Table 22 C-Factor Surveying Results

Site ID

Flow Hyd ID

Residual Hyd #1

ID

Residual Hyd #2

ID

Elevation Residual

#1 (ft)

Elevation Residual

#2 (ft)

Elevation Difference (Elev 2 -

Elev 1) in ft

Distance between Residual

Hydrants (ft) C-1 F-1 P-21 P-1 958.511 962.036 3.525 424 C-2 F-2 P-22 P-2 927.315 936.087 8.772 769 C-3 F-3 P-23 P-3 938.591 919.705 -18.886 809 C-4 F-4 P-24 P-4 922.197 916.64 -5.557 682 C-5 F-5 P-25 P-5 893.655 882.335 -11.32 2236 C-6 F-6 P-26 P-6 948.241 949.591 1.35 372 C-7 F-7 P-27 P-7 1009.39 989.81 -19.58 317 C-8 F-8 P-28 P-8 914.596 921.091 6.495 443 C-9 F-9 P-29 P-9 916.782 907.06 -9.722 492 C-10 F-10 P-30 P-10 935.213 955.51 20.297 763

B.2 Fire Flow Surveying

B.2.1 Fire Flow Survey Procedure

During the Fire Flow Testing prescribed for this project, it was necessary to determine the absolute elevations of the hydrants involved in this testing. The following provides an outline of the appropriate procedure for performing a Fire Flow Hydrant Elevation Survey. NOTE: Prior to any surveying activities, the proper care and operation of the total station and its accompanying equipment should be studied and reviewed.

1) In the area where these tests will be performed, it will be necessary to locate Geodetic Benchmarks to determine the elevations. To accomplish this, surveyors should proceed to the website http://benchmarks.scaredycatfilms.com to find possible locations. http://www.ngs.noaa.gov/ should also provide helpful, more detailed descriptions of individual benchmarks.

2) Proceed to the physical areas where the most useful benchmarks have been identified and determine whether they are readily available or accessible. If not, the surveyors may employ the use of a metal detector and/or a shovel to uncover the desired marker if it is determined to be appropriate.

NOTE: Circumstances involving the actual location of a benchmark vary, so respect and care for private property rights and personal safety should be considered and observed at all times. For instance, if a benchmark is believed to be located in someone’s property, surveyors should not continue their search for said benchmark without permission from

http://www.ngs.noaa.gov/


62

the property owner. Likewise, if a benchmark is located in the middle of a road or some other similarly hazardous area, practical judgment should be used to avoid placing surveyors or equipment in danger.

3) Once a useable benchmark has been located, surveyors should use standard surveying procedure known as leveling to proceed from the benchmark to the desired hydrant locations. This will involve identifying the most accessible and efficient path to take between the benchmark and the hydrants, keeping in mind that the fewer shots that can be taken, the less error can be introduced into the elevation measurements.

4) Position a Leica TC400NL total station and tripod so that it has a clear line of sight to the benchmark and to a point along the path toward the desired hydrant location where a prism rod can be placed.

5) Place the total station on its tripod and make sure the instrument is level.

6) Once level, the height of the instrument should be measured using a tape measure, yard stick or similar device and should then be duplicated on the prism rods. In situations where it is impractical or undesirable for the instruments and rods to have the same height, record each individual height for use in future calculation.

7) At this point, the instrument can be turned on and the rods can be placed at their desired positions – one on the benchmark and one along the path toward the hydrant.

8) Once the rods are placed and steady, the total station operator can take a measurement by sighting the center of the prism located at the benchmark and pressing the “DISP” button on the instrument.

9) After a few moments, the total station will display the slope distance, horizontal distance, vertical angle and the vertical distance between the prism and the instrument’s sight. The vertical distance should be recorded as the “Backsight.”

10) Leaving the total station in place and simply turning it towards the second prism, Step 4 should be repeated. The vertical distance here should be recorded as the “Foresight.”

11) Subtract the Backsight from the Foresight to determine the difference in elevation between the two points and thus the total elevation of the second prism’s location.

12) Repeat steps 4-10 by backsighting to the second prism’s location and foresighting to another point further along the path to the desired hydrant’s location. This should be repeated as many times as is necessary to reach the final location.

13) Once in position to measure the elevation of the fire flow hydrants, place the rods on top of the nut at the center of the hydrant’s desired flow nozzle to approximate the elevation at the center of the flow.

14) Record the elevations of both the flow and residual fire flow hydrants on the Fire Flow Surveying Data Log to use in future calculations.

NOTE: It is good practice to take accurate field notes throughout this procedure to make interested parties aware of any special circumstances involved in the surveyors’ measurements and keep track of any error introduced therein.


63

Appendix C: Tank Information The following tables show volume and depth ratios based on water levels for each storage tank.

Table 23 Lake Street Tank Data

750,000 gal1025.751143.75

40Ovaloid

60 ft

Depth (ft)

Volume (ft³)

Ratio of Depth to

Total Depth

Ratio of Volume to

Total Volume 0 0 0 0.00001 12044 0.025 0.01612 25732 0.05 0.03433 40826 0.075 0.05444 57109 0.1 0.07615 74435 0.125 0.09926 92564 0.15 0.12347 111260 0.175 0.14838 131025 0.2 0.17479 151158 0.225 0.201510 171950 0.25 0.229311 192632 0.275 0.256812 213747 0.3 0.285013 235000 0.325 0.313314 255385 0.35 0.340515 275769 0.375 0.367716 296154 0.4 0.394917 316533 0.425 0.422018 336923 0.45 0.449219 357313 0.475 0.476420 377692 0.5 0.503621 398077 0.525 0.530822 418462 0.55 0.557923 438846 0.575 0.585124 459231 0.6 0.612325 479615 0.625 0.639526 500000 0.65 0.666727 521133 0.675 0.694828 542246 0.7 0.723029 563129 0.725 0.750830 583129 0.75 0.777531 604027 0.775 0.805432 623818 0.8 0.831833 643046 0.825 0.857434 661557 0.85 0.882135 679241 0.875 0.905736 695992 0.9 0.928037 711658 0.925 0.948938 726041 0.95 0.968139 733900 0.975 0.978540 750000 1 1.0000

Inside Diamter

Depth of TankShape

Elevation of Bottom of the TankMax Level

Lake Street Tank Data Size


64

Table 24 Capital Court Tank Data

1,500,000 gal952.5115139.5

Composite86 ft

Depth (ft)

Volume (ft³)

Ratio of Depth to

Total Depth

Ratio of Volume to

Total Volume 1 14185 0.0253 0.00952 31849 0.0506 0.02123 53127 0.0759 0.03544 76900 0.1013 0.05135 102196 0.1266 0.06816 129063 0.1519 0.08607 157546 0.1772 0.10508 187694 0.2025 0.12519 219553 0.2278 0.1464

10 253169 0.2532 0.168811 288591 0.2785 0.192412 325865 0.3038 0.217213 365037 0.3291 0.243414 406156 0.3544 0.270815 449105 0.3797 0.299416 492413 0.4051 0.328317 535722 0.4304 0.357118 579031 0.4557 0.386019 622340 0.4810 0.414920 665648 0.5063 0.443821 708957 0.5316 0.472622 752266 0.5570 0.501523 795575 0.5823 0.530424 838884 0.6076 0.559325 882192 0.6329 0.588126 925501 0.6582 0.617027 968810 0.6835 0.645928 1012119 0.7089 0.674729 1055428 0.7342 0.703630 1098736 0.7595 0.732531 1142045 0.7848 0.761432 1185354 0.8101 0.790233 1228663 0.8354 0.819134 1271971 0.8608 0.848035 1315280 0.8861 0.876936 1358589 0.9114 0.905737 1401898 0.9367 0.934638 1445207 0.9620 0.963539 1488401 0.9873 0.9923

39.5 1509572 1.0000 1.0064

Inside Diamter

Depth of TankShape


Capital Court Tank Data Size


65

Table 25 Stephens Drive Tank Data

750,000 gal966.5114835

Ovaloid68 ft

Depth (ft)

Volume (ft³)

Ratio of Depth to

Total Depth

Ratio of Volume to

Total Volume 0 0 0.0000 0.00001 5928 0.0286 0.00792 14133 0.0571 0.01883 24558 0.0857 0.03274 37428 0.1143 0.04995 52128 0.1429 0.06956 67773 0.1714 0.09047 86160 0.2000 0.11498 106058 0.2286 0.14149 127234 0.2571 0.1696

10 149533 0.2857 0.199411 172743 0.3143 0.230312 196815 0.3429 0.262413 221242 0.3714 0.295014 247242 0.4000 0.329715 273206 0.4286 0.364316 299602 0.4571 0.399517 326337 0.4857 0.435118 353302 0.5143 0.471119 380409 0.5429 0.507220 407569 0.5714 0.543421 434676 0.6000 0.579622 461641 0.6286 0.615523 488376 0.6571 0.651224 514772 0.6857 0.686425 540736 0.7143 0.721026 566268 0.7429 0.755027 591163 0.7714 0.788228 615235 0.8000 0.820329 638445 0.8286 0.851330 660744 0.8571 0.881031 681920 0.8857 0.909232 701818 0.9143 0.935833 720205 0.9429 0.960334 735850 0.9714 0.981135 750000 1.0000 1.0000

Inside Diamter

Depth of TankShape


Stephens Drive Tank Data Size


66

Appendix D: Data Collection

D.1 C-Factor Test

D.1.1 C-Factor Test Procedure

A step by step procedure for conducting the C-Factor Test is shown below. Hydrant Testing Crew Instructions 1. Test shall be made during a period of ordinary demand. Before testing begins, the

Nicholasville WTP plant will need to be notified of the time of testing. This must occur so the Nicholasville WTP can record the required data regarding tank levels, pump operation schedules, plant flow, etc. during each hydrant flow test.

2. Two hydrants designated the “Residual Hydrants”, will be chosen to collect the normal static pressure while the other hydrant in the group, the “Flow Hydrant”, is closed. The residual pressure will also be collected while the other hydrant in the group is flowing. Record the length between these hydrants (should range between 400 and 1200 feet). If the hydrants are not at the same elevation, height of the hydrants will need to be recorded.

3. One hydrant, designated the “Flow Hydrant”, is chosen to be the hydrant where flow pressure will be observed using a Pitot tube (Hydrant Flow Meter). The Pitot tube to be used for this project is a Pollard P669LF.

4. Once the Flow Hydrant has been selected, a valve directly downstream of the Flow Hydrant should be closed. The valve should be closed slowly to prevent pressure surges and water hammers in the system.

5. At this time the flowing hydrant shall be opened, water should be allowed to flow long enough to clear any debris and foreign substances from stream.

6. A 2 ½” cap with pressure gauge that can read approximately 25 psi greater than the system pressure for the hydrant will be attached to the residual hydrants and each residual hydrant opened full. For this project a Pollard item #P67022LF Hydrant Static Pressure gage will be used. A reading (static pressure) is taken when the needle comes to a rest. Record this reading on the C-Factor Data Collection Log.

7. The Hydrant testing crew members for the residual hydrants will then signal the flowing hydrant crew member using 2 way radio device or cell phone. Attach the Pitot tube to the 2 ½” outlet along with the static pressure gage to a remaining outlet and open hydrant again. The hydrant valve should be opened slowly to prevent pressure surges or water hammer in the system. The hydrant should be flowed approximately 2-5 minutes.


67

a. If dechlorination regulations exist for the selected hydrant then dechlorinating diffuser will need to be connected to the flowing hydrant.

8. Observe the Pitot Gauge and static gage reading and record the pressures at the residual hydrants and the flowing hydrant simultaneously (once readings have stabilized). Proper communication will be needed to achieve simultaneous recording.

9. After adequate readings have been recorded, close the flow hydrant to cease flow. Static pressure readings should be recorded at both residual hydrants simultaneously.

10. Complete the other necessary information on the C-Factor Data Collection Log. 11. Make sure to reopen the previously closed valve before leaving the testing site.

Water Treatment Plant Crew Instruction

1. During Field testing the WTP will collect SCADA data showing the tank levels and pump flows in and out of the system.

2. The WTP will also alert the Field crew in the event that testing needs to be halted due to excessive pumping.

D.1.2 C-Factor Calculations

In order to calculate the C-factor, the procedure shown below was followed. The head loss was first calculated using the pressures recorded at the two residual hydrants after the flow hydrant is opened, along with the elevations of each residual hydrant. The equation used to calculate the head loss between the two residual hydrants was found using the Bernoulli Equation (shown in Equation 5). The hydrant labeled Residual Hydrant #2 on the C-Factor Data Collection Log will be the upstream hydrant, while the hydrant labeled Residual Hydrant #1 is located downstream (as shown in Figure 10).

�𝑷𝑷𝟐𝟐𝜸𝜸

+ 𝒁𝒁𝟐𝟐 + 𝑽𝑽𝟐𝟐𝟐𝟐

𝟐𝟐𝑨𝑨� − 𝑴𝑴𝑳𝑳 = �𝑷𝑷𝟏𝟏

𝜸𝜸+ 𝒁𝒁𝟏𝟏 + 𝑽𝑽𝟏𝟏𝟐𝟐

𝟐𝟐𝑨𝑨� (5)

Where, ℎ𝐿𝐿 = Head loss 𝑃𝑃1 = Residual pressure at Hydrant #1 (downstream hydrant) 𝑃𝑃2 = Residual Pressure at Hydrant #2 (upstream hydrant) 𝑍𝑍1 = Gage elevation at Hydrant #1 𝑍𝑍2 = Gage elevation at Hydrant #2 𝑉𝑉1 = Velocity in pipe at Residual Hydrant #1 𝑉𝑉2 = Velocity in pipe at Residual Hydrant #2 𝛾𝛾 = Specific weight of water g = Acceleration due to gravity


68

The difference in velocity heads between the two residual hydrants is considered negligible. Therefore, only the difference in the pressure head and elevation head between the hydrants was considered when calculating the head loss.

𝑴𝑴𝑳𝑳 =

(𝑷𝑷𝟐𝟐−𝑷𝑷𝟏𝟏)∗

⎝

⎜⎛𝟏𝟏𝟐𝟐𝟐𝟐𝒎𝒎𝑫𝑫

𝟐𝟐𝒇𝒇𝑴𝑴𝟐𝟐�

⎠

⎟⎞

𝜸𝜸+ (𝒁𝒁𝟐𝟐 − 𝒁𝒁𝟏𝟏) (6)

Where, ℎ𝐿𝐿 = Head loss (ft) 𝑃𝑃1 = Residual pressure at Hydrant #1 (downstream hydrant) in psi 𝑃𝑃2 = Residual Pressure at Hydrant #2 (upstream hydrant) in psi 𝑍𝑍1 = Gage elevation at Hydrant #1 (downstream hydrant) in feet 𝑍𝑍2 = Gage elevation at Hydrant #2 (upstream hydrant) in feet 𝛾𝛾 = 62.4 lb

ft3� The static pressures recorded at each hydrant prior to flowing the hydrant were also used as a check for the validity of the data. The static pressure head between the two residual hydrants was calculated and compared to the elevation head between the hydrants. The static pressure head should be equal, or relatively close, to the elevation head. The flow rate in the particular pipe was also calculated using the discharge pressure and geometry of the hydrant.

𝑸𝑸 = 𝟐𝟐𝟐𝟐.𝟖𝟖𝟐𝟐 ∗ 𝑪𝑪𝑫𝑫 ∗ 𝑫𝑫𝑴𝑴𝟐𝟐 ∗ �𝑷𝑷𝑫𝑫 (7)

Where, 𝑄𝑄 = Flowrate (gpm) 𝐶𝐶𝑑𝑑 = Coefficient of discharge of hydrant (see Figure 24) 𝐷𝐷𝑜𝑜 = Diameter of hydrant/reducer opening (in) 𝑃𝑃𝑑𝑑 = Discharge or pitot pressure (psi)


69

Figure 24 Hydrant Nozzle Discharge Coefficients

The Hazen Williams Equation was used to calculate the C-factor for each C-factor test performed.

𝑪𝑪 = 𝟑𝟑.𝟖𝟖𝟔𝟔𝟔𝟔 𝑳𝑳𝟔𝟔.𝟖𝟖𝟐𝟐∗𝑸𝑸𝑴𝑴𝑳𝑳

𝟔𝟔.𝟖𝟖𝟐𝟐∗𝑫𝑫𝟐𝟐.𝟔𝟔𝟐𝟐𝟕𝟕𝟕𝟕 (8)

Where, ℎ𝐿𝐿 = Head loss (ft) 𝐿𝐿 = Length of pipe (ft) 𝑄𝑄 = Flowrate (gpm) 𝐶𝐶 = C-Factor 𝐷𝐷 = Diameter of pipe (in)

D.1.3 C-Factor Sensitivity Analysis

In order to perform an accurate sensitivity analysis, every variable used in the C-Factor calculation is taken into account. These variables include length of the pipe, coefficient of discharge of the hydrant, diameter of the hydrant opening, discharge pressure, diameter of the pipe, pressure at both residual hydrants, and elevation at both residual hydrants. The equation used to calculate the C-factor including every variable is shown below.


70

𝑪𝑪 = 𝟑𝟑.𝟖𝟖𝟔𝟔𝟔𝟔 𝑳𝑳𝟔𝟔.𝟖𝟖𝟐𝟐×𝟐𝟐𝟐𝟐.𝟖𝟖𝟐𝟐×𝑪𝑪𝑫𝑫×𝑫𝑫𝑴𝑴𝟐𝟐×�𝑷𝑷𝑫𝑫

�∆𝑷𝑷𝜸𝜸 +∆𝒁𝒁�𝟔𝟔.𝟖𝟖𝟐𝟐

×𝑫𝑫𝟐𝟐.𝟔𝟔𝟐𝟐𝟕𝟕𝟕𝟕 (9)

Where, 𝐿𝐿 = Length of pipe (ft) 𝐶𝐶𝑑𝑑 = Coefficient of discharge of hydrant 𝐷𝐷𝑜𝑜 = Diameter of hydrant opening (in) 𝑃𝑃𝐷𝐷 = Discharge pressure (psi) 𝐷𝐷 = Diameter of pipe (in) ∆𝑃𝑃 = Change in pressure (psi) ∆𝑍𝑍 = Charge in Elevation (ft) The partial derivatives with respect to each variable in the C-Factor equation were calculated and normalized. These equations were combined with the uncertainty due to the precision in measurement of each variable. The following equation illustrates a quantitative uncertainty in the C-Factor (labeled as ∆C) due to the uncertainty in each variable.

∆𝑪𝑪𝑪𝑪

= �𝟔𝟔.𝟖𝟖𝟐𝟐 ∆𝑳𝑳𝑳𝑳� + �∆𝑪𝑪𝑫𝑫

𝑪𝑪𝑫𝑫� + �𝟐𝟐 ∆𝑫𝑫𝑴𝑴

𝑫𝑫𝑴𝑴� + �𝟔𝟔.𝟖𝟖 ∆𝑷𝑷𝑫𝑫

𝑷𝑷𝑫𝑫� + �−𝟐𝟐.𝟔𝟔𝟐𝟐𝟕𝟕 ∆𝑫𝑫

𝑫𝑫� + �� −𝟏𝟏

𝟏𝟏.𝟖𝟖𝟖𝟖𝟐𝟐𝜸𝜸𝑴𝑴𝑳𝑳� ∆(∆𝑷𝑷)

∆𝑷𝑷� +

�� −𝟏𝟏𝟏𝟏.𝟖𝟖𝟖𝟖𝟐𝟐𝑴𝑴𝑳𝑳

� ∆(∆𝒁𝒁)∆𝒁𝒁

� (10)

An example uncertainty analysis is shown below for the data measured at Site C-3.

∆𝐶𝐶

105= �(0.54) �

3 𝑜𝑜𝑑𝑑809 𝑜𝑜𝑑𝑑

�� + �0.10.9

� + �2 �0.1 𝑖𝑖𝑚𝑚2.5 𝑖𝑖𝑚𝑚

�� + �0.5 �1 𝑝𝑝𝑝𝑝𝑖𝑖

76 𝑝𝑝𝑝𝑝𝑖𝑖�� + �−2.627 �

0.03 𝑜𝑜𝑑𝑑0.833 𝑜𝑜𝑑𝑑

��

+ ��−1

1.854𝛾𝛾(14.94)� �(2 × 1) 𝑝𝑝𝑝𝑝𝑖𝑖

6 𝑝𝑝𝑝𝑝𝑖𝑖�� + ��

−11.854(14.94)

� �(2 × 0.1)𝑜𝑜𝑑𝑑−28.8 𝑜𝑜𝑑𝑑

��

= 0.295

∆𝐶𝐶 = 31.0

𝐶𝐶 = 105 ± 31.0 The uncertainty for each C-factor test was calculated. The data needed for calculations in the sensitivity analysis is shown below in Table 26, and the results of the C-factor sensitivity analysis are shown in


71

Table 26 C-Factor Uncertainty Data

Site Length (ft)

Diameter (in)

Discharge Pressure (psi) ΔP (psi) ΔZ (ft) Head

Loss (ft) C-Factor

C-1 424 8 64.5 1 3.5 5.81 149 C-1 424 8 45 4 3.5 15.05 224 C-2 769 6 43.2 21 8.8 55.00 143 C-3 809 10 76 6 -28.8 14.94 105 C-4 682 6 47 12 -5.6 26.74 233 C-4 682 6 47.2 14 -5.6 29.05 211 C-5 2236 12 82 10.5 -11.39 12.86 127 C-5 2236 12 62 17 -11.39 27.88 144 C-6 372 8 50.8 3.4 1.4 9.25 132 C-7 317 8 36 10 19.6 3.50 130 C-8 443 8 54 -7 6.5 5.05 145 C-9 492 6 45 14.25 -8.338 20.23 180 C-10 763 6 26.5 43 20.3 119.63 75 C-10 763 6 28 42 20.3 117.32 78


72

Table 27 C-Factor Sensitivity Results

D.2 Fire Flow Test

D.2.1 Fire Flow Test Procedure

The AWWA M17 guide- Installation, Field Testing, and Maintenance of Fire Hydrants was used to develop the standard operating procedures for the fire flow test. Hydrant Testing Crew Instructions 1. Test shall be made during a period of ordinary demand. Before testing begins the

Nicholasville WTP plant will need to be notified of the time of testing. This is so the Nicholasville WTP can record the required data regarding tank levels, pump operation schedules, plant flow, etc. during each hydrant flow test.

2. One hydrant designated the “Residual Hydrant”, will be chosen to collect the normal static pressure while the other hydrants in the group, the “Flow Hydrant’, is closed. The residual

C-1 C-1 C-2 C-3 C-4 C-4 C-5 C-5 C-6 C-7 C-8 C-9 C-10 C-10

L (ft) 424 424 769 809 682 682 2236 2236 372 317 443 492 763 763ΔL 3 3 3 3 3 3 3 3 3 3 3 3 3 3

Uncertainty L 0.0038 0.0038 0.0021 0.0020 0.0024 0.0024 0.0007 0.0007 0.0044 0.0051 0.0037 0.0033 0.0021 0.0021Cd 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9

ΔCd 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1Uncertainty Cd 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111 0.1111

Do (in) 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5ΔDo 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

Uncertainty Do 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08Pd (psi) 64.5 45 43.2 76 47 47.2 82 62 50.8 36 54 45 26.5 28

ΔPd 1 1 1 1 1 1 1 1 1 1 1 1 1 1Uncertainty Pd 0.0078 0.0111 0.0116 0.0066 0.0106 0.0106 0.0061 0.0081 0.0098 0.0139 0.0093 0.0111 0.0189 0.0179

D (ft) 0.667 0.667 0.500 0.833 0.500 0.500 1.000 1.000 0.667 0.667 0.667 0.500 0.500 0.500ΔD 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03

Uncertainty D 0.1182 0.1182 0.1576 0.0946 0.1576 0.1576 0.0788 0.0788 0.1182 0.1182 0.1182 0.1576 0.1576 0.1576Head Loss (ft) 5.81 15.05 55.00 14.94 26.74 29.05 12.86 27.88 9.25 3.50 5.05 20.23 119.63 117.32

ΔP (psi) 1 4 21 6 12 14 10.5 17 3.4 10 -7 14.25 43 42Δ(ΔP) 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Uncertainty ΔP 2.1E-05 2.0E-06 1.0E-07 1.3E-06 3.7E-07 3.0E-07 8.9E-07 2.5E-07 3.8E-06 3.4E-06 3.4E-06 4.2E-07 2.3E-08 2.4E-08ΔZ (ft) 3.5 3.5 8.8 -28.8 -5.6 -5.6 -11.39 -11.39 1.4 19.6 6.5 -8.338 20.3 20.3Δ(ΔZ) 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2

Uncertainty ΔZ 0.0053 0.0020 0.0002 0.0003 0.0007 0.0007 0.0007 0.0003 0.0083 0.0016 0.0033 0.0006 0.0000 0.0000C-factor 149 224 143 105 233 211 127 144 132 130 145 180 75 78

ΔC/C 0.3262 0.3263 0.3626 0.2945 0.3625 0.3624 0.2775 0.2791 0.3319 0.3299 0.3255 0.3638 0.3698 0.3688Total Uncertainty 48.6 73.1 51.9 31.0 84.5 76.5 35.2 40.2 43.7 42.9 47.2 65.5 27.7 28.7

C-factor with Uncertainty

149± 48.6

224± 73.1

143± 51.9

105± 31.0

233± 84.5

211± 76.5

127± 35.2

144± 40.2

132± 43.7

130± 42.9

145± 47.2

180± 65.5

75± 65.5

78± 28.7


73

pressure will also be collected while the other hydrant in the group is flowing. If the hydrants are not at the same elevation, height of the hydrants will need to be recorded.

3. One hydrant, designated the “Flow Hydrant”, is chosen to be the hydrant where flow pressure will be observed, using a Pitot tube (Hydrant Flow Meter). The Pitot tube to be used for this project is a Pollard P669LF.

4. At this time the flowing hydrant shall be opened, water should be allowed to flow long enough to clear any debris and foreign substances from stream.

5. A 2 ½” cap with pressure gauge that can read approximately 25 psi greater than the system pressure for the hydrant will be attached to the residual hydrant and the hydrant opened full. For this project a Pollard item # P67022LF Hydrant Static Pressure gage will be used. A reading (static pressure) is taken when the needle comes to a rest. Record this reading on the Fire Flow Data Collection Log.

6. The hydrant testing crew members for the residual hydrant will then signal the flowing hydrant crew member using 2 way radio device or cell phone. Attach the Pitot tube to the 2 ½” outlet along with the static pressure gage to a remaining outlet and open hydrant again. The hydrant valve should be opened slowly to prevent pressure surges or water hammer in the system. The hydrant should be flowed approximately 2-5 minutes.

a. If dechlorination regulations exist for the selected hydrant then dechlorinating Diffuser will need to be connected to the flowing hydrant.

7. Observe the Pitot Gauge and static gage reading and record the pressures at the residual hydrant and the flowing hydrants simultaneously (once readings have stabilized). Proper communication will be needed to achieve simultaneous recording.

8. After adequate readings have been recorded, close the flow hydrant to cease flow. Static pressure readings should be recorded at the residual hydrant.

9. Complete the other necessary information on the Fire Flow data Collection Log.

Water Treatment Plant Crew Instruction

1. During Field testing the WTP will collect SCADA data showing the tank levels and pump flows in and out of the system.

2. The WTP will also alert the Field crew in the event that testing needs to be halted due to excessive pumping.

D.2.2 Fire Flow Data

Table 28 displays a summary of all data used during fire flow calculations including pressures recorded at hydrants during testing, pressures found using model simulations, the status of pumps and storage tanks during testing, etc.


74

Table 28 Data Collected from Fire Flow Tests

**Represents Correct Analysis once Demand had been adjusted for the particular event

D.2.3 Fire Flow Calculations

The procedure below shows calculations for the maximum capacity of a hydrant if it is pumped down to a 20 psi residual pressure. The flow rate formula produces a value in gallons per minute (GPM) based on the nozzle diameter and pitot pressure.

𝑸𝑸 = 𝟐𝟐𝟐𝟐.𝟖𝟖𝟐𝟐 ∗ 𝑪𝑪𝑫𝑫 ∗ 𝑫𝑫𝑴𝑴

𝟐𝟐 ∗ �𝑷𝑷𝑫𝑫 (11) Where, Q = flowrate (gpm) Cd = coefficient of discharge Do = diameter of hydrant opening (in) Pd = discharge/pitot pressure (psi)

Site ID F1 F2 F3 F4 F4** F5 F6 F7 F8 F9 F10 Date Performed 17-Oct 9-Nov 17-Oct 9-Nov 9-Nov 9-Nov 17-Oct 17-Oct 17-Oct 9-Nov 9-NovApproximate Time 19:00 19:45 18:10 19:05 19:05 17:45 18:35 19:25 17:50 18:40 18:05Hydrant/Node J-H362* J-H667 J-H218 J-H743 J-H743 J-H287 J-H961 J-H535 J-H638 J-H569 J-H1144Residual Pressure (psi) 85 82 64 79 79 44 64 91 62 78 77Field Static Pressures (psi) 99 88 92 94 94 56 72 101 69 83 84Model Static Pressure (psi) 98.7 89.9 94.3 94 94 57 73.5 99.5 69.8 83.9 82.7Model Residual Pressure (psi) 80.5 78.8 59.5 70.1 76 45.4 64.3 88.9 61 78.2 76.8Static Difference (psi) 0.3 -1.9 -2.3 0 0 -1 -1.5 1.5 -0.8 -0.9 1.3Residual Difference (psi) 4.5 3.2 4.5 8.9 3 -1.4 -0.3 2.1 1 -0.2 0.2Hydrant/Node J-H363 J-H666 J-H219 J-H744 J-H744** J-H286 J-H960 J-H534 J-H635 J-H570 J-H1145Flow (gpm) 1507 1488 2428 1425 1425 950 2279 2760 1300 1354 2480Field Static Pressure (psi) 102 92 94 92 92 46 73 98 75 73 82Model Static Pressure (psi) 98.4 88.7 90.9 89.6 89.6 46.3 71.1 99.8 73.11 71.3 81.1Field Flow Pressure (psi) 80.5 78.5 52 72 72 32 45.5 68 60 65 55Static Difference (psi) 3.6 3.3 3.1 2.4 2.4 -0.3 1.9 -1.8 1.89 1.7 0.9Demand Factor 1.64 1.2 1.29 1.248 1.248 1.265 1.396 1.225 1.29 1.317 1.265 Pump #3 Flow (gpm) 3315 3315 3315 3315 3315 3315Pump #4 Flow (gpm) 3264 3264 3264 3264 3264Capital Ct Actual Level (ft) 1142.3 1142.08 1142.29 1133 1133.04 1132.92 1142.34 1142.23 1142.44 1132.94 1132.97Lake St Level (ft) 1140.57 1140.58 1140.44 1130.1 1130.07 1130.04 1140.56 1140.61 1140.41 1129.99 1130.04Stephens Dr Level (ft) 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5 1109.5PUMP 1 OFF OFF OFF OFF OFF OFF OFF OFF OFF OFF OFFPUMP 2 OFF OFF OFF OFF OFF OFF OFF OFF OFF OFF OFFPUMP 3 OFF On OFF On On On OFF OFF OFF On OnPUMP 4 On OFF On OFF OFF OFF On On On OFF OFFPUMP 5 OFF OFF OFF OFF OFF OFF OFF OFF OFF OFF OFF


75

This formula below calculates available flow based on the readings taken before and during the single outlet flow test (solving for "QR".)

𝑸𝑸𝑹𝑹 = 𝑸𝑸𝑭𝑭 ∗𝑴𝑴𝑨𝑨

𝟔𝟔.𝟖𝟖𝟐𝟐

𝑴𝑴𝒇𝒇𝟔𝟔.𝟖𝟖𝟐𝟐 (12)

Where, QF = Observed flow (gpm) hr = Pressure drop from the static pressure to the desired residual pressure (psi) hf = Pressure drop from the static pressure to the actual residual pressure recorded (psi)

D.3 Boundary Conditions Collection

The following figures display the output from the SCADA system in Nicholasville. One set of graphs display the tank levels of all three elevated storage tanks, while another set show the pump flow rates. The SCADA data was saved for days when hydraulic testing occurred.

Figure 25 SCADA Data (Tank Levels) 10/9/2011


76

Figure 26 SCADA Data (Pump Flow) 10/9/2011



77

Figure 28 SCADA Data (Pump Flow) 10/11/2011



78

Figure 30 SCADA Data (Pump Flows) 10/12/2011


79



80

Figure 32 SCADA Data 10/14/2011



81

Appendix E: Model Validation The following tables show the water levels in all three storage tanks measured by the SCADA data compared to the levels found by the model simulation. These values reflect the graphs shown in section 5.4.1 24 hour-EPS Simulation on page 52.

Table 29 EPS vs. SCADA Data (Tank Levels) 10/10/2011

Model DATA SCADA DATA Difference (Model-SCADA)

Time (hr)

Elevation of Lake St

Tank

Elevation of Capital Ct Tank

Elevation of

Stephens Dr Tank

Elevation of

Lake St Tank

Elevation of

Capital Ct Tank

Elevation of

Stephens Dr Tank

Lake St

Capital Ct

Stephens Dr

0 32.16 28.1 26.18 32 28 26 0.16 0.1 0.18 1 33.94 29.24 27.89 34 29.7 28 -0.06 -0.46 -0.11 2 36.17 30.93 30.1 36 31.5 30 0.17 -0.57 0.1 3 38.58 32.59 32.43 38 33 33 0.58 -0.41 -0.57 4 39.33 32.47 33.35 39.5 33.5 34 -0.17 -1.03 -0.65 5 37.76 30.43 31.84 39.5 32 33 -1.74 -1.57 -1.16 6 36.91 29.23 31.02 39 31 32 -2.09 -1.77 -0.98 7 37.02 29.44 30.84 38.2 30 31.4 -1.18 -0.56 -0.56 8 37.65 30.19 31.44 38 31.2 31 -0.35 -1.01 0.44 9 35.66 27.99 29.46 36.7 29.5 29.5 -1.04 -1.51 -0.04

10 33.91 25.92 27.62 35 28 27.5 -1.09 -2.08 0.12 11 32.47 24.37 26.21 33 26 26 -0.53 -1.63 0.21 12 30.73 22.44 24.09 31 24.6 24 -0.27 -2.16 0.09 13 28.83 20.37 22.03 29 23.3 22 -0.17 -2.93 0.03 14 27.09 18.61 20.09 27.6 22 20 -0.51 -3.39 0.09 15 27.08 19.36 19.86 27.6 23 20.2 -0.52 -3.64 -0.34 16 27.18 19.86 20.07 27.6 23.5 21 -0.42 -3.64 -0.93 17 27.47 20.41 20.61 28 24 21.5 -0.53 -3.59 -0.89 18 27.88 20.97 21.16 28.1 24.3 22 -0.22 -3.33 -0.84 19 28.2 21.37 21.4 28.3 24.8 22.1 -0.1 -3.43 -0.7 20 28.32 21.5 21.28 29 25 22.3 -0.68 -3.5 -1.02 21 28.84 22.26 21.82 29.5 25.6 23 -0.66 -3.34 -1.18 22 29.99 23.8 23.26 30.8 26.8 24 -0.81 -3 -0.74 23 31.23 25.29 24.73 32 28 26 -0.77 -2.71 -1.27 24 32.97 27.23 26.85 34 29.5 28 -1.03 -2.27 -1.15


82


Model DATA SCADA DATA Difference (Model - SCADA)

Time (hr)

Elevation of Lake St Tank


Elevation of

Stephens Dr Tank

Elevation of

Lake St Tank

Elevation of

Capital Ct Tank

Elevation of

Stephens Dr Tank

Lake St

Capital Ct

Stephens Dr

0 34 30 28 34 30 28 0 0 0 1 36.04 31.26 30 36 31 30 0.04 0.26 0 2 38.21 32.74 31.95 37.5 33 32 0.71 -0.26 -0.05 3 39.94 34.36 34.34 39.5 34 34 0.44 0.36 0.34 4 38.56 31.53 32.74 39 32.5 33 -0.44 -0.97 -0.26 5 37.7 30.09 31.83 38 31 32 -0.3 -0.91 -0.17 6 36.6 28.86 30.69 37 29.7 30 -0.4 -0.84 0.69 7 35.29 27.42 29.2 35.5 28 28 -0.21 -0.58 1.2 8 33.67 25.65 27.33 33 26.5 26 0.67 -0.85 1.33 9 31.96 23.71 25.34 31 25 24 0.96 -1.29 1.34

10 30.25 21.81 23.45 29 23.4 22 1.25 -1.59 1.45 11 28.39 19.82 21.43 27 21.5 19.5 1.39 -1.68 1.93 12 26.3 17.56 19.1 25 20 17.5 1.3 -2.44 1.6 13 24.35 15.65 17.27 23 18.5 15.5 1.35 -2.85 1.77 14 22.44 13.78 15.41 21 17 13.5 1.44 -3.22 1.91 15 21.94 13.97 14.78 20 16 12 1.94 -2.03 2.78 16 22.81 15.75 15.72 20.5 18 13 2.31 -2.25 2.72 17 23.85 17.4 17.02 21 19 14 2.85 -1.6 3.02 18 24.4 18.01 17.47 21.7 19.5 14.6 2.7 -1.49 2.87 19 24.61 18.07 17.68 21.9 19.9 15 2.71 -1.83 2.68 20 24.98 18.38 18.23 21.9 19.9 15 3.08 -1.52 3.23 21 25.15 18.48 18.19 22 19.9 15.2 3.15 -1.42 2.99 22 25.48 18.8 18.7 22.2 20 15.9 3.28 -1.2 2.8 23 25.83 19.15 19.1 23 20.5 16.3 2.83 -1.35 2.8 24 26.22 19.54 19.56 23.4 21 17.2 2.82 -1.46 2.36


83


Model DATA SCADA DATA Difference ( Model - SCADA)

Time (hr)



Elevation of

Stephens Dr Tank



Elevation of

Stephens Dr Tank

Lake Street

Capital Ct

Stephens Dr

0 32.84 28.47 27.19 32.6 28.5 27 0.24 -0.03 0.19 1 34.02 28.57 28.05 33.7 29 28 0.32 -0.43 0.05 2 35.15 29.09 29.1 34.5 30 29 0.65 -0.91 0.1 3 36.29 29.87 30.14 35.5 30.8 30 0.79 -0.93 0.14 4 37.36 30.65 31.08 36.5 31.5 31 0.86 -0.85 0.08 5 38.36 31.43 31.93 37.5 32 32 0.86 -0.57 -0.07 6 39.23 32.14 32.63 38 32.8 32.8 1.23 -0.66 -0.17 7 39.93 32.82 33.13 38.5 33 33 1.43 -0.18 0.13 8 40 33.14 33.14 39.5 33.2 33 0.5 -0.06 0.14 9 40 33.34 33.19 39.5 33.2 33 0.5 0.14 0.19

10 40 33.36 33.06 39.5 33.2 33 0.5 0.16 0.06 11 38.03 31.09 31.66 37.3 31.5 31 0.73 -0.41 0.66 12 35.78 28.27 29.51 35.5 29.5 29 0.28 -1.23 0.51 13 34.04 26.11 27.71 33 27.5 26.8 1.04 -1.39 0.91 14 32.14 23.9 25.57 31 25.8 25 1.14 -1.9 0.57 15 30.42 21.98 23.72 29.5 24 23 0.92 -2.02 0.72 16 28.67 20.18 22.01 28 23 21.5 0.67 -2.82 0.51 17 27.98 20.02 21.31 26.8 22 20 1.18 -1.98 1.31 18 28.4 21.05 21.65 26.7 22.6 20 1.7 -1.55 1.65 19 28.64 21.58 21.77 26.8 23.4 20 1.84 -1.82 1.77 20 28.98 22.08 22.19 27 23.8 20.2 1.98 -1.72 1.99 21 29.3 22.49 22.52 27.3 24 20.9 2 -1.51 1.62 22 29.68 22.92 22.96 28 24.1 21.4 1.68 -1.18 1.56 23 30.13 23.43 23.57 28.7 25 22.4 1.43 -1.57 1.17 24 30.94 24.34 24.72 29.5 25.7 23.6 1.44 -1.36 1.12


84

Appendix F: Demand Factor Development When the hydraulic model is used for future simulations, demand factors are needed to estimate demand patterns during the desired time of simulation. These demand factors will adjust water usage throughout the system based on time and location.

These demand factors were found by collecting data over a three day period (October 10, 11, and 13, 2011). Because this data was collected on a Monday, Tuesday, and Thursday, the demand factors reflect a typical weekday. The demand factors were calculated throughout the 24 hour period separately for each of the three days. Then the demand factors at each hour interval were averaged to find an overall demand factor. Graphs are included showing the demand factors over a 24 hour period for each day. The tables below show the data used to calculate the Diurnal factors at hour intervals for each of the three days. There is also a table that shows the demand pattern during the C-factor Testing. (10/12/2011).

Figure 34 Demand Factors (10/10/2011)

0.4

0.6

0.8

1.0

1.2

1.4

1.6

0:00 6:00 12:00 18:00 0:00

Diur

nal D

eman

d Fa

ctor

Time (hr)

Demand Factor (10/10/2011)


85

Table 32 Demand Factor Data (10/10/2011)

Time (hr)

Amount of Storage in

Lake St Tank (gal)

Amount of Storage in Capital Ct Tank (gal)

Amount of Storage in Stephens Dr (gal)

High Service Pump 3

Flow

Gallons From High

Service Pumps

Amount In (gal)

Total Storage in

Tanks (gal)

Demand of System

(gal)

Diurnal Curve Factor

0:00 623818 1012119 566268 4400 264000 22022051:00 661557 1085743 615235 4400 264000 264000 2362535 103670 0.7362:00 695992 1163699 660744 4400 264000 264000 2520435 106100 0.7543:00 726041 1228663 720205 3250 195000 264000 2674909 109526 0.7784:00 741950 1250317 735850 0 0 195000 2728117 141792 1.0075:00 741950 1185354 720205 0 0 0 2647509 80608 0.5736:00 733900 1142045 701818 733 43980 0 2577763 69746 0.4957:00 727612 1098736 689879 4112 246720 43980 2516227 105516 0.7498:00 726041 1150706 681920 0 0 246720 2558667 204280 1.4519:00 701932 1077082 649594 0 0 0 2428608 130059 0.924

10:00 679241 1012119 615235 0 0 0 2306595 122013 0.86711:00 643046 925501 566268 0 0 0 2134815 171780 1.22012:00 604027 838884 514772 0 0 0 1957683 177132 1.25813:00 563129 752266 461641 0 0 0 1777036 180647 1.28314:00 533800 665648 407569 3280 196800 0 1607017 170019 1.20815:00 533800 687302 412990 3280 196800 196800 1634092 169725 1.20616:00 533800 708957 434676 3280 196800 196800 1677433 153459 1.09017:00 542246 734942 448158 3280 196800 196800 1725346.4 148886.6 1.05818:00 544334 752266 461641 3280 196800 196800 1758241 163905.4 1.16419:00 548510 752266 461641 3280 196800 196800 1762417 192624 1.36820:00 563129 752266 461641 3585 215100 196800 1777036 182181 1.29421:00 573129 795575 475008 4400 264000 215100 1843712 148424 1.05422:00 599847 838884 514772 4400 264000 264000 1953503 154209 1.09523:00 623818 925501 566268 4400 264000 264000 2115587 101916 0.7240:00 661557 1012119 615235 4400 264000 264000 2288911 90676 0.644

Total Daily Demand (gal) 3378894Average Hourly Demand 140787.25


86

Figure 35 Demand Factors (10/11/2011)

Table 33 Demand Factor Data (10/11/2011)

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

0:00 6:00 12:00 18:00 0:00

Diur

nal D

eman

d Fa

ctor

Time (hr)

Demand Factors (10/11/2011)

Time (hr)


Lake St Tank (gal)



High Service Pump 3

Flow

Gallons From High

Service Pumps

Amount In (gal)

Total Storage in

Tanks (gal)

Demand of System

(gal)


For Model

0:00 661557 1098736 615235 4400 264000 23755281:00 695992 1142045 660744 4400 264000 264000 2498781 140747 0.904 0.9212:00 718849 1228663 701818 4400 264000 264000 2649330 113451 0.729 0.7423:00 741950 1271971 735850 0 0 264000 2749771 163559 1.050 1.0704:00 733900 1207008 720205 0 0 0 2661113 88658 0.569 0.5805:00 726041 1142045 701818 0 0 0 2569904 91209 0.586 0.5976:00 711658 1085743 660744 0 0 0 2458145 111759 0.718 0.7317:00 687616 1012119 615235 0 0 0 2314970 143175 0.919 0.9378:00 643046 947155 566268 0 0 0 2156469 158501 1.018 1.0379:00 604027 882192 514772 0 0 0 2000991 155478 0.998 1.01710:00 563129 812898 461641 0 0 0 1837668 163323 1.049 1.06911:00 521133 730611 393989 0 0 0 1645733 191935 1.232 1.25612:00 479615 665648 339819 0 0 0 1485082 160651 1.032 1.05113:00 438846 600685 286404 0 0 0 1325935 159147 1.022 1.04114:00 398077 535722 234242 1175 70500 0 1168041 157894 1.014 1.03315:00 377692 492413 196815 4700 282000 70500 1066920 171621 1.102 1.12316:00 387884 579031 221242 4600 276000 282000 1188157 160763 1.032 1.05217:00 398077 622340 247242 4080 244800 276000 1267659 196498 1.262 1.28618:00 412346 643994 262820 3290 197400 244800 1319160 193299 1.241 1.26519:00 416423 661317 273206 3290 197400 197400 1350946 165614 1.063 1.08420:00 416423 661317 273206 3290 197400 197400 1350946 197400 1.268 1.29221:00 418462 661317 298485 3290 197400 197400 1378264 170082 1.092 1.11322:00 422538 665647 316962 3290 197400 197400 1405147 170517 1.095 1.11623:00 438846 687302 307622 3290 197400 197400 1433770 168777 1.084 1.1040:00 447000 708957 331730 3290 197400 197400 1487687 143483 0.921 0.939

3737541155730.88Average Hourly Demand

Total Daily Demand (gal)


87

Figure 36 Demand Factors 10/12/2011

Table 34 Demand Factor Data 10/12/2011

0.7

0.8

0.8

0.9

0.9

1.0

1.0

1.1

1.1

1.2

1.2

9:30 11:00 12:30 14:00 15:30

Diru

nal D

eman

d Fa

ctor

Time (hr)

Demand Factor 10/12/2011

Time (hr)


Lake St Tank (gal)



Stephens Dr (gal)

High Service Pump 3

Flow

Gallons From High Service

Pumps

Amount In (gal)

Total Storage in

Tanks (gal)

Demand of System

(gal)


9:30 577129 922902 496558 3255 97650 199658910:00 578129 919870 495238 3255 97650 97650 1993237 101002 1.11710:30 578129 922036 492335 3255 97650 97650 1992500 98387 1.08811:00 578729 922036 497614 3255 97650 97650 1998379 91771 1.01511:30 579129 923335 490223 3255 97650 97650 1992687 103342 1.14312:00 579929 918571 497350 3255 97650 97650 1995850 94487 1.04512:30 579929 918571 497350 3255 97650 97650 1995850 97650 1.08013:00 583129 933729 506853 3255 97650 97650 2023711 69789 0.77213:30 584382 937194 510812 3255 97650 97650 2032388 88973 0.98414:00 584382 937194 510812 3255 97650 97650 2032388 97650 1.08014:30 587308 949320 519964 3255 97650 97650 2056592 73446 0.81315:00 590443 957982 523599 3255 97650 97650 2072024 82218 0.91015:30 592951 960148 526715 3255 97650 97650 2079814 89860 0.99416:00 594831 964479 531648 3255 97650 97650 2090958 86506 0.957

117508190390.846

77280Average Half Hourly DemandModel Average Half Hourly

Total Daily Demand (gal)


88

Figure 37 Demand Factors 10/13/2011

Table 35 Demand Factor Data 10/13/2011

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

0:00 6:00 12:00 18:00 0:00

Diru

nal D

eman

d Fa

ctor

Time (hr)

Demand Factor 10/13/2011


89

Time (hr)


Lake St Tank (gal)



High Service Pump 3

Flow

Gallons From High

Service Pumps

Amount In (gal)

Total Storage in

Tanks (gal)

Demand of System

(gal)


0:00 635354 1033773 591163 3280 196800 22602901:00 656003 1055428 615235 3280 196800 196800 2326666 130424 0.8492:00 670399 1098736 638445 3280 196800 196800 2407580 115886 0.7543:00 687616 1133383 660744 3280 196800 196800 2481743 122637 0.7984:00 703825 1163699 681920 3280 196800 196800 2549444 129099 0.8405:00 718849 1185354 701818 3280 196800 196800 2606021 140223 0.9126:00 726041 1220001 716527 3280 196800 196800 2662569 140252 0.9127:00 729970 1228663 720205 3255 195300 196800 2678838 180531 1.1758:00 733900 1237324 720205 3250 195000 195300 2691429 182709 1.1899:00 733900 1237324 720205 3250 195000 195000 2691429 195000 1.269

10:00 733900 1237324 720205 1300 78000 195000 2691429 195000 1.26911:00 715972 1163699 681920 0 78000 2561591 207838 1.35212:00 687616 1077082 638445 0 0 0 2403143 158448 1.03113:00 643046 990464 586186 0 0 0 2219696 183447 1.19414:00 604027 916839 540736 0 0 0 2061602 158094 1.02915:00 573129 838884 488376 0 0 0 1900389 161213 1.04916:00 542246 795575 448158 328 19680 0 1785979 114410 0.74417:00 516906 752266 407569 3233 193980 19680 1676741 128918 0.83918:00 514793 778251 407569 3233 193980 193980 1700613 170108 1.10719:00 516906 812898 407569 3233 193980 193980 1737373 157220 1.02320:00 521133 830222 412990 3233 193980 193980 1764345 167008 1.08721:00 527466 838884 431965 3233 193980 193980 1798315 160010 1.04122:00 542246 843214 445462 3233 193980 193980 1830922 161373 1.05023:00 556864 882192 472335 3233 193980 193980 1911391 113511 0.7390:00 573129 912508 504213 3260 195600 193980 1989850 115521 0.752

3688880153703.33

Total Daily Demand (gal)Average Hourly Demand


90

Appendix G: Data Collection Logs


91


92


93


94


95

Water Distribution System Calibration Report Security Deliverables/HSD... · Utility Operational...

Documents

Transcript of Water Distribution System Calibration Report Security Deliverables/HSD... · Utility Operational...