Analyzing Interval Data Using the Energy Charting and Metrics (ECAM) Tool

1PNWD-SA-9511

Adapted from a presentation by:

Srinivas Katipamula, Ph.D.Staff Scientist, Battelle Pacific Northwest Division

Bill Koran, P.E.Senior Engineer, Quantum Energy Services and Technologies

Presentation Outline

• ECAM Capabilities

• Analysis of Interval Meter Data with ECAM

2PNWD-SA-9511

• A tool to assist with analysis of building performance data

• Facilitate the types of analyses that energy engineers, commissioning providers, and building operators might need to make

• “Semi-automation,” not complete automation of the analyses

What is ECAM?

3PNWD-SA-9511

• Because analyses is not completely automated, the tool is flexible, supporting additional analysis without limiting any normal spreadsheet functions

ECAM Version 2.0 New Features

• Excel 2007/2010 compatibility

• Full time history chart, filterable to selected months or days

• Load profile as box plots

• Load duration chart (histograms/frequency distributions)

• Chart to check input schedule

• Matrix charts

4PNWD-SA-9511

• Matrix charts

• Metrics by daytype, occupancy, and month

• Data summaries

• Additional scheduling periods for startup and shutdown

• Integrated user guide/help

• PNNL re-tuning charts (See http://buildingefficiency.labworks.org/large.stm for more details)

ECAM Detection/Diagnostic Features

• Detection

– Tool does not provide detection directly

– Assists with performance tracking

– Provides the ability to investigate the relationships between variables

5PNWD-SA-9511

between variables

• Diagnostics

– Tool does not provide diagnostics

– Future versions may provide some limited diagnostic capabilities, but this is not the focus of the tool

– General intent is to maximize the user’s ability to benefit from whatever data is available

Five Easy Steps

1. Select data from existing spreadsheet

2. Map points

3. Create schedules

6PNWD-SA-9511

3. Create schedules (optional)

4. Input energy project dates (optional)

5. Create metrics and charts

User’s Data — Example

DateTime

Outside

Dry Bulb

325

COMMON

CHWS

326

COMMON

CHWR

328

CHILLER

1 AMPS

342 COM

CDW

SUPPLY

343 COM

CDW

RETURN

345 A

CHILLER

ON

414

SCHWP-5

KW

415

SCHWP-6

KW

416

SCHWP-7

KW

8/23/04 12:00 AM 67.4 41.9 48.5 226.1 69.2 77.3 100 0 5 0.8

8/23/04 12:15 AM 67.5 42 48.7 224.3 69 77.1 100 0 4.8 1.3

8/23/04 12:30 AM 67.3 41.6 47.6 206.9 68.9 76.3 100 0 4.7 1.6

8/23/04 12:45 AM 67.1 41.8 47.7 206.7 68.5 75.8 100 0 4.6 1.3

8/23/04 1:00 AM 67.1 42.2 47.6 195.9 68.2 75 100 0 4.7 1.4

8/23/04 1:15 AM 67 41.8 47 193.8 68.1 74.8 100 0 4.6 0

8/23/04 1:30 AM 67 42 46.2 168.1 67.6 72.9 100 0 4.7 0

7PNWD-SA-9511

8/23/04 1:45 AM 66.8 41.9 45.3 152.1 67.5 71.9 100 0 4.6 0.6

8/23/04 2:00 AM 67 42.2 45.3 150.2 67.8 71.9 100 0 4.6 0.6

8/23/04 2:15 AM 67.1 42 48.5 0 67.6 74.6 100 0 4.8 0.9

8/23/04 2:30 AM 67.1 41.6 47.7 0 68.8 75.7 100 0 4.5 0.5

8/23/04 2:45 AM 67 42 48.4 0 68.2 75.1 100 0 4.7 0.8

8/23/04 3:00 AM 67.1 41.6 47.5 0 68.4 75.1 100 0 4.5 0.9

8/23/04 3:15 AM 67.2 41.7 47.5 0 68.3 74.8 100 0 4.7 0.8

8/23/04 3:30 AM 67.4 42.4 47.9 0 68.2 74.4 100 0 4.6 0.8

8/23/04 3:45 AM 67.4 43.5 48.6 0 68.3 73.8 100 0 4.7 0.9

8/23/04 4:00 AM 67.5 45 49.4 0 68.1 73 100 0 4.8 0

Step 1: Select data

1a. 1b.

8PNWD-SA-9511

1c. 1d.

Step 2: Map points

9PNWD-SA-9511

User’s points are mapped to common point names as perUser’s points are mapped to common point names as per

A Specifications Guide for Performance Monitoring SystemsA Specifications Guide for Performance Monitoring Systems

Step 3: Create Schedule (optional)

10PNWD-SA-9511

Forms are similar to eQuestForms are similar to eQuest

Step 4: Input Energy Project Dates (optional)

11PNWD-SA-9511

Step 5: Create Metrics or Charts

12PNWD-SA-9511

Preprocessing of Data

• Time stamp disaggregation

• Daytyping

• Occupancy

13PNWD-SA-9511

• Pre/Post Energy Project

• Normalizations

• Calculations

ECAM Metrics Creation Capabilities

• Can be normalized:

– Building area (e.g. W/sqft)

– Cooling tons (e.g. kW/ton)

– cfm (e.g. Watts/cfm)

• Can be filtered by:

– Year

– Month

– Pre/post time periods

In ECAM, metrics are typically data In ECAM, metrics are typically data averaged over a particular time periodaveraged over a particular time period

14PNWD-SA-9511

– cfm (e.g. Watts/cfm)

– gpm (e.g. gpm/ton)– Pre/post time periods

– Daytype

– Time of day

– Occupancy

– Weather conditions

– Status of equipment

– CombinationsAll filters and many normalizations All filters and many normalizations are automatically set up by ECAMare automatically set up by ECAM

Automatic Creation Of Additional Fields For Metrics

• Equipment Status

• Demand (kW)

• Chilled water tons

• from demand (kW) or

current (amps)

• from current (amps)

approximate calculation

• whenever flows and

temperatures are available

15PNWD-SA-9511

• Watts per square foot

• CFM per square foot

• kW per ton

• gpm per ton

temperatures are available

• for all electrical demand

points (kW)

• for all air flow points

• for all related points

• for all related points

Interval Data Visualization

• Load profiles

– Raw data

– Averages

– Pre/post

– By daytype

– Calendars

• Scatter charts

– Raw data

– Averages

– Pre/post

– By daytype

– By occupancy

16

– Calendars – By occupancy

– By time of day

– Choice of independent variable

– Multiple points

Filtering capabilities are providedfor all types of charts

Load Profile by Daytype

2.5

3.0

3.5

4.0

4.5

Avg

Ele

cM

tr_W

att

s_p

erS

F

Weekday

Year (All)

Month (All)

MonthYr Sep 2006

Weekday (All)

17PNWD-SA-9511

0.0

0.5

1.0

1.5

2.0

12:0

0 A

M

2:0

0 A

M

4:0

0 A

M

6:0

0 A

M

8:0

0 A

M

10:0

0 A

M

12:0

0 P

M

2:0

0 P

M

4:0

0 P

M

6:0

0 P

M

8:0

0 P

M

10:0

0 P

M

Avg

Ele

cM

tr_W

att

s_p

erS

F

Saturday

Sunday

Holiday

Weekday (All)

Day (All)

Holiday (All)

5degBin (All)

1degBin (All)

TempRng (All)

Filtering Options

Calendar Load Profile

October

Monday Tuesday Wednesday Thursday Friday Saturday Sunday

1 2

3 4 5 6 7 8 9

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

6

8

6

8

6

8

6

8

6

8

6

8

6

8

18PNWD-SA-9511

10 11 12 13 14 15 16

17 18 19 20 21 22 23

24 25 26 27 28 29 30

31

0

2

4

0

2

4

0

2

4

0

2

4

0

2

4

0

2

4

0

2

4

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

0

2

4

6

8

3-D Load Profile

7/1/20067/2/2006

7/3/20067/4/2006

7/5/20067/6/2006

7/7/20067/8/2006

7/9/20067/10/2006

7/11/20067/12/2006

7/13/20067/14/2006

7/15/20067/16/2006

7/17/2006

7/18/2006

7/19/2006

7/20/2006

7/21/20064.00

5.00

12:00 AM

19PNWD-SA-9511

12:00

AM3:00

AM6:00

AM9:00

AM12:00

PM3:00

PM6:00

PM9:00

PM

7/21/2006

7/22/2006

7/23/2006

7/24/2006

7/25/2006

7/26/2006

7/27/2006

7/28/2006

7/29/2006

7/30/2006

7/31/2006

0.00

1.00

2.00

3.0012:00 AM

2:00 AM

4:00 AM

6:00 AM

8:00 AM

10:00 AM

12:00 PM

2:00 PM

4:00 PM

6:00 PM

8:00 PM

10:00 PM

7/1

/20

06

7/2

/20

06

7/3

/20

06

7/4

/20

06

7/5

/20

06

7/6

/20

06

7/7

/20

06

7/8

/20

06

7/9

/20

06

7/1

0/2

00

6

7/1

1/2

00

6

7/1

2/2

00

6

7/1

3/2

00

6

7/1

4/2

00

6

7/1

5/2

00

6

7/1

6/2

00

6

7/1

7/2

00

6

7/1

8/2

00

6

7/1

9/2

00

6

7/2

0/2

00

6

7/2

1/2

00

6

7/2

2/2

00

6

7/2

3/2

00

6

7/2

4/2

00

6

7/2

5/2

00

6

7/2

6/2

00

6

7/2

7/2

00

6

7/2

8/2

00

6

7/2

9/2

00

6

7/3

0/2

00

6

7/3

1/2

00

6

4.69-5.00

4.38-4.69

4.06-4.38

3.75-4.06

3.44-3.75

3.13-3.44

2.81-3.13

2.50-2.81

2.19-2.50

1.88-2.19

1.56-1.88

1.25-1.56

0.94-1.25

0.63-0.94

0.31-0.63

0.00-0.31Surface Chart

Box Plot Load Profile

20PNWD-SA-9511

3.0

4.0

5.0

6.0

Avg

Ele

cM

tr_W

att

s_p

erS

F

During Changes

Before 11-15-06

After 1-24-07

Scatter Chart by Date Range

21PNWD-SA-9511

0.0

1.0

2.0

0 20 40 60 80 100 120

AHU_TempOa

Avg

Ele

cM

tr_W

att

s_p

erS

F

Excel won’t create

Scatter Charts based on

PivotTables (PivotCharts).

ECAM includes

Scatter Charts based on

PivotTables.

Scatter Chart by Occupancywith Binned Weather

800

1,000

1,200

Av

g E

lecM

tr_k

W

22PNWD-SA-9511

0

200

400

600

20 30 40 50 60 70 80 90 100

1degBin

Av

g E

lecM

tr_k

W

Occ

Unocc

INTERVAL METER DATA ANALYSIS EXAMPLES

23

ANALYSIS EXAMPLES

Example: Load Profile by Daytype

24 Hour Occupied

24PNWD-SA-9511

24 Hour Occupied Schedule

Example: Load Profile by Daytype

Morning

Building occupied

(6 AM)

Unoccupied/ shutdown

•

25PNWD-SA-9511

Morning startup

Example: Typical Office Building

50% load reduction

between peak and nighttime

Long shutdown

period

26PNWD-SA-9511

5 day occupied schedule and unoccupied on

weekends

Example: Good Office Building

80% Peak Reduction

27PNWD-SA-9511

Good startup/shutdown

(6 AM/6 PM)

Building not occupied on weekend

Example: Comparing data before and after improvements

50% reduction

15% Peak load reduction

28PNWD-SA-9511

Additional 30% reduction at night

Comparing Consumption from Two Identical Buildings

29PNWD-SA-9511

Comparing Regression Analysis from Two Identical Buildings

30PNWD-SA-9511