NREL Research Support Facility

• 355,000 ft2 Office Building

• 1350 Occupants • 25 kBtu/ft2 + 6 (data center)+ 4 (180

additional people) = 35kBtu/ft2 • 50% energy savings (before PV) • $250/ft2

• LEED Platinum • Zero Energy based on Site, Source,

Carbon, & Utility Cost • 60’ Section for daylight & Natural Vent. 0 Energy

• Optimal glass to wall area ratio • Engineered glass shading • Light louvers & Dynamic Glazing • Transpired Solar Collectors • Thermal Storage • N/S Orientation • Radiant Hydronic Heat & Cool with

evaporative cooling • raised floor dedicated outdoor air

Challenge: Implementing New Energy Efficient Technologies

Ron Judkoff, Chief Engineer

CONTROLS

When Materials Selection was Easy!

New Problem: Too much technology to choose from

Evolutionary elimination of bad A&Es

A Lot of Things Need to Go Right to “Actually” Achieve High Efficiency

• Owner wants a new high efficiency building or deep retrofit

• Selected A&E knows how to design it • Constructors and installers are up to speed • Controls are properly set up with the correct

control strategy and tested in operation • Building operator understands system • Occupants are comfortable and don’t

sabotage system

180,000 ft2 Big Box, Cold Climate: Retrofit OptEplus Retrofit Package Optimization

90.1-2004 EUI:76 kbtu/ft2/y

Pre-Retrofit Base 54 kbtu/ft2/y

SG1 Analysis 35 kbtu/ft2/y

+30% +35%

On the Path to Commercial ZEBs Oberlin Lewis Center for Environmental Studies

-14,000 ft2 Classroom and Offices - 60 kW PV

- Daylighting - Natural Ventilation

- Ground Source Heat Pump

The RJ BLOOPERS Archive

0

200

400

600

800

1,000

1,200

1,400

1,600

Jan-

00F

eb-0

0M

ar-0

0A

pr-0

0M

ay-0

0Ju

n-00

Jul-0

0A

ug-0

0S

ep-0

0O

ct-0

0N

ov-0

0D

ec-0

0Ja

n-01

Feb

-01

Mar

-01

Apr

-01

May

-01

Jun-

01Ju

l-01

Aug

-01

Sep

-01

Oct

-01

Nov

-01

Dec

-01

Jan-

02F

eb-0

2M

ar-0

2A

pr-0

2M

ay-0

2Ju

n-02

Jul-0

2A

ug-0

2S

ep-0

2O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb

-03

Dai

ly A

vera

ge M

onth

ly C

onsu

mpt

ion

and

Pro

duct

ion

(kW

h/da

y)

Utility BillsEquipmentTotal LightsTotal CoolingTotal HeatingPV Production

Oberlin Before & After Fixes

Control fix to EB-1 HP swap-out

Measured: 54 kBtu/ft2/yr

30

ASHRAE 90.1-2001: 54 kBtu/ft2/yr

45

N. Facing Trombe Wall. Oops!

S. Facing Direct Gain Problem fix?

They flipped the plan. The idea was to have North Daylighting and a South Facing Thermal Storage Wall with a clerestory and a light shelf for some bi-directional daylight.

Even the same A&E firm that has just done two really well designed energy efficient buildings can lose its corporate memory. And by the time I realized what had happened, it was too late. Lesson: Never Sleep at the Wheel.

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

11

Image Processing Occupancy Sensor (IPOS)

• IR Occupancy sensing not robust • Detection of motion not sufficient • Complaints may lead to no energy saved

Source (top): Rab Lighting; Photo: Luigi Gentile Polese, NREL

Cre

dit:

Mar

jorie

Sch

ott(N

REL

)

Correlate local measurements to occupant feedback

“Phidgets” collect temperature, humidity, and light readings through a USB port on occupant’s computer

Temperature Comfort

Humidity Comfort

Lighting Comfort

Glare Comfort

Noise Comfort

Air Movement

Comfort

Temp C°

Lighting Lux

Humidity %

Timestamp

Cold Dry Glare Noisy 23 354.0 28 05/23 17:13

18

IPOS (Image Processing Occupancy Sensor): Fusion of Image Sensing

Face 3

Images: Luigi Gentile Polese, N

REL

1 Motion

2 People

19

IPOS

Building system

sensors

Building system actuators

Building Energy Management System (EMS)

Robust Algorithms for Finely Grained: • Lighting Controls • Daylighting • Set Point

Management • Demand

Controlled Ventilation • Open Embedded OS

• OpenCV Computer Vision Library

• IPOS Algorithms • BACnet stack

Top and bottom left: Dennis Schroeder, 24420, 24418 N

REL; Control Panel: Warren Gretz, N

REL 12762

Supplemental Slides

20

End

21

Functional Model Physical Sensor

In-memory Buffers

Space Illuminance Virtual Sensing Integration

Occupant Count & Location

Image Acquisition

Occupant Activity

Face Detection

People Detection

Motion Detection

• Human Occupancy • Confidence Levels • Number & Location of Occupants • Occupant Activity Levels

• Illuminance/Light Meter • Light Dimming Power Levels • Multiple Zones Outputs

IPOS Outputs over BACnet

22

Potential Energy Savings En

ergy

Sav

ings

%

60.3

87.8 85.9

41.2

80.3 75.6

15.5

54.1

43.3

0

10

20

30

40

50

60

70

80

90

100

Weekdays (6AM-6PM) Weekdays (6PM -9PM) Weekend (7AM-7PM)

Theoretical Best IPOS PIR (15min)

Measured Versus Modeled Monthly and Cumulative EUI

23

Cre

dit:

Cha

d Lo

bato

/NR

EL

0

4

8

12

16

20

24

28

32

36

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

Mod

el

RS

F

PV

October2010

November2010

December2010

January2011

February2011

March2011

April2011

May2011

June2011

July2011

August2011

September2011

Annu

al E

UI (

kBtu

/ft2 )

Mon

thly

EU

I (kB

tu/ft

2 )

PVData CenterMechanical SystemsPlug LoadsLightingHeatingCoolingModel Annual EUIRSF Annual EUI

The PV system is sized for an annual EUI of 35.1 kBtu/ft2.

Note: The annual EUI values are demand side values and do not include the PV generation. Note: Model predictions are based on TMY Weather Data

Innovation for Our Energy Future

Transpired Solar Collector, Conserval

Electrochromic Window Sage

Light- Louver

Solatube

NREL/Industry developed technologies Independently specified

by Haselden Team

BIPV Product Examples

Light, flexible PV roofing shingle for direct rooftop mounting

PV roofing shingles

Flexible roof-top PV shingles

Building-integrated PV metal roofing modules

PowerGuard® system

Neutral Cost Point: Greensburg BEopt Analysis

(2000 ft2, 2-story, 16% window to floor area ratio, unconditioned basement)

IECC 2003

BEopt Beta 0.8.04

Energy Star

Greensburg Target:

Neutral Cost

Cost Effective

Technology Gap

Technology Gap

Example: Greensburg Neutral Cost Package1

• R22 wall assembly (2x6: R-19 batts+ foam sheathing) • R50 ceiling assembly • R10 basement • .0001 SLA (2 ACH50) • Low e/low SHGC glazing, Argon Fill (0.28 U-value, 0.37 SHGC) • 80% CFL Lighting • SEER 18 AC • AFUE 90+ furnace • Gas tankless hot water, EF 0.8+ • Tight ducts (Mastic, 5% Leakage), in conditioned space • Energy Star Appliances • 1.5 kWDC PV System • BA QA (moisture control, …) Estimated cost increase relative to standard home2,3: +$10.00-$13.00/ft2

Notes: 1. Equivalent packages may be substituted, based on specific builder preferences 2. Does not include costs associated with builder/contractor training and changes in business practices. 3. Incremental costs will depend on current builder practice

Where we are today

Code

Case Study Bldgs

NREL RSF ZEB 25 kBtu/ft2/yr

(79 kWh/m2/yr)

(284 kWh/m2/yr)

(223 kWh/m2/yr)

Shading Device: Enough steel to build the Eiffel Tower? (and doesn’t work very well)

My Zero Energy House In Kedougou, Senegal. 250km from Kayes, Mali

(hottest regularly inhabited place on

earth).

1972

Buildings are for people, so where are they in the feedback

loop?