www.arpa-e.energy.gov

BUILDING ENERGY EFFICIENCY

Ravi Prasher, Technology Development Program Manager, Intel Corporation

Colin McCormick, Building Technologies Program, U.S. Department of Energy

March 1, 2010

ARPA-E Pre-Summit Workshop

www.arpa-e.energy.gov

“There is a growing realization that we should be able to build buildings that

will decrease energy use by 80 percent with investments that will pay for

themselves in less than 15 years. Buildings consume 40 percent of the energy

in the U.S., so that energy efficient buildings can decrease our carbon

emissions by one third.”

Secretary Chu, Caltech Commencement, June 12, 2009

VISION

Buildings use 72% of nation’s electricity and 55% of its natural gas.

Buildings construction/renovation contributed 9.5% to US GDP and employs

approximately 8 million people. Buildings’ utility bills totaled $370 Billion in 2005.

Source: Buildings Energy Data Book 2007

By 2030, Business as Usual

• 16% growth in electricity

demand

• Additional 200 GW of

electricity at cost of $500-

1000B, or $25-50B/yr

Buildings Can Provide

Grid-Level Storage

BUILDINGS PLAY A KEY ROLE IN

AMERICA’S ENERGY FUTURE

China

India

8.5%/yr growth

New: 80% reduction

Existing: 50% reduction

COMMERCIAL BUILDINGS OFFER A

SIGNIFICANT OPPORTUNITY FOR

ENERGY SAVINGS

• HVAC (space heating,

space cooling, and

ventilation): 31.4%

• Lighting: 14.1%

• Water heating: 8.6%

• Increasing

miscellaneous loads

• Air conditioning is a

key driver of peak

electricity demand.

Source: 2009 Buildings Energy Data Book

(http://buildingsdatabook.eren.doe.gov/TableView.aspx?table=1.1.5)

Space heating, 20.0%

Lighting,14.1%

Space cooling, 9.9%

Water heating, 8.6%

Electronics, 7.2%

Refrigeration, 5.2%

Wet clean, 3.1%

Computers, 2.7%

Cooking, 2.3%

Ventilation, 1.5%

Other,17.3%

Adjust to SEDS, 8.1%

Total: 41.0

Quads

TOTAL PRIMARY ENERGY

CONSUMPTION

BY BUILDINGS (2010)

• Buildings are

responsible for ~ 40%

of CO2 emissions in

the U.S.

• Buildings are

responsible for ~ 2,300

Tg CO2 eq. (MMT)

• Cooling alone is

responsible for ~ 5%

of CO2 emissions in

the U.S.

Source: 2009 Buildings Energy Data Book

(http://buildingsdatabook.eren.doe.gov/TableView.aspx?table=1.1.5)

TOTAL CO2 EMISSIONS BY

BUILDINGS

(2010)

Space heating, 447

Lighting, 336

Space cooling, 236Water

heating, 197

Electronics, 172

Refrigeration, 125

Wet clean, 73

Computers, 64

Cooking, 54

Ventilation, 35

Other, 405

Adjust to SEDS, 194

Total: 2,338

Tg CO2 eq.

Courtesy: World Business Council for Sustainable Development (WBCSD) Report

on Energy Efficiency in Buildings, July 2008

Need:

• Tools to integrate process & communities

• Tools to integrate building design and operations

• Align incentives

FRAGMENTATION OF INDUSTRY

AND PROCESS

SYNERGISTIC APPROACH

David Culler

University of

California,

Berkeley

Member of National

Academy of Engineering

Credited with pioneering

work on sensor networks &

computer architecture

Sanjay Sarma

Massachusetts

Institute of

Technology

Credited with developing

many standards and

technologies that form the

foundation of the commercial

RFID industry.

Sam Baldwin

DOE, Office of

Energy Efficiency

and Renewable

Energy

Chief Technology Officer and

Member, Board of Directors

Satyen Mukherjee

Phillips Research

Chief scientist and senior

director for research strategy

in North America

Ron Judkoff

National Renewable

Energy Laboratory

Director, Center for Buildings

and Thermal Systems

Jeff Snyder

California Institute of

Technology

Thermoelectric Materials

and Devices

Breakdown of 76 attendees

by affiliation

Representative Attendees

ARPA-E & EERE BUILDINGS

WORKSHOPDECEMBER 15, 2009

25%

26%30%

18%

National Labs Academia

IndustryFederal Agencies

DOE, DOD, NSF, NIST, GSA, OSTP

What is the state-of-the-art and where do we want to be?

What are the gaps, challenges, and barriers?

What are the potential approaches to overcoming the barriers?

What are the specific performance and cost metrics?

What is the time horizon for goals?

What are the challenges and barriers for validation and adoption?

What levels of investment do we need to develop and deploy these

technologies?

What is the return on investment?

KEY QUESTIONS

ACHIEVING EFFICIENT BUILDINGS IS VERY

CHALLENGING, BUT EXISTING EXAMPLES &

INVESTMENT AREAS PROVIDE HOPE

Dr. J Michael McQuade,

United Technologies Corp.

Buildings Systems Grand Challenges to Increase

Energy Efficiency

Greatest efficiency improvements may lie in

exploitation of otherwise detrimental sub-system

interactions

Efficient designs must also deal with uncertainties

and bottlenecks in design and operation

Investments are required to achieve aware,

interactive, reconfigurable buildings

WE SHOULD STRIVE TO BRIDGE THE TRADEOFF

THAT EXISTS BETWEEN ENERGY CONSUMPTION

AND PERFORMANCE

Yi Jiang

Tsinghua University

What is the approach to reach the low energy goal for

buildings?

There is a huge difference in building energy

consumption between OECD and developing

countries such as China

Difference due to lifestyle difference - energy

efficiency designs must take lifestyle into account

Decentralized, “part time part space” service

system may be the US solution

Other areas for study - independent control of

temperature and humidity, evaporative cooling

1. Measurement & Communication (sensors, data protocols, power)

• Shyam Sunder (NIST)

• David Culler (UC Berkeley)

2. Simulation & Computation (computational models, calibration)

• Ron Judkoff (NREL)

• Michael Wetter (LBNL)

3. Systems Approach to Fault Diagnostics and Controls

• Scott Bortoff (Mitsubishi Electric)

• Srinivas Katipamula (PNNL)

4. Active and Passive Thermal Devices and Components

• Ravi Prasher (Intel)

• Sam Baldwin (DOE/EERE)

BREAKOUT SESSIONS

Highly efficient buildings exist today, but measured performance rarely

meets design performance.

Achieving designed performance requires labor-intensive, high-skill

continuous operation & maintenance. This must be made easier.

Solution will be part technical, part behavioral, part economic, part policy.

Enormous opportunities exist in improved systems integration.

Investable areas for building science components include:

– Lighting (Solid State lighting program in EERE)

– Active building envelope technologies (controllable)

– Long-duration, fast-charging/discharging thermal storage

– Enhanced passive ventilation systems

– Fast humidity control systems

– High-efficiency, low-GWP refrigerant replacements/not-in-kind systems

KEY LESSONS

The SpreadEUI in kBTU/sq.ft.-yr

Analysis of 121 LEED-Rated BuildingsLow-to-Medium Energy Use Intensity Buildings

Measured to Design Ratio

Towards Zero-Net EnergyM. Frankel, “The Energy Performance of LEED Buildings,” presented at the Summer Study on Energy Efficient

Buildings, American Council of Energy Efficiency Economy, Asilomar Conference Center, Pacific Grove, CA, August

17-22, 2008.

Labels/codes are for Design Performance, NOT based on Measured Performance.

Gap

• Lack of Measurements & Policies

Requiring it

MISMATCH BETWEEN MEASURED &

DESIGNED PERFORMANCE

MEASUREMENT & COMMUNICATIONSENSOR PARAMETERS

Key parameters Future/desirable parameters

Mass flow (air, water, steam, fuel) Carbon dioxide

Temperature, pressure, humidity VOCs

Light levels Biologics

Electrical power (sub-metering major

equipment, tenant groups)

Occupant density

Indoor environmental quality Plug load power consumption

Building boundary conditions or

configuration (windows, doors, etc)

Granular equipment consumption (zonal)

Occupancy and comfort levels, acoustics Parameterized personal comfort

External conditions (weather, insulation)

Measurement Type I:Conservation Laws

Utility Line

Distributed

Use

Y

xi

Is Y = Σxi ?

Measurement Type II: Model Validation

Heat Loss

Heat Load

Inlet Outlet

Measurement ModelInverse

ForwardCalibrated

Model

People

,,TYm

,,TYm

,T

MEASURED PERFORMANCE IS

LINKED TO MEASUREMENT TYPE

MEASUREMENT & COMMUNICATIONSENSOR CHARACTERISTICS/BARRIERS

Key characteristics Key barriers

Sensor density based on gradients Sensor drift and mis-calibration

Location at equipment, exterior faces Large dataset handling (search, etc)

Ease of access/maintenance Inadequate meta-data

Dynamically configurable Connectivity (wired/wireless?)

Integration with controls Power (energy harvesting?)

Robust to single-point failure Interoperability

Fully autonomous Legacy systems, protocol changes (3-4 years)

Long sensor lifetime (20+ years?) Expense/difficulty of mesh network?

SIMULATION & COMPUTATIONSIMULATION NEEDS/CHALLENGES

Issues Needs

Modeling inefficient buildings is more difficult

than efficient ones

Tools/interface tailored to different

user groups

Occupant behavior in larger buildings is highly

stochastic

Fast/real-time simulation

Weather data is hourly – hard to simulate at finer

timescales

Ties into other toolsets (MATLAB,

Mathworks, etc)

Wind data is at a single elevation (10 m) – large

impact on buildings

Smooth transition from design

model to operating model

Moisture is not well-modeled Better inverse-modeling capability

Ground coupling is not well-modeled Better theory of in-building

convection

Equipment performance (esp. legacy) is not well-

modeled

Communicating anticipated loads

to utilities

SIMULATION & COMPUTATIONTIME AND LENGTH SCALES

Thanks to Michael McQuade, UTC

Reduce energy consumption by cooling equipment

Alternate refrigerants with GWP < 1

On demand local cooling

CHALLENGE: REDUCING GHG

EMISSIONS FROM SPACE COOLING

Standard residential A/C working fluid

is moving from R-22

(production/import ban starts in 2010)

to R-410A and other HFCs.

Standard vehicle A/C working fluid is

moving from R-12 (production/import

ban started in X) to R-134a.

R-410A 100-year GWP: 2,088.

R-404A 100-year GWP: 3,922.

AZ-20 (Honeywell)

Puron (Carrier)

Suva 410A (DuPont)

Forane 404A (Arkema)

Sources:

1) Velders et al, PNAS 106, 10949 (2009),

http://www.pnas.org/content/suppl/2009/06/22/0902817106.DCSupplemental/09028171

06SI.pdf#nameddest=ST2

2) EPA: http://www.epa.gov/Ozone/snap/refrigerants/lists/homeac.html

GHG EMISSIONS FROM HFC

WORKING FLUIDS

Under a 450-ppm stabilization

scenario, HFCs are likely to

contribute 28-45% of total GHG

emissions in 2050, on a CO2-eq

basis.

Growth is driven by demand for A/C

and refrigeration in the developing

world.

Source: Velders et al, PNAS 106, 10949 (2009),

http://www.pnas.org/content/suppl/2009/06/22/0902817106.DCSuppleme

ntal/0902817106SI.pdf#nameddest=ST2

HFC EMISSIONS: AIR CONDITIONING

& REFRIGERATION

CURRENT COOLING SYSTEMS

Cooling type COP Source

Air cooled systems ~4 EERE, DOE

Water cooled systems (evaporative

cooling of condensers)

~7 EERE, DOE

COP = Cooling load/ electrical energy supplied

Can we get COP same as water cooled chiller but without loss of water?

ENERGY END USE INTENSITIES IN

BUILDINGSTH1

TH2

BJ1

BJ2

BJ3

US2

FR1

SH1

US1

照明办公电耗kWh/m2.a

26.1 24.5 18.4 23.1 26.5

152.7 149.4

20.8N/A

0

50

100

150

200

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Lighting & appliances (kWh/m2.a)冷机电耗kWh/m2.a

8 6.59

19.526.7

13.7

48.152

43.97

22.2

0

10

20

30

40

50

60

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Chiller (kWh/m2.a)

风机电耗kWh/m2.a

0 7.7 0 8.121.5

168.7197.1

13.525.7

0

50

100

150

200

250

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Fan for AC system (kWh/m2.a)水泵电耗kWh/m2.a

02.9

0

8.96.4

25.3

17.519.5

9.8

0

5

10

15

20

25

30

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Pump for AC system (kWh/m2.a)

耗冷量GJ/m2.a

0.07 0.12 0.18

0.44

0.20

1.041.12

0.63

0.36

0.0

0.2

0.4

0.6

0.8

1.0

1.2

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Cooling (GJ/m2.a) 耗热量GJ/m2.a

0.28 0.30.2

0.3650.25

0.82 0.88

0.45

0.17

0.0

0.2

0.4

0.6

0.8

1.0

TH1 TH2 BJ1 BJ2 BJ3 SH1 US1 US2 FR1

Heating (GJ/m2.a)

USA USA

China

France

Courtesy: Yi Jiang, Tsinghua University

the Building's Cooling Electricity Consumption with

Different Operations in Shanghai21.8

12.4

8.8

5.33.3

0

5

10

15

20

25

Standard Case Raise the

Starting

Temperature

Open

Windows

Part Time

Operation

Part Space

operation

Co

oli

ng

Ele

ctri

city

Co

nsu

mp

tio

n o

f th

e

Wh

ole

Yea

r(kW

h/m

2·a)

HOW BUILDINGS ARE OPERATED IS

IMPORTANT

Comparing with the energy

consumption of the five cases– Total cooling electricity

consumption：

21.8kWh/m2 to 3.3kWh/m2!

Measured data & simulation

data in Beijing

0123456789

101112131415

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

住户编号

空调能耗指标

(kW

h/m

2)

平均值 2.3 kWh/m2 Average 2.3kWh/m2

En

erg

y C

on

sum

pti

on

of

Co

olin

g S

yste

m

(kW

h/m

2)

Apartment No.

0123456789

101112131415

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

住户编号

空调能耗指标

(kW

h/m

2)

平均值 2.3 kWh/m2 Average 2.3kWh/m2

En

erg

y C

on

sum

pti

on

of

Co

olin

g S

yste

m

(kW

h/m

2)

Apartment No.

Ele

ctri

city

Co

nsu

mp

tio

n o

f C

oo

ling

Sys

tem

0123456789

101112131415

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

住户编号

空调能耗指标

(kW

h/m

2)

平均值 2.3 kWh/m2 Average 2.3kWh/m2

En

erg

y C

on

sum

pti

on

of

Co

olin

g S

yste

m

(kW

h/m

2)

Apartment No.

0123456789

101112131415

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

住户编号

空调能耗指标

(kW

h/m

2)

平均值 2.3 kWh/m2 Average 2.3kWh/m2

En

erg

y C

on

sum

pti

on

of

Co

olin

g S

yste

m

(kW

h/m

2)

Apartment No.

Ele

ctri

city

Co

nsu

mp

tio

n o

f C

oo

ling

Sys

tem

the Building's Cooling Electricity Consumption with

Different Operations in Beijing

16.5

9.8

5.0

2.0 1.2

0

5

10

15

20

Standard

Case

Raise the

Starting

Temperature

Open

Windows

Part Time

Operation

Part Space

operation

Co

oli

ng

ele

ctri

city

co

nsu

mp

tio

n o

f th

e

Wh

ole

Yea

r(kW

h/m

2·a)

Cooling On Demand in

Space & Time

Courtesy: Yi Jiang, Tsinghua University

>10X

REHEAT FIGHTING WITH COOLING

A typical Air Handling Process in a US bldg: VAV + Reheat

22 ℃

17℃ 21℃ 14℃ 20℃

Cooling CoilVAV box withRe-heaterC ooling: 264kW

H eating: 228kW

22 ℃

17℃ 21℃ 14℃ 20℃

Cooling CoilVAV box withRe-heaterC ooling: 264kW

H eating: 228kW

Courtesy: Yi Jiang, Tsinghua University

HEATING & COOLING IN AN USA

CAMPUS IN PHILADELPHIA

2006年度UPENN逐月总冷量和热量

0

50000

100000

150000

200000

250000

7月 8月 9月 10月 11月 12月 1月 2月 3月 4月 5月 6月

GJ 冷量

热量

Heating and Cooling Load

0

20

40

60

80

100

120

140

160

180

200

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun

MJ/m2

cooling load

heating load

Heating and Cooling Load

0

20

40

60

80

100

120

140

160

180

200

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun

MJ/m2

cooling load

heating load

Courtesy: Yi Jiang, Tsinghua University

Waste

Latent load

Sensible load

THERMAL LOADS ON A/C

Reduction of temperature and humidity

Two types of loads

– Sensible load: mCp (Tamb – Tspace)

– Latent/humidity load: mhfg (xamb-xspace)

m = mass flow rate of air

Cp = specific heat of air

hfg = latent heat of water

x = humidity ratio (mass of water

vapor/mass of air)

Minimum theoretical work required to go from 1 to 2 = Availability

(exergy) of moist air

Sensible load

Latent load

Reheat load Extra cooling load

SEPARATE CONTROL OF HUMIDITY

& SENSIBLE LOAD

Which is a more energy efficient path?

DESICCANT SYSTEMS

BACKGROUND ON DESICCANTS

Desiccants material that absorb moisture:

During moisture adsorption/absorption (sorption)

– Latent is converted to sensible heat: Water vapor converts to liquid

– Heat is released due to sorption (typically small compared to L to S

conversion)

– Therefore process ~ isoenthalpic

During desorption in the regeneration of moisture to atmosphere

– Regeneration energy = sum of

• Latent heat

• Heat necessary to raise the desiccant to a temperature high

enough to make its surface vapor pressure higher than that of

surrounding air

– Desorption rate (r) = A*exp (-Ea/kBT): Arhenius Rate

Extra energy needed to to desorb the desiccant

amb

PvaporambambPvapor

amb

PairambambPairsT

TcTTTcx

T

TcTTTcw ln)(ln)(

ambvapor

vapor

ambvapor

ambvapor

vapor

ambairlP

PTxR

PP

PPTRw

__

lnln

MINIMUM THEORETICAL WORK FOR

AIR CONDITIONING OF MOIST AIR

Mina, Newell,& Jacobi, Int. J. Refrigeration, 28. 784 – 790 (2005)

For sensible or mechanical cooling (Carnot)

For dehumidification or latent load cooling (Entropy of mixing):

Availability (Exergy)

~ 60 – 120 oC

~500 oC

CarnotWs

Q1

Tamb

Thot

COPlatent Ql

Qd

Ql

Wl /Carnot

Ideal COPlatent = ~ 2.5-6.5

Real COP = ~ 0.7 – 1.0

Sensible HeatLatent Heat

IDEAL SYSTEM AT THERMODYNAMIC

LIMIT

Difference between current energy input and ideal system can be reduced by 50% by

•Increasing the COP of vapor compression (without loss of water!)

•Increasing the COP of the desiccant

REAL SYSTEMS VS. THEORETICAL

LIMIT

Current

systems

FOA

targets

Primary energy use

0

20

40

60

80

100

120

140

160

180

200

1 2 3 4 5 6 7 8

COP_Vapor-compression

Pri

mary

En

erg

y (

kJ/k

g)

Vapor

compression/mechanical

cooling

Desiccant + Vapor compression

(COPlatent = 1.4)Theoretical limit

Current

Systems

Tamb = 90 oF, RH = 0.9

Tsupply = 55 oF, RH = 0.5

Pri

ma

ry E

ne

rgy (

kJ

/kg

)

COPvapor-compression

REFRIGERANTS WITH GLOBAL

WARMING POTENTIAL (GWP) ≤ 1

Magnetic

refrigerants

Thermoacoustic

refrigeration

Thermoelectric

refrigeration

Is it possible to get 1 Ton (3.5 kW) of cooling with COP >4?