What May Emerge as Renewables B L S l ?Become Large Scale?

J A tJay Apt

Tepper School of Business andD t t f E i i & P bli P liDepartment of Engineering & Public Policy

Carnegie Mellon University

March 8, 2011

2 www.RenewElec.org

Operating Wind Farms February 21, 2011p g y ,Wind farms > 5 MW

4

Does wind power have atmospheric risks?Does wind power have atmospheric risks?

“…the wind farm significantly slows down the wind…usually leading to a warming and drying of the surface air…”

Models show wind causes temperature h 1°Cchanges ~ 1°C

Annual surface temperature (GFDL d l)

90N

0.8

1

response (GFDL model)

30N

60N

0.4

0.6

0

-0.2

0

0.2

60S

30S

-0.6

-0.4

180W 120W 60W 0 60E 120E 180E90S -1

-0.8

Units: °KSurface (2 m) air temperature response. Drag perturbation of 0.005.Surface (2 m) air temperature response. Drag perturbation of 0.005. Showing 20 yr of perturbed run versus 20 years of control.

8

9

Wind sometimes fails for many days

BPA Balancing Authority Total Wind Generation

1250

1500 Sum of ~1000 turbines

500

750

1000

WM

5 10 15 20 25 30

250

500

5 10 15 20 25 30Date in January 2009

[CEC:2010] R. Masiello, K. Vu, L. Deng, A. Abrams, K. Corfee, J. Harrison, D. Hawkins, and K. Yagnik, “Research evaluation of wind generation, solar generation, and storage impact on the California grid,” tech. rep., Public Interest Energy Research (PIER), California Energy Commission, 2010. [C ] C il f l (C ) “ l f h i i f i d[CEER:2009] Council of European Energy Regulators (CEER), “Regulatory aspects of the integration of wind generation in European electricity markets.” 2009. [DOE:2008] “20% Wind Energy by 2030: Increasing Wind Energy’s Contribution to U.S. Electricity Supply,” tech. rep., US Dept. of Energy, 2008. [Doherty:2005] R. Doherty and M. O’Malley, “A new approach to quantify reserve demand in systems with significant installed wind capacity,” IEEE Transactions on Power Systems, vol. 20, no. 2, pp. 587–595, 2005. [ERCOT:2008] “ERCOT Wind Impact / Integration Analysis,” tech. rep., GE Energy Consulting, 2008. [EWIS:2010] EWIS, “European wind integration study (EWIS): Towards a successful integration of large scale wind power into European electricity grids,” tech. rep., European Wind Integration Study, 2010. [EWIS:2010] Wilhelm Winter, ed., “Towards A Successful Integration of Large Scale Wind Power into European Electricity Grids,” tech. rep., European Wind Integration Study, 2010. [MacCormack:2010] J. MacCormack, A. Hollis, H. Zareipour, and W. Rosehart, “The large-scale integration of wind generation: Impacts on price, reliability and dispatchable conventional suppliers,” Energy Policy, vol. 38, pp. 3837–3846, 2010. [MN:2006] R. Zavadil, “Final Report - 2006 Minnesota Wind Integration Study: Volume I,” tech. rep., EnerNex Corporation, 2006. [NREL:2010] D. Corbus, “Eastern wind integration and transmission study,” tech. rep., National Renewable Energy Laboratory/EnerNex, 2010. [NYSERDA:2005] John Saintcross, ed. “The Effects of Integrating Wind Power on Transmission System Planning, [ ] , g g y g,Reliability, and Operations, Report on Phase 2: System Performance Evaluation.” New York State Energy Research and Development Authority. [SPP:2010] T. B. Tsuchida, P. A. Ruiz, A. Rudkevich, P. W. Sauer, G. G. Lorenz ́on, R. Yeu, R. Baxter, J. Cooper, D. Cross-Call, S. L. Englander, and J. Goldis, “SPP WITF Wind Integration Study,” tech. rep., Charles River Associates, 2010.

11

[Strbac:2007] G. Strbac, A. Shakoor, M. Black, D. Pudjianto, and T. Bopp, “Impact of wind generation on the operation and development of the UK electricity systems,” Electric Power Systems Research, pp. 1214–1227, 2007.

All of these assume Gaussian (normal)statistical distributions for wind

12

15 Days of 10-Second Time Resolution Data

13

Fourier Transform to get the Power Spectrum

2.6 Days

30 Seconds30 Seconds

14

We can learn some importantthings from the power spectrumg p p

103

Smoothing by Adding Wind Farms

102

101 Wind Plant

24 Hours

100

101

al D

ensi

ty

10-1

ower

Spe

ctra

10-3

10-2Po

10-7 10-6 10-5 10-4 10-310-4

Frequency (Hz)

16

103

Smoothing by Adding Wind Farms

102

101 Wind Plant4 Wind Plants24 Hours

100

101

al D

ensi

ty

10-1

ower

Spe

ctra

10-3

10-2Po

10-7 10-6 10-5 10-4 10-310-4

Frequency (Hz)

17

103

Smoothing by Adding Wind Farms

102

101 Wind Plant4 Wind Plants20 Wind Plants

24 Hours

100

101

al D

ensi

ty

10-1

ower

Spe

ctra

10-3

10-2Po

10-7 10-6 10-5 10-4 10-310-4

Frequency (Hz)

18

103

Smoothing by Adding Wind Farms

102

101 Wind Plant4 Wind Plants20 Wind Plants

24 Hours

100

101

al D

ensi

ty

f-1.67

10-1

ower

Spe

ctra

Source: Katzenstein, W., E. Fertig, and J. Apt, The Variability of Interconnected Wind Plants. Energy Policy 2010 38(8): 4400 4410

10-3

10-2Po

f-2.56

Energy Policy, 2010. 38(8): 4400-4410.

10-7 10-6 10-5 10-4 10-310-4

Frequency (Hz)

19

Smoothing by Adding Wind Farmshas diminishing returns… has diminishing returns

Source: Katzenstein, W., E. Fertig, and J. Apt, The Variability of Interconnected Wind Plants. Energy Policy, 2010. 38(8): 4400-4410.

Each hour decompose wind energy into the following components

Up Energy

Down Energy ← qH

qForecasted

Hourly

Ene

rgy

Compo

nent

Wind Energy

Load Following Component

↓ Capacity

↑ Capacity

← qH

Regulation

Component0 15 30 45 60 min

t

nt

21

2008

11

12

9

10

$/M

Wh)

7

8

bilit

y C

ost (

$

5

6

Varia

b

0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.53

4

Capacity Factor

22

2008 and 2009

11

12

20082009

9

10

/MW

h)

7

8

bilit

y C

ost (

$

5

6

Varia

b

0 2 0 25 0 3 0 35 0 4 0 45 0 53

4

0.2 0.25 0.3 0.35 0.4 0.45 0.5Capacity Factor

23

11

12

Highest Variability Cost Wind Plant as Starting PointMedian Variability Cost Wind Plant as Starting PointLowest Variability Cost Wind Plant as Starting PointIndividual Wind Plants

9

10

$/M

Wh)

7

8

abilit

y C

ost (

6

Varia

0 2 4 6 8 10 12 14 16 18 204

5

Number of Interconnected Wind Plants

Number of Interconnected Wind Plants

24

Hydroelectric Power has Droughts

26

Wind Probably Does Too

Source: Katzenstein, W., E. Fertig, and J. Apt, The Variability of Interconnected Wind Plants.

Energy Policy, 2010. 38(8): 4400-4410.

Operating Solar PV February 21, 2011p g y ,Units > 5 MW

28

Solar

29

Comparison of Wind with Solar PV4.6 MW TEP Solar Array (Arizona)

4000

y ( )

2000

3000

4000

kW

0

1000

1400000 1450000 1500000 1550000

Seconds since 00:00:00 Jan 1, 2007,

2000

3000

W

0

1000

2000

kW

(b)

kW

30

250 750 1250Minutes

Comparison of wind and solar PV

Solar PV

Wind

31

Source: CEIC Working Paper CEIC-07-12, available at www.cmu.edu/electricity

Work with Warren Katzenstein

NOx and CO2 Emissions from Gas Turbines Paired with Wind or Solar for Firm Power

GE LM6000

32

GE LM6000sealegacy.com

Siemens-Westinghouse 501FDsummitvineyardllc.com

Siemens-Westinghouse 501FD Regression Analysis

3.5

2.5

3

min

)

1.5

2

Em

itted

(lb/

m

GE Document GER-3568G

0 5

1NO

x E

0 20 40 60 80 100 120 140 160 180 2000

0.5

Power (MW)

33

Power (MW)

Emissions Factors(a) LM6000

0 4(b) LM6000

LM6000Steam,no SCR

0.2

0.25

0.3

0.35

(tonn

es/M

Wh)

Predicted

0 2

0.25

0.3

0.35

0.4

ns (k

g/M

Wh)

0.05

0.1

0.15

CO

2 Em

issi

ons

Expected

0

0.05

0.1

0.15

0.2

NO

x Em

issi

o

Expected

Predicted

501FDDLN

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

α (Penetration Factor)0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

α (Penetration Factor)

0.35)

(c) 501FD0.4

(d) 501FD

DLN,SCR

0 15

0.2

0.25

0.3

ons

(tonn

es/M

Wh)

Predicted

0.2

0.25

0.3

0.35

sion

s (k

g/M

Wh) Predicted

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.05

0.1

0.15

(P t ti F t )

CO

2 Em

issi

o

Expected

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.05

0.1

0.15N

Ox E

mis

s

Expected

34

α (Penetration Factor) α (Penetration Factor)

Will Electric Vehicles Come to the Rescue?

V2G energy arbitrage profit sensitivity to battery pack replacement costV2G energy arbitrage profit sensitivity to battery pack replacement cost with perfect information in the three cities studied. The symbol indicates

the median annual profit for the years studied and the range indicates the most and least profitable years.

35

But, a very small battery installation can buffer a lot of wind variabilitycan buffer a lot of wind variability

36

Final Comments• None of this means that wind,

geothermal or solar (if costs evergeothermal, or solar (if costs ever come down) can't be used at large

l b t i d/ l ill iscale, but wind/solar will require a portfolio of fill-in power (some with p p (very high ramp rates, some with slow) and R&D is required to optimizeand R&D is required to optimize emissions control for fast and deep ramping

37

ramping.

Thank you.

Jay Aptapt@cmu.edu

38