1

Incorporating Wind into a Natural-gas Turbine Baseload Power System Increases NOx and CO2 Emissions from the Gas Turbines

CMU - FUTURE ENERGY SYSTEMS: EFFICIENCY, SECURITY, CONTROL

Warren Katzenstein and Dr. Jay Apt

warren@cmu.eduapt@cmu.edu

March 11th, 2008

22

Today I will discuss

• Background and Motivation• Research Question• Approach

– Actual wind data– Actual emissions data from two types of natural-gas turbines

• Model Construction– General Electric LM6000 turbine– Siemens-Westinghouse 501FD turbine

• Results• Implications

3

Renewables Portfolio Standards

State Goal

☼ PA: 18%¹ by 2020

☼ NJ: 22.5% by 2021

CT: 23% by 2020

MA: 4% by 2009 +1% annual increase

WI: requirement varies by utility; 10% by 2015 goal

IA: 105 MW

MN: 25% by 2025(Xcel: 30% by 2020)

TX: 5,880 MW by 2015

☼ AZ: 15% by 2025

CA: 20% by 2010

☼ *NV: 20% by 2015

ME: 30% by 200010% by 2017 - new RE

State RPS

☼ Minimum solar or customer-sited RE requirement* Increased credit for solar or customer-sited RE

¹PA: 8% Tier I / 10% Tier II (includes non-renewables)

HI: 20% by 2020

RI: 16% by 2020

☼ CO: 20% by 2020 (IOUs)*10% by 2020 (co-ops & large munis)

☼ DC: 11% by 2022

DSIRE: www.dsireusa.org September 2007

☼ NY: 24% by 2013

MT: 15% by 2015

IL: 25% by 2025

VT: RE meets load growth by 2012

Solar water heating eligible

*WA: 15% by 2020

☼ MD: 9.5% in 2022

☼ NH: 23.8% in 2025

OR: 25% by 2025 (large utilities)5% - 10% by 2025 (smaller utilities)

*VA: 12% by 2022

MO: 11% by 2020

☼ *DE: 20% by 2019

☼ NM: 20% by 2020 (IOUs)10% by 2020 (co-ops)

☼ NC: 12.5% by 2021 (IOUs)10% by 2018 (co-ops & munis)

ND: 10% by 2015

4

Intermittent Power

Time

Power

1 Hour 2 Hour

5

Research Question

Does operating one or more gas turbines to fill in intermittent wind power result in increased NOx and CO2 emissions compared to full-power steady-state operation of natural-gas turbines?

6

Approach

+

+

+

1

2

n

=

Firm PowerVariable PowerCompensating Power

Time

Power

Gas

Wind

7

Approach

Wind Power Output

Calculate Power Leveland Ramp Rate Needed

GT’s Control and Regression Map

GT Emissions Model

+-

Error

Calculated Emissions

Ideal Fill Power

Realized Fill Power

8

Objective Function for Baseload Plant

9

Calculating Pollutant Mass Emissions of Baseload Plant

10

0 0.5 1 1.5 2 2.50

10

20

30

40

50

60

70

80

90

Time (hours)

Pow

er (M

W)

Data Slice from Output of Two Wind Farms in Pennsylvania

Wind Data Obtained

• Wind Data– Output of 3 existing

wind farms• Eastern• Southern Great

Plains• Central Great

Plains– 1 to 10 seconds

resolution– 32 total days of data– From anonymous

source

11

Gas Turbine Data Obtained• NOx emissions & heat

rate for 7 CTs & 1 NGCC– 1 minute resolution– Ranges from 38 days of

data to 135 days of data– CTs are LM6000s– Have

• Gas flow (HSCFH)• Load (MW)• NOx ppm and lbs• NOx ppm corrected to

15% O2

• O2 %• Heat rate (mBtu/hr)

– From anonymous source

90 95 100 105 110 115 1200

5

10

15

20

25

30

35

40

45

50

Time (hours)

Pow

er (M

W)

Data Slice of Power Output of LM6000 Data Obtained

12

GE LM6000 – Rated 40-45MW

Source: www.sealegacy.com Oct. 4th, 2007

13

CO2 Emissions vs Power for LM6000

(idle)

14

0 5 10 15 20 25 30 35 40 45 50-20

-15

-10

-5

0

5

10

15

20

Power (MW)

Ram

p R

ate

(MW

/Min

)

LM6000 NOx Emissions as a Function of Power Level and Ramp Rate

0

0.2

0.4

0.6

0.8

1

1.2

Model Construction - Regression AnalysisN

Ox

Em

itted (lbs/min)

15

Region 4

Region 2

Region 1

Region 3

Idle Power

16

Siemens-Westinghouse Combined-Cycle Turbine – Rated 200 MW

www.summitvineyardllc.com

17

Operating Limit Constraint

Operating Limit Constraint

18

CO2 Emissions vs Power for SW 501FD

19

Siemens-Westinghouse - Regression Analysis

0 20 40 60 80 100 120 140 160 180 2000

0.5

1

1.5

2

2.5

3

3.5

Power (MW)

NO

x E

mitt

ed (l

b/m

in)

GE Document GER-3568G

Results

21

0 0.5 1 1.5 2 2.50

10

20

30

40

Time (hours)

Pow

er L

evel

s (M

W)

Wind + CT Operating Parameters

Ideal Fill PowerWind PowerActual Fill Power

0 0.5 1 1.5 2 2.5-20

-10

0

10

20

Time (hours)

Ram

p R

ate

(MW

/min

)

Ramp Rates

Results (1) – LM6000

22

Expected Emission Reductions Calculation

⎥⎥⎦

⎤

⎢⎢⎣

⎡

∗

−∗= +

nPenetratioWindMMM

NaturalGasTotal

GasNatuarlWindTotalral GasTotal,Natu

,

,100ReductionEmissionExpected

%1005.0)5.01(

100ReductionEmissionExpected,

=⎥⎥⎦

⎤

⎢⎢⎣

⎡

∗−

∗=NaturalGasTotal

ral GasTotal,Natu

MM

Example Calculations

If Wind Penetration is 0.5, then expect MTotal, Wind + NaturalGas = 0.5·MTotal,NaturalGas

If Wind Penetration is 0.3, then expect MTotal, Wind + NaturalGas = 0.7·MTotal,NaturalGas

%1003.0)7.01(

100ReductionEmissionExpected,

=⎥⎥⎦

⎤

⎢⎢⎣

⎡

∗−

∗=NaturalGasTotal

ral GasTotal,Natu

MM

23

Expected Emissions ReductionEastern Wind Farm

6,400 lbs1,500 lbs-240%± 250%

NOx

6,968 tons732 tons76%± 1%

CO2

501FD (NGCC, DLN)

1,595 tons176 tons80%± 1%

CO2

8,300 lbs290 lbs29%± 4%

NOxLM6000 (CT)

Mass Emitted by Wind + NG

Emissions Reduced

Expected Emissions Reduction

Turbine

24

Expected Emissions Reduction

-530% + 150% - 5%

-600% + 100% -75%

-240% ± 250%

NOx

78.9% ± 0.1%

76.8% ± .2%

76% ± 1%

CO2

501FD (NGCC, DLN)

77% ± 1%

77% ± 1%

80% ± 1%

CO2

31% ± 4%

21% ± 3%

29% ± 4%

NOxLM6000 (CT)

Central Great Plains

Southern Great Plains

EasternWind Farm

25

Generating Smaller Wind Data Sets

1000 min

(16.6 hrs)

+1 min

26

Implications• 1 MWh of wind energy does not eliminate 1 MWh of emissions• Impacts

– Clean Air Interstate Rule (CAIR)• Significant penetration of wind power will make it harder for

CAIR to achieve emission reduction goals– Emission displacement studies

• Overestimating the amount emissions are displaced by wind– Life Cycle Analyses

• Don’t account for wind’s effect of decreasing emission efficiencies of conventional generators

– Technology• Not all gas turbines are equally suitable for pairing with wind• R&D program to improve emissions of heavily cycled gas

turbines

27

AcknowledgementsAllen Robinson, Cliff Davidson, Lester Lave, Mitchell Small

Anonymous resource for power plant emissions data,

Anonymous resource for wind data

Funding CEICNETL CIT Dean’s Fellowship

Warren Katzensteinwarren@cmu.edu

Dr. Jay Aptapt@cmu.edu

Questions?

Renewable Portfolio Standard which would require

electric utilities to obtain 15 percent of their electricity

from wind, solar, or biomass energy by 2020

– NYTimes June 15th, 2007

RPS dsireusa.org C.A.I.R epa.gov

LA Smog Apt