1 PJM/MISO Update on CO- 114 Frank Koza – PJM Dave Zwergel – Midwest ISO
Feb 11, 2016
1
PJM/MISO Update on CO-114
Frank Koza – PJMDave Zwergel – Midwest ISO
2
CO 114 Behavior• TLR results on PJM Coordinated Flowgates
have been questionable since AEP and DPL joined PJM
• Calculations produce “unrealistically excessive flow” across all priorities (7-Firm, 6-Non-Firm Network, and 2-Hourly)
• PJM flows have often increased tenfold• Results add uncertainty to operations
– How much is relief should I ask for? – How much is relief will I get?
3
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
Examples
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
t Im
pact
ED2
ED6
NNL
101 Kammer #8 xfmr l/o Kammer-South Canton 765 kV line
Flowgate Limit is 3695
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
4
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
Examples
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time
Meg
awat
t Im
pact
ED2
ED6
NNL
1719 Mt. Storm-Doubs 500/Mt. Storm-Meadow Brook 500
Flowgate Limit is 2598
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
5
Examples
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
2337 Cook-Palisades345/BentnHrbr-Palisades345
Flowgate Limit is 2094
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
t Im
pact
ED2
ED6
NNL
6
Results and Consequences
RCs may have to request much more relief than is actually necessary to effect curtailments
RCs may believe flows represent available relief, when in fact, the flows are reported erroneously and no relief is available
Operator Flow Change Request: -1500 MW
7
Analysis• Since the RTO is responsible for calculating Market
Flow, PJM and MISO jointly investigated the discrepancies to find the cause
• Assumptions hid an inequality in the basic calculation mechanism– Testing was done with generators at the same output levels, but
shifts in output levels would have demonstrated the inequality• Load Shift Factor aggregations must be consistent
between pre- and post-expansion models• Since CE joined PJM as a separate Control Area,
problem remained hidden until AEP and DPL joined and the LSFs were merged
8
Historic Footprint
ControlArea 1
ControlArea 2
Gen 11100 MW Output
GSF = 0.10
Gen 2900 MW Output
GSF = -0.20
Load 11000 MW Demand
LSF = -0.20
Load 21000 MW Demand
LSF = -0.30
Flowgate X
Gen Out(Scaled to meet
Load) GSF LSF GLDF Impact on Flowgate
Control Area 1 1000 0.10 -0.20 0.30 300
Control Area 2 900 -0.20 -0.30 0.10 90
390 NNLPTP Tag would be 100MW * (.10 – (-.20)) = 30MW
9
New Control Area
Merged
ControlZone 1
ControlZone 2
Gen 11100 MW Output
GSF = 0.10
Gen 2900 MW Output
GSF = -0.20
Load 11000 MW Demand
LSF = -0.20
Load 21000 MW Demand
LSF = -0.30
Flowgate X
Gen Out(Raw,
Unscaled) GSF LSF GLDF Impact on Flowgate
Control Zone 1 1100 0.10 -0.25 0.35 385
Control Zone 2 900 -0.20 -0.25 0.05 45
430 MFNo Point to Point Tag (becomes internalized)
10
Comparison
Gen Out GSF LSF GLDF Impact on Flowgate
Control Area 1 1000 0.10 -0.20 0.30 300
Control Area 2 900 -0.20 -0.30 0.10 90
30
420 NNL
Gen Out GSF LSF GLDF Impact on Flowgate
Control Zone 1 1100 0.10 -0.25 0.35 385
Control Zone 2 900 -0.20 -0.25 0.05 45
430 MF
Plus Point to Point Impact
11
Solution• Considered multiple solutions
– Threshold change– Netting– Partial netting
• Best Solution Found– RTOs use load shift factors similar to those used in the firm
usage calculation• Control Zone impacts will be determined as if Historic CAs remain in
place (GLDF = GSF – Historic LSF)• Inter-Zone Transfer Impacts will be determined as if Historic CAs
remain in place (Xfer TDF = Historic TDF – Historic TDF)• Generation in a Zone in excess of Zone Load will be considered
transfer MW (but sum of all Zonal Gen + Transfer Gen will not exceed RTO Load)
• Market flow values will regain accuracy consistent with firm usage
12
New Control Area
Solution Step 1
ControlArea 1
ControlArea 2
Gen 11100 MW Output
GSF = 0.10
Gen 2900 MW Output
GSF = -0.20
Load 11000 MW Demand
LSF = -0.20
Load 21000 MW Demand
LSF = -0.30
Flowgate X
Gen Out (Min of Gen or Load) GSF LSF GLDF Impact on Flowgate
Control Area 1 1000 0.10 -0.20 0.30 300
Control Area 2 900 -0.20 -0.30 0.10 90
390 MFCalculate GTL Impacts as if Historic Footprint still existed
13
New Control Area
Solution Step 2
ControlArea 1
ControlArea 2
Gen 11100 MW Output
GSF = 0.10
Gen 2900 MW Output
GSF = -0.20
Load 11000 MW Demand
LSF = -0.20
Load 21000 MW Demand
LSF = -0.30
Flowgate X
Excess Gen TDF 1 TDF 2 PTP TDF Impact on Flowgate
1 to 2 Transfer 100 0.10 -0.20 0.30 30
30 MFCalculate PTP Impacts of Transferring Excess MW
14
Comparison
Gen Out GSF LSF GLDF Impact on Flowgate
Control Area 1 1000 0.10 -0.20 0.30 300
Control Area 2 900 -0.20 -0.30 0.10 90
30
420 NNL
Gen Out GSF LSF GLDF Impact on Flowgate
Control Zone 1 1000 0.10 -0.20 0.30 300
Control Zone 2 900 -0.20 -0.30 0.10 90
1 to 2 Transfer 100 0.10 -0.20 0.30 30
420 MF
Plus Point to Point Impact
15
New vs. Old Process
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
t Im
pact
ED2
ED6
NNL
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
101 Kammer #8 xfmr l/o Kammer-South Canton 765 kV line
Moderate Increase in Flows
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
New Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
16
New vs. Old Process
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time
Meg
awat
t Im
pact ED2
ED6
NNL
New Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time
Meg
awat
ts Im
pact ED2
ED6
NNL
1719 Mt. Storm-Doubs 500/Mt. Storm-Meadow Brook 500
Mild Increase in FlowsSome Firm became Non-Firm
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
17
Current Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
t Im
pact
ED2
ED6
NNL
New vs. Old Process
Pre Control Area Consolidation
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
2337 Cook-Palisades345/BentnHrbr-Palisades345
Moderate Increase in Flows
PJM Impact Based on Distribution Factors calculated with PSS/e MUST and Generator Output from Jan 5, 2005
New Process
0
200
400
600
800
1000
1200
1400
1600
1800
2000
1 3 5 7 9 11 13 15 17 19 21 23
Time
Meg
awat
ts Im
pact
ED2
ED6
NNL
18
Questions?
19
ReferenceMathematical Explanation of Inequality
Assume 2 Control Areas A and B. Each Control Area has a single Generators G and a single Load L. Each Generator and the Aggregate Load has a shift factor S. Impact I on a flowgate for a generator is calculated as follows:
GSSI LGG )( The impacts of multiple Control Areas can be summed to determine the impacts of both Control Areas.
BAAB III Now assume the Control Areas merge to create Control Area C. Create an Aggregate Load Shift Factor for the new Control Area SLC as follows:
C
BLB
C
ALALC L
LS
LL
SS
Impact I on a flowgate for a generator in the merged Control Area is calculated as follows:
GSSI LCGG )( The impacts of the two generators can be summed to determine the impact of the new Control Area:
BA GGC III Hypothesis: CAB II
CBA III
CBLBGBALAGA IGSSGSS ))(())((
BA GGBLBGBALAGA IIGSSGSS ))(())((
))(())(())(())(( BLCGBALCGABLBGBALAGA GSSGSSGSSGSS
BLCBGBALCAGABLBBGBALAAGA GSGSGSGSGSGSGSGS
C
BLB
C
ALALC L
LSLLSS
BC
BLB
C
ALABGBA
C
BLB
C
ALAAGA
BLBBGBALAAGA
GLL
SLL
SGSGLL
SLL
SGS
GSGSGSGS
BC
BLB
C
ALAA
C
BLB
C
ALABLBALA G
LL
SLL
SGLL
SLL
SGSGS
C
BBLB
C
BALA
C
ABLB
C
AALABLBALA L
GLSL
GLSL
GLSL
GLSGSGS
BBLBBALAABLBAALABCLBACLA GLSGLSGLSGLSGLSGLS
BBLBBALAABLBAALABBLBBALBABLAAALA GLSGLSGLSGLSGLSGLSGLSGLS
ABLBBBLBBALABBLBBALBABLA GLSGLSGLSGLSGLSGLS
ABLAABLBBBLBBALABBLBBALB GLSGLSGLSGLSGLSGLS
)()( BLABLBABLBALABLBALBB LSLSGLSLSLSLSG
)5653(7)53465343(2
)3015(7)15241512(2
)45(7)18(2
31536
CAB II