Power Market Participation of Flexible Loads and Reactive Power - - PowerPoint PPT Presentation

▶
power market participation of flexible
SMART_READER_LITE
LIVE PREVIEW

Power Market Participation of Flexible Loads and Reactive Power - - PowerPoint PPT Presentation

Aug 19, 2022 216 likes •549 views

Power Market Participation of Flexible Loads and Reactive Power Providers: Real Power, Reactive Power, and Regulation Reserve Capacity Pricing at T&D Networks DIMACS, Rutgers U January 21, 2013 Michael Caramanis mcaraman@bu.edu Outline

slide-1
SLIDE 1

Power Market Participation of Flexible Loads and Reactive Power Providers: Real Power, Reactive Power, and Regulation Reserve Capacity Pricing at T&D Networks

DIMACS, Rutgers U January 21, 2013 Michael Caramanis mcaraman@bu.edu

slide-2
SLIDE 2

2

Outline

  • How can Flexible Loads Provide Fast Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-3
SLIDE 3

3

Outline

  • How can Flexible Loads Provide Fast Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-4
SLIDE 4

4

Characteristic Generation Demand Dispatchability- Schedulability Low/Med/High Wind, Run of Riv /Neuclear, L.E.P,/ HydroFossil Capacity Loads, dependent on

  • env. e.g.,Light/ Ind. Energy

Loads Aluminum idle/Schedulable production of

  • electr. energy intensive storable

products (gas liquif.) Flexibility Low/med/high No Ramp – steady output e.g., nucl, min gen, start up cost and delay/ Inertia and medium storage/high ramp- low inertia large storage Thermal or work inertia (Allum. Smelter)/Enegy Demand with small storage to capacity ratio (HVAC)/ Large storage to capacity ratio (ice, molten salt, batteries in Evs) Forecastaility Low/Med/High Wind, Solar. RoR Hydro/reliable fossil/unreliable fossil Inflexible loads (lighting cooking)/Weather dependent/scheduled loads Voltage Control Synchronous Generators with dynamic Var compensators, DC-AC Converters Distributed Power Electronics accompanying EVs. HVAC, Roof top PV.

Generation and Demand Share Functional Characteristics that are Key to the Efficient and reliable Operation of the Electricity Grid

slide-5
SLIDE 5

5

  • 0.8
  • 0.6
  • 0.4
  • 0.2

0.2 0.4 0.6 0.8 1 D: 10/1/11 S: 9/1/10

Instance of PJM Regulation Signal, y(t). Note Constant Average over relatively short period of Time

Flexible Loads Require Energy by some deadline => Capable of Regulation Reserves

slide-6
SLIDE 6

6

Example of Generator providing Super Fast Reserves: Frequency control and  40MW of Secondary Reserves

Source: Courtesy of EnThes Inc., March 2007

Today Generating Units are Only Reserve Providers

Frequency Control Secondary Reserves 320MW50MW

slide-7
SLIDE 7

7

Outline

  • How can Flexible Loads Provide Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-8
SLIDE 8

Distribution Network Low Voltage Bus Marginal Cost Based Dynamic Prices (DLMP) Result from Augmenting Transmission System High Voltage Prices (LMP) by Marginal cost of: Line Losses, Reactive Power, Voltage control, Transformer Life Loss

8

HV, Bus n LV, n(k)

LMP at Bus n

DLMP at n(i)=mn(i)(LMP at n)+…Where mn(i)=(1+ML at n(i))… LV, n(i)

slide-9
SLIDE 9

9

Outline

  • How can Flexible Loads Provide Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-10
SLIDE 10

Examples of Flexible Loads: State Dynamics Determine Preferences

10

  • Distributed PHEV Charging
  • Centralized Pumped Storage Hydro Units

1 ( ) ( ) ( ) ( ) deptime ( )

ˆ

j j j j j

F F F F F

t t t t n i n i n i n i n i

x x d x 

   

1 ( ) ( ) ( ) ( ) ( ) 2 ( ) ( ) ( ) ) 4 ( ( ) psh psh t g t r R t n p n p n p ps t t p n p n p n h p n p n p p n p p h n s

x x p g g x x   

    

slide-11
SLIDE 11

11

Strategic Flexible PHEV Load Behavior

   

, ( ) ( ) ( )

,

( ) ( ) , , , , , , , ( ) ( ) ( ) ( ) 1 ( ) ( ) ( ) ( ) ( ) ( )

{ [ λ λ ] ( )} . . ˆ e.g., state dyn of EV dem.

min

j j

j E t R t t F F R n n n i t t j j n i n i j j j j j j j

F F F F F F F F R

F t E t t n i n n i m j t d d j t F R t R t t t t n i n n i n i n i t t t t n i n i n i n i t t n i n i

E m d m d U x s t x x d d d 

 

     

λ λ

,

( ) ( ) ( )

up/dn nature of Reg. Res. ˆ [ ] Local Constraint

j j

F F R

t t t n i n i n i j

d d C  

slide-12
SLIDE 12

Use of Current Bidding Rules to Self Schedule

12

 

, * * ( ) ( ) , * ( ) ( )

Bid Energy at a very high price Bid Energy 2 at energy price ~ λ and Regulation Service Rate at 0.

j j j

F F R t t n i n i F R t t E t n i n i n

d d d m 

slide-13
SLIDE 13

Using the Current Bidding Rules. Bids described on the previous slide, induce the ISO/DSO to almost surely Schedule Energy and Reserves to the * values, and thus effectively self dispatch.

13

( ) ( ) ( )

* ( ) ( ) ( ) ( ) ( ), ( ) , ( ) ( )

( ) ( ) ( ) ( ) ( ) ( ) , , , , , ( ) * 2 ( ) , * ( ) ( ) ( ) , ( )

. . 0,

max ( ) 2

(

)

j

c t t R t n i n n j j

c c R F t t c t n n i n i n n i j n i c t E u t n i n n i

t t t t t t n i n i n n n n d g g t t i F n i t n i j F R R t t n n i n j n i

s t g d d d t

u d c g r g d g d

 

 

      

    

     

   

   

,

( ) ( )

& other capacity and ramp constr. for conv. gen. and dem.

R u t n

c t n i n i

t

Loss R

  

 

slide-14
SLIDE 14

14

Outline

  • How can Flexible Loads Provide Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-15
SLIDE 15

Hierarchical Game Dynamics

  • Undamped Oscillations when Flex Load

Updates Clearing Price Estimates Myopically to Most Recent ex-post ISO/DMO value

  • Convergence to stable equilibrium when

Flex Load Updates Clearing Price Estimates Factoring in History, for example sets them Equal to their Time Average

15

slide-16
SLIDE 16

16

0.00% 1.00% 2.00% 3.00% 4.00% 5.00%

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 129 133 137 141 145 149 153 157 161 165 169 173

% Convergence Iteration

UPQB LMP Convergence by Iteration (No Congestion)

Base Case Discrete Smooth Case Quadratic Case

slide-17
SLIDE 17

17

0.00% 1.00% 2.00% 3.00% 4.00% 5.00%

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 129 133 137 141 145 149 153 157 161 165 169 173

% Convergence Iteration

UPQB LMP Convergence by Iteration (With Congestion)

Base Case Bus 1 Discrete Smooth Case Bus 1 Quadratic Case Bus 1

slide-18
SLIDE 18

18

0.00% 0.50% 1.00% 1.50% 2.00% 2.50% 3.00%

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 129 133 137 141 145 149 153 157 161 165 169 173

% Convergence Iteration

Step Size Impact on UPQB Convergence

Discrete Smooth Case Bus 1 (Steepest Descent) Discrete Smooth Case Bus 1 (Averaging) Quadratic Case Bus 1 (Steepest Descent) Quadratic Case Bus 1 (Averaging)

slide-19
SLIDE 19

19

(0.50) (0.40) (0.30) (0.20) (0.10)

  • 0.10

0.20

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

$/MWh Hour

(UPQB LMP - TCCB LMP) by Iteration

Bus 1 Iterations 148, 151, 154, etc Bus 1 Iterations 149, 152, 155, etc Bus 1 Iterations 150, 153, 156, etc

slide-20
SLIDE 20

20

Outline

  • How can Flexible Loads Provide Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-21
SLIDE 21

Under New Bidding Rule allowing Flex Load to Express True Utility, ISO/DNO will Solve

21

, ( ) ( ) ( ) ( ) ( ) ( )

( )

( ) ( ) , , , , , , , , , , ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ), ( ) ( ) ( )

[ ( )] . . ( ) 2

max

F F R c t t R t t t W t j j n i n n n i n i n j j j j

c c F F R c t n i

t t n i n i d g g d d g t i t t t t t t t n n n n n i n i F t t c t n n i n i n n i j n i F n i t n i j

d

u d c g r g U x s t g d d d

  

         

     

    

( ) ( ) ( )

2 , ( ) ( ) ( ) ( ) ( ), ( ) ( ) 2 , ( ) 1 ( ) ( ) ( ) ( )

0, ( ) 0, 2 ˆ ( ) Flex Load Dynamics & other Local ( ) and System constrai

j j j j j j

n i c t n i n i F F F F

E x t n F t t c t n n i n i n n i j n i F n i t E x t n i n j t t t t n i n i n i n i

d

t g d d d t x x d n i n i

 

   

             

 

   

nts

slide-22
SLIDE 22

Complex Bid ISO/DNO Market Clearing Achieves Hierarchical Game Equilibrium

22

  • Theorem:

– First order Optimality Conditions – Complementary Slackness, and – Feasibility Conditions Coincide if we combine Hierarchical game problems and compare to ISO/DSO problem, Except when Flex loads dominate in a Distr. Location (competitive assumption fails?)

( )

( ) ( ) , , , ( ) ( ) ( ) ( ) ) ( ) , (

£ 1

( )

( )

j

t n Fj i j

F

E t ISO n d R c t t t t j t j t n n i n i n i t n i n i n i j F R t n i n i

d d d

               

slide-23
SLIDE 23

Additional term in ISO/DNO problem

23

( )

( ) ( ) , , , ( ) ( ) ( ) ( ) ) ( ) , (

£ 1

( )

( )

j

t n Fj i j

F

E t ISO n d R c t t t t j t j t n n i n i n i t n i n i n i j F R t n i n i

d d d

               

( ) ( ) ( )

,

Becomes negligible, i.e. 0 as

  • r as the relative size of flex load Reg. Res. Transactions

j j

t n i t c t n i n i

F R F

d d d

   

slide-24
SLIDE 24

24 (250) (200) (150) (100) (50)

  • 50

100 (0.40) (0.30) (0.20) (0.10)

  • 0.10

0.20

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

$/MWh $/MWh Hour

Impact of Competitiveness Assumption

UPQB Price Spread - TCCB Price Spread UPQB IFL Costs - TCCB IFL Costs

slide-25
SLIDE 25

Conclusion

  • Flexible Loads at Distribution Level may

participate in Expanded ISO/DNO Centrally Cleared Power Market bringing significant benefits, particularly w.r.t. Sustainable Renewable Generation Integration to the Grid

  • Expanded ISO/DNO-Operated Power Market

Clearing is Practical from Information and Computational Tractability Point of view.

  • Inclusion of Other Important Distribution Network

Costs, such as Reactive Power Compensation and Voltage Control is also Practical.

25

slide-26
SLIDE 26

26

Outline

  • How can Flexible Loads Provide Reserves
  • How do Distribution Network Injections Differ From

Transmission System Bus Injections?

  • Current Market Bidding Rules Motivate Flexible

Distributed Loads to Exercise Strategic Behavior Resulting in a Hierarchical Game

  • Conditions for Hierarchical Game to Converge
  • Revised Bidding Rules Remove Strategic Behavior

Incentives and Allow ISO/“DNO” to Clear Market in a Socially Optimal Manner

  • Detailed Distribution Market Pricing Real and

Reactive Power

slide-27
SLIDE 27
  • Ex. of Var. Speed HVAC - PV Collaboration:

Action in the small by Distr. Flex. Loads

27

slide-28
SLIDE 28

Load and Other Resources May Participate fully in Future Distribution Markets:

28

Centralized Dispatchable Generation Distributed Consumers Centralized Non-Dispatchable Generation Distributed Generation e.g., Wind Parks Centralized Prosumers (e.g., cogeneration) Distributed Prosumers

Current market: Only Centralized generation is a full market participant. All others communicate their capabilities and needs without feedback and response

Market Centralized Dispatchable Generation Distributed Consumers Centralized Non-Dispatchable Generation Distributed Generation e.g. Wind Parks Centralized Prosumers(e.g., cogeneration) Distributed Prosumers

Future Market: Many more non-dispatchable cedtralized generators, distributed generators and prosumers. On the distributed side, “feedback” renders Non-Dispatchable generation and distributed consumers- producers (prosumers) full market participants

Market

slide-29
SLIDE 29

29

Example of a Radial Distribution Network: One Medium Voltage Branch is Shown with three feeders, each with three building loads. Substation is the Slack Bus

slide-30
SLIDE 30

Distribution Market Problem formulation:

Minimize Utility Loss, Real and React. Power Cost (incl Losses), Asset Life Loss, and Volt. Control Cost s.t. Load Flow , Capac., Volt. Magnitude Const

 

 

, , , ,

2 2 2 , , (1) , (1)

1

m n m n i i i i M M m n m n

f f g d d P P P V g m m m m b b m i i f f

tM c P u P P C C Q c V   

       

              

   

Minimize Subject to

30

slide-31
SLIDE 31

Distribution Market Problem formulation (cont.)

31

, , , , ,

, 2 1, 2, , ,

15000 15000 exp , 383 273 , ,

b m b m b m b m b m

f b m H f H A f f f b m b m b b b

f k k S f V V V b A   

                    

slide-32
SLIDE 32

Distribution Market Benefits

  • Marginal Losses Reflected in DLMPs=>Demand

Adaptation

  • Reactive Power Pricing motivates Dual Use of

Power Electronics whose presence is expected to Become Ubiquitous while accompanying Distributed Clean Generation (e.g., PV) installations and Flexible Loads (e.g., EVs, Heat Pumps)

  • Marginal Voltage Control Cost Reflected in DLMPs

=>Demand Adaptation

  • Distribution Asset Degradation Marginal Costs

Reflected in DLMPs =>Demand Adaptation

  • Full Distr. Net Price Unbundling =>Distr. Net Rent

32