epe trg 2005-06 del2

7/30/2019 EPE Trg 2005-06 Del2

1/52

Slovenian electricity marketSlovenian electricity market

openingopening -- lessons learnedlessons learned

by

Robert Golob, Ph.D.

EESTEC International Workshop

Power Engineering Today and Tomorrow

Ljubljana, Slovenia

11. December, 2003

7/30/2019 EPE Trg 2005-06 Del2

2/52

2

LectureLecture TopicsTopics

1. What is Deregulation All About?

2. Situation in Slovenia

3. GENCO in Deregulated Environment

4. Power Market Simulator

5. Illustration: GENCO Case Studies

7/30/2019 EPE Trg 2005-06 Del2

3/52

1.1. WhatWhat isis DeregulationDeregulation AllAllAboutAbout??

Regulated vs. Deregulated System

Official Goals of Deregulation

How Are This Goals Being Met?Deregulation & Prices

So What Have We Learned?

7/30/2019 EPE Trg 2005-06 Del2

4/52

4

WhatWhat isis DeregulationDeregulation AllAll AboutAbout??Regulated System

TRANSMITION SYSTEM

OPERATOR

(ELES)

PRODUCERS

DISTRIBUTION SYSTEM

OPERATOR(S)

CONSUMERS

Physical Power Flow

Money Flow

Vertically IntegratedPower System

7/30/2019 EPE Trg 2005-06 Del2

5/52

5

What is Deregulation All About?What is Deregulation All About?Deregulated System

TRANSMITION SYSTEM

OPERATOR

(ELES)

PRODUCERS

CONSUMERS

DISTRIBUTION SYSTEM

OPERATOR(S)

ORGANIZED MARKET

(BORZEN)

ELIGIBLE CONSUMERS TARIFF CONSUMERS

BROKERS

Physical Power Flow

Money Flow

Money Flow

TRANSMITION SYSTEM

OPERATOR DOES NOT

HANDLE ANY MONEY FLOW

7/30/2019 EPE Trg 2005-06 Del2

6/52

6

Getting rid of vertically integrated monopolies

Customer gets a right to choose

Price / tariff reduction and alignment

Entrance of new players and private capital

Improving the efficiency of operation

WhatWhat WeWe HaveHave LearnedLearned

Official Goals of Deregulation

7/30/2019 EPE Trg 2005-06 Del2

7/52

7

Deregulation:

(market) power has been transferred from transmission operatorsto generators

Liberalization:due to lack of transparency (liquidity), the mobility of consumersstays rather low

Privatization:large incumbents control the scene


How Are They Being Met?

Efficiency:revenues are increasing and costs have been (drastically)decreased

7/30/2019 EPE Trg 2005-06 Del2

8/52

8

What We Have Learned

Deregulation & PricesHighly dereglated

Poorly dereglated

Price decreasePrice increase

7/30/2019 EPE Trg 2005-06 Del2

9/52

9

No unique solution and no guarantee that it willwork

Yet there are some extremely positive examples

Liquidity of the markets transparent priceindicators are crucial

The process is still evolving (continental Europe)

Consolidation will mark the future outcomes


So What Have We Learned?

7/30/2019 EPE Trg 2005-06 Del2

10/52

2.2. SituationSituation inin SloveniaSlovenia

Situation in 1998

How Are This Goals Being Met?

Deregulation & PricesSo What Have We Learned?

7/30/2019 EPE Trg 2005-06 Del2

11/52

11

Liberalization what is it? Operation of sector with great losses (120 mill

in 1998)

Number of employees up to 40% technologicalsurpluses in individual companies

Electricity prices one of the lowest in Europeand not covering production costs

SituationSituation inin SloveniaSlovenia

Situation in 1998

7/30/2019 EPE Trg 2005-06 Del2

12/52

1.1. PowerPowerMarketMarket OverviewOverview

Different Market Designs

Day-Ahead Market

Auction Mechanism

7/30/2019 EPE Trg 2005-06 Del2

13/52

13

Two distinct trading mechanisms:

Central auction Bilateral trading

Central auction mechanism:

Simple central auctionmechanism(without economic dispatch) Highly complex auction mechanisms (based on unit commitment algorithms)

Auction types:

Double-sided auction demand side can be fully envolved

Single-sided auction onlysupplycan submitbids to the market

PowerPowerMarketMarket OverviewOverview

Different Market Designs

Transmission & ancillary services:

Electricity trading is done independetly of transmission and ancillary services The energy is traded with the transmission and ancillary service markets (usually

leads to nodal or zonal pricing market models)

7/30/2019 EPE Trg 2005-06 Del2

14/52

14

Hourly bids.

Central auction mechanism.

Double-sided auction.

24 simultaneous auctions for next day period.

Single Market Clearing Price (MCP) for all

participants.

Billateral trade is allowed.


Day-Ahead Market

7/30/2019 EPE Trg 2005-06 Del2

15/52

15


Auction Design

Supply bids

Q

P

Demand bids

Q

P

Market ClearingMechanism

Q

P

Auction Results

(prices & quantities)

Physical system

contrsaints

7/30/2019 EPE Trg 2005-06 Del2

16/52

16


Bids

All bids are Price / Quantitybids.

Supply bids:bids to sell energyincreasing or constant

Demand bids:

bids to buy energydecreasing or constant

Q

P

Q

P

7/30/2019 EPE Trg 2005-06 Del2

17/52

17


Market Clearing Mechanism

Quantity [MWh/h]

Price

[/M

Wh]

Aggregated

demand bids

Aggregated

supply bids

Sold / Bought

Quantity

(Traded Volume)

Marke

tClear

ing

Price

(MCP

)

Single price discovery

Supply meets demand

7/30/2019 EPE Trg 2005-06 Del2

18/52

2. GENCO2. GENCO OverviewOverview

GENCO Types

Bids & Bidding Strategies

SchedulingBillateral Trade

7/30/2019 EPE Trg 2005-06 Del2

19/52

19

Different types of GENCOs:

Thermal (coal-fired, gas-fired, oil-fired)Combined-cycle (CHP)

HydroPumped storage

Nuclear

Alternative sources

WindSolar...

Different types of strategies & market bids.

GENCOGENCO OverviewOverview

GENCOGENCO O iO i

7/30/2019 EPE Trg 2005-06 Del2

20/52

20

Passive bid (strategy) must-run bid:

Bid for selling electricity by any price.

Commonly used by hydro, nuclear and some types of

thermal producers.

GENCOGENCO OverviewOverviewBasic Bid

Quantity

Price

GENCOGENCO O iO er ie

7/30/2019 EPE Trg 2005-06 Del2

21/52

21

Thermal GENCO:

Marginal production cost bid:

GENCOGENCO OverviewOverviewBids & Bidding Strategies

Production cost curve:

C = a + b*Q + c*Q2

Quantity

Cost

Marginal production costs:

MC = = b + 2*c*QdC

dQ

Quantity

MarginalCosts

Marginal production

cost bid

Quantity

Price

GENCOGENCO O iOverview

7/30/2019 EPE Trg 2005-06 Del2

22/52

22

Thermal GENCO:

Incorporating technical constraints into bid Pmin & Pmax:


Modified marginal

production cost bids

Quantity

Price

Pmin Pmax

Possibi lity for Pminconstraint violation

Possibil ity for Pmaxconstraint violation

Quantity

Price

Pmin Pmax

Quantity

Price

Pmin PmaxIncorporated

must-run bid


7/30/2019 EPE Trg 2005-06 Del2

23/52

23

Hydro GENCO:

Popt is optimisation problem variable

No marginal production costs


Must-run bid

Quantity

Price

Popt

Popt is hydrology dependant

Must-run bid ensures all availableenergy is sold and no spillage occurs

Quantity

Price

Popt Pmax

If unithas water storagecapacity large enough it can

incorporate gaming into its bids:

7/30/2019 EPE Trg 2005-06 Del2

24/52

24

Combined GENCO:

It contains more than one unit (powerplant).

According to market rules it can submita single bid for all itscapacity.

After the auction onlyuniform price and combined quantity isknown.

In fully liberalised market environment there is no centraldispatching.

The problem of optimal unitcomitment (UC) and economicdispatch (ED) remains within combined GENCO.

Different approaches to UC and ED can be used:

Classical optimization and hydro-thermal coordination techniquesGenetic algorithms, evolutionary computation, ....


Scheduling


7/30/2019 EPE Trg 2005-06 Del2

25/52

25

Combined GENCO - example:

Contains several different type units:

GENCOGENCO OverviewOverviewScheduling

Quantity

Price

Pmax250 MW

Unit type Pmin [MW] Pmax [MW] MC [/MWh]

Hydro 0 50 0

30

60

Thermal

(coal-fired)70 130

Thermal

(gas-fired) 20 70

Submitted combined bid forsingle hour:

Sum of all available

capacity within GENCO

MCP

Final economic dispatchdone by GENCO:

Psold230 MW

Unit type P [MW]

Hydro 50

Thermal

(coal-fired)130

Thermal

(gas-fired)50

In this simple example the ED

is done by dispatching units

in MC merit order.

7/30/2019 EPE Trg 2005-06 Del2

26/52

3. GENCO in3. GENCO in DeregulatedDeregulatedEnvironmentEnvironment

General Overview

Power Market Simulator as Optimization Tool

Market Power

7/30/2019 EPE Trg 2005-06 Del2

27/52

27

Significant shift in optimization criteria:

GENCO inGENCO in DeregulatedDeregulated EnvironmentEnvironment

General Overview

REGULATED

Optimization criteria:operating costs minimization

DEREGULATED

Optimization criteria:profits maximization

The need for new

optimization approachesand tools.

Power Market

Simulator

7/30/2019 EPE Trg 2005-06 Del2

28/52

28

Bidding strategies optimization:

Power market simulator can be on of the tools used in optimalbidding strategy development for GENCO

With simulation of different scenarios and approaches the

optimalbidding strategy can be found:- by simulating GENCOs own different approaches and

tactics

- by modeling expected actions of the competition andreviewing impacts a suitable counter-action can be formed

- by forecasting different trends (market prices, scheduledplant decomissions or openings, ...)


Power Market Simulator as Optimization Tool

GENCO i D l d E i

7/30/2019 EPE Trg 2005-06 Del2

29/52

29


Market Power

MARKET POWER the ability of a single or a group ofparticipants to raise prices above competit ive levels.

Increases in

market prices

Increases in

profits

Due to the GENCOs focus on profitmaximization ahigh probability of market abuse exists.

The posibility (probability) of market abuse is usuallycorrelated with the GENCOs market power.

GENCO iGENCO i D l dD l t d E iE i t

7/30/2019 EPE Trg 2005-06 Del2

30/52

30


Market Power - Detection

=

=N

i

iSHHI

1

210000

Common indices:

Hirschmann-Herfindahl (HHI) Index:

Lerner Index:

Si market share of producer i

N number of producers in the system

( )%100

=

P

MCP

P market priceMC short-term marginal costs ofproduction

- Static index

- Gives the level of concentration of production

- Lacks the information about possibility ofmarket power abuse

- Much more dynamic

- Troubles with obtaining trueproduction

short-term marginal costs

All indices share the low ability to detectparticipants abusive behavior patterns.

GENCO iGENCO i D l t dD l t d E i tE i t

7/30/2019 EPE Trg 2005-06 Del2

31/52

31

Power Market Simulator:

Ability to simulate reference scenario.Possibilty of dynamic observations of prices, market shares,quantities, profits...

Ability to forecastmarket trends and impacts of differentevents(planned plant maintainance outage, plantdecomission or opening,...)

Simulation tool can be extremely efficient in detecting possible

participants abusive behavior patterns.


Market Power - Detection

Power market simulator already used as a monitoring tool at

Agency for energy a market monitoring body in Slovenia.

7/30/2019 EPE Trg 2005-06 Del2

32/52

4.4. PowerPowerMarket SimulatorMarket Simulator

General Overview

Technical & Financial Parameters

Bids

Bidding Strategies

Bilateral Contracts

Clearing Mechanism

Results

PP M k t Si l tM k t Si l t

7/30/2019 EPE Trg 2005-06 Del2

33/52

33

SUPPLY

PowerPowerMarket SimulatorMarket Simulator

General Overview Modell ing of Day-Ahead Market

Technical & Financial

Parameters

Bids

Bidding Strategies

DEMAND

Technical & Financial

Parameters

Bids

Bidding Strategies

Clearing Mechanism

System Parameters

Results:Quantities & Prices

Bilateral Contracts Bilateral Contracts

Billateral contracts

are alsoconsidered.

PPower M k tMarket Si l tSimulator

7/30/2019 EPE Trg 2005-06 Del2

34/52

34

SUPPLY:

Installed Capacity [MW]

Start-up Costs (thermal units)

Operating Costs (thermal units)

Hydrology reservoir inflow (hydro units)

DEMAND:

Load curve

Load forecast (short & long-term)

PowerPowerMarketMarket SimulatorSimulator

Technical & Financial Parameters

PowerPower MarketMarket SimulatorSimulator

7/30/2019 EPE Trg 2005-06 Del2

35/52

35

Bids are made of blocks of quantities and prices.


Bids

SUPPLY bid:(bid for sale)

DEMAND bid:(bid for purchase)


7/30/2019 EPE Trg 2005-06 Del2

36/52

36

Bidding strategy is basically a series of appropriate one-hour bids for

corresponding number of simultaneous auctions.


Bidding Strategies


7/30/2019 EPE Trg 2005-06 Del2

37/52

37

Bilateral contracts are entered in the form of their load over time.

The actual negotiated prices in bilateral contracts are unknown.


Bilateral Contracts


7/30/2019 EPE Trg 2005-06 Del2

38/52

38

Clearing mechanismmimics the actual clearing process on day-aheadpower market.

For each trading hour the Market Clearing Price (MCP) and quantitiesof bought/sold electricity are determined according to market rules.


Clearing Mechanism


7/30/2019 EPE Trg 2005-06 Del2

39/52

39

Price levels:


ResultsQuantities:

Profits:

Market shares:

7/30/2019 EPE Trg 2005-06 Del2

40/52

5.5. CaseCase StudiesStudies

Abusive Behavior Illustration

Scheduled Outages Planning

Pumped-storage Plant Operation

CaseCase StudiesStudies

7/30/2019 EPE Trg 2005-06 Del2

41/52

41

Test systemcharacteristics:


Abusive Behavior IllustrationProducer Unit

Installed capacity

[MW]

Units marginal cost

[c/kWh]

1 300 1.74

2 50 3.041 200 2.392 150 2.833 50 5.221 300 1.522 100 5.65

3 50 6.96

C

B

A

500

600

700

800

900

1000

1100

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 (h)

Loa

d(MW

)

Test system24-hour inelasticdemand diagram:


7/30/2019 EPE Trg 2005-06 Del2

42/52

42

DemandQuantity

Price

MCP

Extra profit for

producer A

Example of possible price manipulation by producer A inparticular trading hour:


Abusive Behavior Illustration Witholding Strategy

MCP

DemandQuantity

Price

Producer A plays

unilateral witholding

strategy

Producer A

Costs

Profit

Producer B

CostsProfit

Producer C

CostsProfit

Producers A

unit 2 to be

witheld from

market

Price mark-up in

particular trading hour


7/30/2019 EPE Trg 2005-06 Del2

43/52

43


Abusive Behavior Illustration Witholding Strategy

Producer A Producer B Producer C

Reference scenario

A view fromthe simulator - 6th hour of the 24-hour period:

Price mark-up in

particular trading hour

Producer A is witholding

Producers A uni t

2 is to be witheld


7/30/2019 EPE Trg 2005-06 Del2

44/52

44

0

50

100

150

200

Producer A Producer B Producer C

Pro

fit[*1000]

Reference scenario Producer A withholding unit 2

Results:

CaseCase StudiesStudiesAbusive Behavior Illustration Witholding Strategy

2

3

4

5

6

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 [h]

MCP[c/kWh]

Reference scenario Producer A withholding unit 2

MCP levels duringobserved period.

Raised MCP levels as a

result of Producer Aabusive behavior.

Raised profit levels as a

result of Producer Aabusive behavior.


7/30/2019 EPE Trg 2005-06 Del2

45/52

45

Results fromPower Market Simulator:

CaseCase StudiesStudiesAbusive Behavior Illustration Witholding Strategy

Rised MCP levels as a result

of Producer A abusive

behavior.

Reference scenario

Producer A is witholding


7/30/2019 EPE Trg 2005-06 Del2

46/52

46

Test system:

Supply:



Producer type No. of producersCombined installed capacity

[MW]

Thermal (coal-fired) 7

11

Hydro 3 81212

Thermal (gas-fired)

857

386337Nuclear

2392Total:

Demand:Peak demand: 1891 MW

Base load: 692 MWAnnual consumption: 11.3 TWh

Neighbouring systems:2 neighbouring systems

Maximumtransfer capacity: 300 MW

Simulated period:

1 year

Nuclear producer is

planning a 1 month

maintenance outage.


7/30/2019 EPE Trg 2005-06 Del2

47/52

47



Changes in prices:

Moderate price spikes as a

result of outage in May.

Daily average prices

No scheduled outage


Scheduled outage in May

Av. price

levels

No outage

[c/kWh]

Outage

[c/kWh]

In May 2.40

3.78

3.87 (+61%)

Yearly 3.91 (+3%)

Outage planned inMay


7/30/2019 EPE Trg 2005-06 Del2

48/52

48

CaseCase StudiesStudiesScheduled Outages Planning Uncoordinated Planning

Changes in prices:


No scheduled outage


Scheduled outage in August

Av. price

levels

No

outage

[c/kWh]

Outage

[c/kWh]

In Aug. 2.63

3.78

7.47 (+184%)

Yearly 4.20 (+11%)

Outage planned inAugust

Special event: in August also one ofthe biggest thermal power plants

consideres 1-month scheduled

outage.!


7/30/2019 EPE Trg 2005-06 Del2

49/52

49


Pumped-storage Plant OperationPumped-storage plant characteristics:

- Maximumproduction capacity: 50 MW

- Maximumconsumption capacity: 50 MW- Upper reservoir capacity: unlimited

Minimal di fference between sell price

and buy price needed to cover the

energy transformation losses.

Bids submitted each trading hour:

Quantity Quantity

Price

Price

Buy bid Sell bid


7/30/2019 EPE Trg 2005-06 Del2

50/52

50

CaseCase StudiesStud es

Pumped-storage Plant OperationResulting price levels during observed period:

During work-day peak hours the

electricity prices are high.During work-day night hours and

during the weekend the electricity

prices are lower.

Pumped-storageplant is taking

advantage of suchperiodic pricefluctuations.


7/30/2019 EPE Trg 2005-06 Del2

51/52

51

Case Studies

Pumped-storage Plant OperationSimulation results (1 week period):

Bought hourly quantities: Sold hourly quantit ies:

During work-day peak hours theenergy is produced and sold to the

market.

During work-day night hours and

during the weekend the energy is

consumed for pumping water into

upper reservoir.

7/30/2019 EPE Trg 2005-06 Del2

52/52

THANK YOU FOR YOUR

ATTENTION

epe trg 2005-06 del2

Documents