Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

IITMandi Bid

Download as pdf or txt
Download as pdf or txt
You are on page 1of 27

Bidding Strategies of

Renewable Energy Sources

S. N. Singh

Department of Electrical Engineering


Indian Institute of Technology Kanpur
Kanpur-India
Introduction
Future power generation will not be limited to
the conventional power plants.
Wind power (off-shore and on-shore ) and
Solar power are potential candidates in non-
conventional power generation.
Large scale integration in the electricity
system presents planning and operational
difficulties due to mainly the intermittent and
difficult to predict the availability.
Percentage of RES is increasing due to
technology and efficiency improvements,
government financial support and energy
policies.
Introduction
Economic analysis often reveals that wind /solar
power is competitive with conventional power.
But such analysis strips out distortions such as input
fuel subsidies, taxes and cross-subsidies between
different class of consumers and uses the opportunity
cost of capital as the discount rate.
The RES may not be competitive to the conventional
energy sources.
The energy production cost issue becomes more
challenging due to ongoing restructuring process
motivated to provide competition.
Introduction
Wind generation is accommodated in day-ahead and
hour-ahead energy markets without imbalance penalties
In some electricity market, wind/DER generators are not
allowed to bid and they are taken into the system as and
when these powers are available.
This talk analyses and proposes the pricing mechanism
for RES power integrated into the electricity market.
Several other considerations such as market collusion,
ancillary services and market power should be taken.
It could be guideline for the policy makers and market
operators to promote the wind power with system
reliability and security.
Bidding Strategies
Theoretically, in perfectly competitive market,
suppliers should bid at, or very close to, their marginal
production costs to maximize returns.
However, the electricity market is not perfectly
competitive, and power suppliers may seek to benefit
by bidding a price higher than marginal production
cost.
Each suppliers objective is to maximize benefit,
therefore, given its own costs and constraints and its
anticipation of rival and market behavior by
constructing its offer (bid) price.
Market Clearing Process
Single Price Market Clearing Process
Pay-as-Bid Market Clearing Process
Example-1
Let genenerator-1 has capacity of 100 MW with production cost
$75/MW per unit time. Demand is 100 MW. Bid of other gen at the
same bus is
50 MW 101 $/MW
50 MW 121 $/MW

What should be bid of gen-1?


Option-1: bid at 100 $/MW
win all demand of 100 MW. The profit will be 100(100-75)= 2500

Option-2: bid at 120 $MW


win 50 MW capacity. The profit will be 50(120-75)= 2250
This shows the option-1 is preferred.
If second block of other gen. is 131 $/MW. Profit at bid price of 130 will
be 50(130-75) = 2750.
Now this is preferred.
Bidding Strategies
Deterministic formulation
Assumptions
uncertainty are not included
No temporalities
All rivals are clubbed together
Rival:
Block: 1, 2, ... , J MW
Price : 1, 2, ... , J $/MW (j+1 > j ; j)
Seller:
Block: H1, H2, ... , HI MW
Cost : c1, c2, ... , cI $/MW (ci+1 > ci ; i)
Bids : p1, p2, ... , pI $/MW
Seller will maximize it profit
Bidding Strategies
Seller will maximize
H i ( pi ci )vi
iI

vi is 1 if dispatched, otherwise zero.


( pi > j )and ( pi < j +1 ), and
1
iff j i 1
vi = k + H k' D
k =1 k '=1

0 otherwise

Seller maximize
H i i ( zi )vi
iI

where i is preference function and zi =pi -ci


Preference function
it depends on the operational and economic reasons
y= xn
n=1 ---> Fig b, n> 1 ---> fig c and n <1 ---. Fig a
c b


Fig c: greater desire to sell even at low gain

Fig a: greater desire to sell even at high gain

Fig b: constant sell

Uncertainty and Temporality


Z (Gain)
uncertainty of bid price of rivals
uncertainty of demand
Gi(pi) is a distribution function which denotes probability that block i will be
sold if it waw offered at price pi

H i i ( z i )Gi ( pi )
iI
supplier must forecast the rivals bid price on previous
data.
Temporality is due to consumer response with price
change.
Example: Supplier has two unit block as following cost

H1 =400 MW C1 = $50/MWh
H2 =200 MW C2 = $65/MWh

Let preference function be linear.

Rivals bids, If know


Block : 400, 400, 200, 200 MW
Price : 56, 66, 71, 76 $/MWh

Demand is 1000 MW
Objective Function

400( p1 50)v1 + 200( p 2 50)v 2

( pi > j )and ( pi < j +1 ), and


1
iff j i 1
vi = k + k'H D
k =1 k '=1

0 otherwise

Options are

P1= 55, P2= 65


P1= 65, P2= 65
P1=70
Bidding in a constrained network
Transmission constraints restricts the flow of power from
low cost node to high value nodes.
Constraints on the system will cause different prices at
different nodes.
Revenue collected from the consumer will be more than the
money paid to the generators.
Unsubscribed revenue can be given to transmission right
holders.
Cost
pd

ps
D
G1= 10 $, G2=12 $
Pflowlimit = 100 MW,
Load is 120 MW qcap
quantity
Bidding Strategies
Linear and Block Bids
Bidding Strategies
Bid function of Suppliers (linear bid)
qi ( p ) = p / msi
Single Side Bidding
For the fixed demand D, MCP (p*) will be
Ng
1
p *

i =1 msi
=D

Demand Curve
d i ( p) = ( pi 0 p ) / m di
Bidding Strategies
MCP in double side bidding
Nd
pi 0
*

i =1
mdi
p =
Ng 1 Nd
1


msi
+
mdi
i =1 i =1

MCP with Wind Power


Option-I: Allowed to bid into the market
and take MCP with some premium
Bidding Strategies
Option-II:MCP is calculated without
wind power. Wind power get this MCP,
if available.

Other Aspects of Wind Power Trading


Mitigation of Market Power
Market Collusion
Ancillary Service Provisions
Case Studies
The output of RES generator is assumed to be 5 MW. Following cases are
studied:
Case-A: Linear supply bid with fixed demand
Case-B: Linear supply bid with linear demand bid

Table I: Linear Bid data


msi qgmax qgmin
($/MW2) (MW) (MW)
Bidder-1 0.10 20 100
Bidder-2 0.25 10 50
Bidder-3 0.20 10 100
Case Studies
Case-A: Linear supply bid with fixed demand
Demand D= 100 MW
If RES would not be available, the total demand
would have been met by the three supply
bidders 1, 2, and 3 (52.63 MW, 21.05 MW and
26.32 MW, respectively).
MCP will be $5.26/MWh
If RES power of 5 MW is available well in
advance, the MCP would be $5.00/MWh.
Generators share 95 MW.
Due to varying nature of RES, the MCP $5.00 is
not good.
Case Studies
Let wind power is allowed to bid. The MCP
and output will be
Case Studies
TABLE II: OUTPUT POWER AND PAYMENTS (CASE-A)
Output (MW) Payments ($) Payment in
ms>10 ms <1 ms>10 ms<1 Option II
Bidder-1 52.63 50.0 277.01 250.00 263.16
Bidder-2 21.05 20.0 110.80 100.00 105.26
Bidder-3 26.32 25.0 138.51 125.00 138.58
Wind Power 0.00 5.0 0.00 25.00 26.32
Total 100.0 100.0 526.32 500.00 526.32

- Bidding rate (ms) from 1 to 10 will be reduced the


RES power dispatch.
- In option II, even in the case of non-availability of
RES power due to absence of wind/solar, the market
operator is not going to have deficit of money
Case Studies
CASE-B: Double side bidding
Two different demand bidders are considered as
TABLE III: LINEAR DEMAND BID DATA
Data mdi ($/MW/MW) pi0 ($/MW)
Demand-1 0.050 7.0
Demand-2 0.075 8.0

The MCP, which is the intersection of cumulative


supply and total demand curves.
Without wind power MCP = $ 4.713/MWh and
required demand would be 45.73 & 43.82 MW.
Case Studies

Nd
pi 0

i =1 m di
5
p* = With zero RES power bid.
Ng
1 Nd
1
+
i =1 m si i =1 mdi
Case Studies
TABLE IV: OUTPUT POWER AND PAYMENTS (CASE-B)
Without wind power With wind power
Power Payment Power Payment Payment
(MW) (at 4.713) (MW) (at 4.618) (at 4.713)
Bidder-1 47.13 222.14 46.18 213.26 217.64
Bidder-2 18.85 88.86 18.47 85.32 87.05
Bidder-3 23.57 111.10 23.08 106.63 108.81
Wind 0.00 0.00 5.00 23.09 23.56
Demand-1 45.73 -215.56 47.64 -220.00 -224.53
Demand-2 43.82 -206.54 45.09 -208.20 -212.53

If market provision has the imbalance penalties for


down loading, the generators can be paid at actual
MCP (i.e. $4.618/MWh) with zero bid of wind power
whereas RES power can be paid at maximum MCP.
Case Studies
Money given to the wind generation is higher
with MCP calculated at zero RES power.
The payment to be collected from the
demand bidder will be at the maximum MCP
rate
In the event of non-availability of RES
power, the suppliers will be asked to
increase their output and the RES power
money will be given to them.
Conclusions
A suitable market mechanism is required to
accommodate the non-conventional energy
sources due to several technical and non-
technical reasons.
Having intermittent nature of availability, a
proper trading option must be used for these
sources to recover their costs in competitive
power market.
A proper use of these sources can also
avoid the abuse of competitive power market
by which the efficiency of the market can be
increased.
Conclusions
RES power can play a vital role in mitigating
the market power, ancillary services but their
costs must be recovered for successful
promotion of wind power energy.
The trading with RES power must be
transparent for free and fair trade of
electricity.
This talk could be a guideline for the policy
makers and market operators to promote the
RES power with system reliability and
security.

You might also like