review of economic consequences of alternative solution ...best.eng.buffalo.edu/research/fall...

12
Review of Economic Consequences of Alternative Solution Methods for Centralized Unit Commitment in Day- Ahead Electricity Markets (Sioshansi, O'Neill et al. 2008) Mehrdad Sheikholeslami The Best Group

Upload: others

Post on 17-Jul-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Review of Economic Consequences of Alternative Solution

Methods for Centralized Unit Commitment in Day-Ahead Electricity Markets (Sioshansi, O'Neill et al.

2008)Mehrdad Sheikholeslami

The Best Group

Page 2: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

What is Unit Commitment (UC)?

Unit commitment (UC) in electric power systems is tooptimize generating resources to supply system loadwhile satisfying prevailing constraints, such asminimum on/off time, ramping up/down,minimum/maximum generating capacity, and fuel andemission limit (Tao and Shahidehpour 2005).

Page 3: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Another Definition of UC

Fundamentally, unit commitment has theresponsibility to find the least-cost commitment anddispatch of a set of generating units to meet expectedload over a time horizon consisting of a fixed numberof periods (e.g. a 24 hour day)(Sioshansi, O'Neill et al.2008).

Page 4: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Types of Unit Commitment

• Security-Constrained Unit Commitment (SCUC): SCUC isutilized by an independent system operator (ISO) to clear theday-ahead market (Tao and Shahidehpour 2005).

• Price-Based Unit Commitment (PBUC): PBUC is used byindividual generating companies (GENCOs) and refers to theoptimization of generating resources in order to maximizeGENCOs’ payoffs. This UC has a different objective than thatof SCUC and emphasizes the importance of the price signal(Tao and Shahidehpour 2005).

Page 5: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Means of Solving Unit Commitment

Many efficient near optimal methods have beendeveloped. The most successful approaches are:

1. Up until recently, the Lagrangian relaxation (LR)algorithm was the only practical means of solvingan ISO-scale unit commitment problem (Sioshansi,O'Neill et al. 2008).

Page 6: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Means of Solving Unit Commitment

2. Recent advances in computing capabilities andoptimization algorithms now make solution of themixed-integer programming (MIP) formulation bymeans of branch and bound (B&B) tractable(Sioshansi, O'Neill et al. 2008).

Branch & Bound Concept

Page 7: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

MIP Advantages

the advantages of the MIP method over LR include:

1) Global optimality;

2) Direct measure of the optimalityof a solution;

3) More flexible and accurate modeling capabilities

4) Adding constraints in MIP does not require modifications

to the solution algorithm as required in the LR method(Tao and Shahidehpour 2005).

Page 8: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

MIP Disadvantages

The disadvantage of the MIP approach is the computational

complexity. For the UC problem, the computation time of LR

is almost linear to the product of the number of units and the

scheduling period. (Xiaohong, Qiaozhu et al. 2003)(Tao and

Shahidehpour 2005)

Page 9: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Arising Issues From LR Solutions

(Johnson, Oren et al. 1997)

1. Equity: Different payoffs to individual resources

2. Incentive: Generators may profitably mispresent their cost

and constraint to affect the outcome of the market.

3. Efficiency: Without submitting truthful offers, the efficiency

of the underlying unit commitment solution is questionable.

Page 10: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Conclusion

The raised issues remain regardless of the solution technique

used. This market design issue will loom regardless of how

accurate the unit commitment solution is, unless an optimum

can be found (Sioshansi, O'Neill et al. 2008).

Page 11: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

References:

Sioshansi, R., R. O'Neill and S. S. Oren (2008). "Economic

Consequences of Alternative Solution Methods for Centralized Unit

Commitment in Day-Ahead Electricity Markets." Power Systems,

IEEE Transactions on 23(2): 344-352.

Tao, L. and M. Shahidehpour (2005). "Price-based unit commitment:

a case of Lagrangian relaxation versus mixed integer programming."

Power Systems, IEEE Transactions on 20(4): 2015-2025.

Xiaohong, G., Z. Qiaozhu and A. Papalexopoulos (2003).

Optimization based methods for unit commitment: Lagrangian

relaxation versus general mixed integer programming. Power

Engineering Society General Meeting, 2003, IEEE.

Page 12: Review of Economic Consequences of Alternative Solution ...best.eng.buffalo.edu/Research/Fall 2015/Unit Commitment.pdf · a case of Lagrangian relaxation versus mixed integer programming."

Johnson, R. B., S. S. Oren and A. J. Svoboda (1997). "Equity and

efficiency of unit commitment in competitive electricity markets."

Utilities Policy 6(1): 9-19.