Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/210895
Authors: 
Sarfati, Mahir
Hesamzadeh, Mohammad Reza
Holmberg, Pär
Year of Publication: 
2018
Series/Report no.: 
IFN Working Paper 1254
Abstract: 
In part I of this paper, we proposed a Mixed-Integer Linear Program (MILP) to analyze imperfect competition of oligopoly producers in two-stage zonal power markets. In part II of this paper, we propose a solution algorithm which decomposes the proposed MILP model into several subproblems and solve them in parallel and iteratively. Our solution algorithm reduces the solution time of the MILP model and it allows us to analyze largescale examples. To tackle the multiple Subgame Perfect Nash Equilibria (SPNE) situation, we propose a SPNE-band approach. The SPNE band is split into several subintervals and the proposed solution algorithm finds a representative SPNE in each subinterval. Each subinterval is independent from each other, so this structure enables us to use parallel computing. We also design a pre-feasibility test to identify the subintervals without SPNE. Our proposed solution algorithm and our SPNE-band approach are demonstrated on the 6-node and the modified IEEE 30-node example systems. The computational tractability of our solution algorithm is illustrated for the IEEE 118-node and 300-node systems.
Subjects: 
Modied Benders decomposition
Multiple Subgame Perfect Nash equilibria
Parallel computing
Wholesale electricity market
Zonal pricing
JEL: 
C61
C63
C72
D43
L13
L94
Document Type: 
Working Paper

Files in This Item:
File
Size
637.04 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.