Data Envelopment Analysis and Stochastic Mixed Complementarity Problem for Solving Nash Equilibrium in Two-level Coal-fired Electricity Market

Abstract

Under the trend of electricity liberalization the electricity market becomes more competitive In fact due to the energy-type features of electricity power systems even the market with several generation firms can exercise the market power if there are some limitations such as transportation restrictions caused by geographical locations or political and economic factors Thus for some countries like China the market structures of electricity generation plants and transmission system operators (TSO) are still imperfectly competitive which may lead to productive inefficiency This study connects coal-fired plants in upstream market and grid operators in downstream market together and then investigate the Nash equilibrium which claims “rational inefficiency” but bring more profits for power industries We propose the data envelopment analysis (DEA) and mixed complementarity problem (MiCP) to identify the Nash equilibrium in the estimated production possibility set of the two-level electricity market Two cases of deterministic models based on different assumptions about operation of market power are discussed Then the directional distance function approach is applied to estimate the direction toward the Nash equilibrium and measure the efficiency Additionally to understand the impacts of price parameters on the equilibrium results the sensitivity analysis and optimization approaches for uncertainty are employed This study develops the stochastic mixed complementarity problem for solving the Nash equilibrium with stochastic programming and robust optimization respectively An empirical study of the electricity market in both north and northeast regions of China is conducted to validate the proposed models The results show the optimal equilibrium solution in the two-level electricity market and the efficiency analysis with respect to the Nash solution provides the managerial insights to drive the productivity in the two-level electricity market system The discussion for the parameter reveals that the price in the upstream market have significant influence to the equilibrium solution And the analysis of Nash solutions generated from stochastic mixed complementarity problem illustrates the necessity to model with uncertain parameters

Date of Award	2018 Jul 16
Original language	English
Supervisor	Chia-Yen Lee (Supervisor)