Markov过程理论在发电商报价策略选择中的应用

Optimal Bidding Strategy Based on Markov Decision Process for Generation Companies

在电力市场的环境下,发电商在报价策略选择的过程中将面临许多不确定性因素,如各时段的系统负荷、市场边际价格、对手的报价策略、自身发电成本等,本文从发电商的角度出发对竞价策略的选择问题进行研究,将竞价策略选择的过程设计成为Markov决策过程——一种抽象的随机优化方法,先将每个时段的报价简化为报价参数的选择,利用Markov过程理论对不确定性因素的出现进行概率估计,将发电商报价策略的选择问题表示为离散的随机优化过程—— Markov决策过程;通过使期望收益最大来计算最优策略,从而确定了每个时段发电商报价参数的最优选择,并将报价参数还原成为报价结果。

There are various uncertainties such as system load, market marginal price, competitors＇ bidding strategies, generation cost and so on, which must be considered when the power supplier is choosing the bidding strategy in a spot market environment. The bidding decision-making problem is studied from a power supplier＇s viewpoint and the problem is formulated as a Markov Decision Process-a discrete stochastic optimization method. This paper simplifies every period＇s bidding to a form of selecting bidding parameters, and utilize the Markov theory to estimate the probabilities that uncertainties occur, however, the power supplier＇s bidding strategies choosing can be described as discrete stochastic optimization process-Markov decision process. The optimal selection of bidding parameters can be determined at every period based on the optimal strategy that maximizes the expected profit, and these parameters are converted to the bidding results.