构造发电公司最优报价策略的机会约束规划方法

在采用暗标拍卖的电力市场环境中,发电公司在构造报价策略时需要对竞争对手的报价行为或市场电价进行估计,即报价决策是在不完全的信息基础上作出的,这样就不可避免地会带来一定的风险,因而需要进行风险管理.为避免现有的用方差度量风险时所存在的问题,作者借用了金融理论中的风险价值的思想,提出了与发电公司报价策略相关的风险的新的度量方法.在此基础上,构造了发电公司最优报价策略的机会约束规划模型,提出了将遗传算法嵌入蒙特卡罗随机模拟的求解方法.最后,用算例对所提出的模型和方法进行了验证.

A CHANCE CONSTRAINED PROGRAMMING BASED APPROACH FOR BUILDING OPTIMAL BIDDING STRATEGIES FOR GENERATION COMPANIES WITH RISK MANAGEMENT

In a sealed auction based electricity market, it is necessary for a generation company to develop a bidding strategy based on estimations of bidding behaviors of competitive generation companies or the market clearing price of electricity, or in other words, the bidding strategy has to be built based on incomplete information. As a result, risk will inevitably be introduced and has to be properly managed. In order to get ride of the disadvantages of the conventional variance-based risk evaluation method, the well-developed concept of value at risk (VaR) from financial theories is employed for developing a new evaluation criterion of risk associated with the bidding strategy. On this basis, a new model is presented for building a risk-constrained optimal bidding strategy for a generation company in the framework of chance-constrained programming, and a genetic algorithm (GA) based solving approach proposed with the well-known Monte Carlo simulation embedded. Finally, a numerical example of a simulated electricity market with six participating generation companies is served for demonstrating the feasibility of the developed model and method.