发电商多输入决策因子竞价的智能代理模拟方法

电力行业从垄断到竞争，市场成员的决策方式会发生很大的转变；受市场参与者成本、风险偏好、电力市场供求关系等因素的影响，其决策行为将受到多种因素的共同影响。在对诸多影响市场成员竞价决策的相关因素分析基础上，提炼关键影响因子分类建模，建立能够模拟发电商日前市场竞价行为的多输入决策因子模型，应用VRE-learning强化学习算法，在5节点测试系统上进行市场成员竞价行为模拟。实验结果表明，运用所建立的竞价决策模型可以较好地表征市场成员的风险特性和决策从属目标，验证了所提方法的有效性。

Intelligent Agent Based Bidding Simulation Method for Multi-input Decision Factors of Power Suppliers

The decision-making behavior of the market participants will change dramatically when the power industry evolves from monopoly to market competition. The decision-making behavior will be influenced by such factors as the cost and risk preference of market participants and supply and demand relationship of electricity market. Based on the analysis of related factors that affect the bidding decision of market participants, this paper refines the key influencing factors and builds a model that can simulate the bidding behavior of market participants in the day-ahead market. The VRE-learning reinforcement-learning algorithm is introduced into the model, which is simulated on the 5-node test system. The case analysis shows that the established bidding decision model can better characterize the risk characteristics and subordinate decision-making objectives of market participants, thus the validity of the proposed method is verified.