学术活动发布与订制系统

In the context of liberalized markets, market outcomes generally result from the strategic interactions of all market players. Generation company (Genco), as the distributed players, build their subjective demand evaluations (SDFs) about market for optimal bidding purpose. The picture of a real electricity market game in Genco’s eye is ‘playing is believing’. Therefore, a question naturally comes to the table: how those SDFs with the heterogeneous manner impact individual player’s decision and game results. To answer this question, this presentation relaxes a conventional assumption, commonly used in the classical oligopolistic equilibrium model, that a correct and uniform demand knowledge is shared by all Gencos. The investigations are carried out from three perspectives:Firstly, the impact of SDFs is analyzed in a dynamic Cournot game and a transmission constrained supply function bidding market game respectively. The economic value of perfect information in the two game settings is discussed. Secondly, to alleviate the system oscillations caused by SDFs in the conventional conjectural variation (CV)-based learning bidding method, a data filter is designed to make bidding system stable. Finally, the stochastic learning technique, i.e. agent-based reinforcement learning, is employed to answer the research questions from another point of view. The comparisons are made between the equilibrium-oriented (EO) approach mentioned above and the multi-agent learning (MAL) approach.

The findings from this presentation provide insights on several interesting issues aroused by assuming players’ heterogeneous beliefs in computational game theory under the background of electricity market. The results can provide important guidelines in multiple perspectives, e.g. the effective design of a dynamic bidding learning process considering players’ heterogeneous demand knowledge; the policy implications of certain bidding behaviors caused by the imperfect demand evaluations, etc.

裘智峰，比利时（荷语）鲁汶大学ELECTA/ESAT博士，博士后。2001年和2004年获得中南大学学士和硕士学位。长期从事电力系统优化调度运行以及电力市场建模等相关方向的研究，例如，在新的市场环境下以利益最大化成本最低为目标的机组组合的经济调度模型及算法研究；在间歇能源环境下基于风险对冲的传统发电机组调度优化问题；在非完全竞争电力市场下基于智能体的决策者模型以及相关增强学习算法；基于同/异质信息认知的博弈模型在电力市场中的应用等。相关研究论文在多个著名国际期刊上发表，并多次在各种学术会议上交流。