多智能体代理下电力双边谈判中的模糊贝叶斯学习模型

在基于多智能体(multi-Agent)代理技术的电力双边谈判中，提高Agent的环境自调节能力对提高谈判效率有着重要意义。该文通过将贝叶斯学习模型引入到基于Agent环境下的电力双边谈判中来赋予Agent学习的能力，使其能够根据动态环境调整自我认识，从而达到提高谈判效率的目的。与此同时，考虑到谈判环境的模糊不确定性，分别运用模糊集和模糊概率理论构建Agent的模糊贝叶斯学习模型，并给出在这种环境下Agent的点报价策略和区间报价策略。最后结合算例证明，谈判双方在都采取模糊贝叶斯学习情况下的谈判效率要比一方或双方都不采取贝叶斯学习情况下的谈判效率要高，而双方采取区间报价策略能有效缩减谈判时间，为决策者提供了更有弹性的决策空间。

A Fuzzy Bayesian Learning Model in Agent-based Electric Power Bilateral Negotiation

In Agent-based electric power bilateral negotiation, it is important to enhance the Agent’s ability to adjust according to the environment so as to improve the negotiation efficiency. By introducing Bayesian learning to Agent-based electric power bilateral negotiation, the Agent was empowered the ability to learn according to a dynamic environment and adjust itself, so that the efficiency of the negotiation was improved. At the same time, taking into account the fuzzy uncertainties of the negotiating environment, the fuzzy set and fuzzy probability theories were used to construct an Agent Fuzzy Bayesian learning model, and then its point bidding strategy and interval bidding strategy were designed. Finally, an example was given to prove that the negotiation where both the negotiators adopt fuzzy Bayesian learning is more efficient than the negotiation where neither or only one negotiator adopt fuzzy Bayesian learning. And the interval bidding strategy can save more time for negotiators, so it is more suitable for time-limited negotiation.