论文部分内容阅读
基于代理的仿真已成为电力市场研究的重要方法。目前,强化学习算法在代理中应用的研究已较为深入,通过随机数进行策略选择是强化学习算法的一个重要环节,因此对随机数在其中的影响研究十分必要。文中以Roth-Erev学习算法为例,分析了随机数对电力市场仿真结果的影响,比较了几种随机数生成方法在仿真中应用的优劣。同时,通过深入研究电力市场仿真中随机数与代理的关系,探讨了对随机数性质的要求,提出一种改进方法生成了可重复、高效、均匀的随机数,从而提高了基于代理的电力市场仿真实验的效率和质量。
Agent-based simulation has become an important method in power market research. At present, the research on the application of reinforcement learning algorithm in agent is more in-depth. The choice of strategy by random number is an important part of the reinforcement learning algorithm. Therefore, it is necessary to study the influence of random number in it. Taking Roth-Erev learning algorithm as an example, the paper analyzes the influence of random number on power market simulation results and compares the advantages and disadvantages of several random number generation methods in simulation. At the same time, by studying the relationship between the random number and the agent in the electricity market simulation, this paper discusses the requirements of the nature of the random number, and proposes an improved method to generate repeatable, efficient and uniform random numbers, which improves the agent-based electricity market Simulation efficiency and quality of experiments.