基于Q学习的微网二次频率在线自适应控制

由于包含微源的多样性及运行模式的多样性,微网的二次频率控制面临着系统参数不确定性的挑战。文中提出了在多代理(Agent)分层混合控制模型中嵌入一种基于Q学习的智能算法。首先,动态预测出微网系统实时二次调频功率缺额值。其次,同时考虑微网运行经济性和环境效益,并采用模糊化方法和粒子群优化算法实现二次调度功率的分配。最后,在C++Builder环境下搭建了包括不同微源的本地层Agent和具有不同控制功能的中央层Agent的微网混合能量管理仿真平台,结果证明了所提出的基于Q学习的微网二次频率自适应控制器可以自适应微网系统结构及其参数的动态变化,实现微网二次调频的智能控制。

Q learning Algorithm Based Secondary Frequency Adaptive Online Control in Real time Operation for Microgrids

Because of the diversity of micro sources and operation modes in a microgrid, secondary frequency control faces great challenges from the uncertainty of system parameters of the microgrid. To solve the problem, a Q learning intelligent algorithm integrated in the hierarchical multi agent model is proposed. Firstly, by the proposed method, the power to be regulated, which is called the microgrid regulation error (MRE), is dynamically calculated in the secondary control for real time operation. Secondly, the generation schedule of distributed generators and batteries is modified in real time with the MRE by the fuzzy theory and particle swarm optimization method by taking both economy and environmental benefits into consideration. Finally, a Q learning algorithm based multi agent hybrid energy management system for the microgrid simulation platform in terms of client server frame is established in C++ Builder. Simulation results have verified that by the proposed Q Learning method, secondary intelligent and adaptive frequency control is realized under the condition of variable structure and parameters in the microgrid. This work is supported by National Basic Research Program of China (973 Program) (No. 2009CB219708), National Natural Science Foundation of China (No. 51077033), Specialized Research Fund for the Doctoral Program of Higher Education (No. 201301111110005), and the Introduction of Innovative Team of Guangdong Province Project (No. 2011N015).