基于Pareto协同进化算法的多个FACTS元件协调控制  

Coordinative control of multiple FACTS controllers based on Pareto co-evolution algorithm

作　　者：刘青[1] 王增平[1] 

机构地区：[1]华北电力大学电力系统保护与动态安全监控教育部重点实验室,河北保定071003

出　　处：《电力自动化设备》2009年第7期79-81,96,共4页Electric Power Automation Equipment

摘　　要：针对多个FACTS装置的控制器之间存在的负交互影响,提出一种基于Pareto协同进化算法的协调控制方法。利用协同进化算法对控制器参数进行种群划分,对各种群采用遍历组合法构造新的个体并计算目标函数值。采用多种群合作策略进行非支配排序,取排序后的最优个体进行遗传操作,最后得到一组Pareto最优解集。与常规多目标优化算法相比,避免了一般多目标优化进化算法中难以处理的适应度值的分配问题,且利用协同进化算法的并行性大幅提高了收敛速度。利用该方法对静止同步串联补偿器(SSSC)与静止同步补偿器(STATCOM)2种典型FACTS装置的协调控制器进行了设计,并通过仿真验证了设计的协调控制器的控制效果良好。Aiming at the negative interaction among FACTS controllers,a coordinative control method based on Pareto co- evolution algorithm is proposed. The controller parameters are divided into populations by co-evolution algorithm and the ergodic combination method is used for each population to construct new unit and calculate the objective function value. The multi- population cooperation strategy is applied for non -dominated sorting and the inheritance operation is then carried out for the best unit to get a group of Pareto optimal solution. Compared with the conventional multi-objective optimization algorithm,the difficulty of adaptation degree distribution is avoided and the convergence rate is improved. The coordinative controller of two typical FACTS controllers,static synchronous series compensator and static synchronous compensator,is designed based on the proposed strategy. Simulation verifies the better control effect of designed controller.

关 键 词：FACTS控制器 多目标优化 协同进化算法 协调控制 遗传算法 

分 类 号：TM744[电气工程—电力系统及自动化] TP18[自动化与计算机技术—控制理论与控制工程]