基于遗传算法的UPFC模糊阻尼控制器参数优化及与梯度下降法的比较  

Comparison between GA and gradient descent algorithm in parameter optimization of UPFC fuzzy damping controller

作　　者：刘皓明[1] MOKTK 李扬[1] 倪以信 

机构地区：[1]东南大学电气工程系 [2]香港大学电机电子工程系 [3]清华港大深圳电力系统研究所,广东深圳518055

出　　处：《电力自动化设备》2005年第11期5-10,共6页Electric Power Automation Equipment

基　　金：国家重点基础研究发展计划资助项目(2004CB217900);国家自然科学基金资助项目(50337010)~~

摘　　要：设计了一种统一潮流控制器(UPFC)模糊辅助控制器,能够阻尼互联系统之间功率低频振荡,分别使用遗传算法和梯度下降法实现了模糊阻尼控制器(FDC)的参数优化。建立了UPFC的频域模型,给出了主控制和辅助控制的框图,推导了模糊阻尼控制器参数优化的公式。互联多机系统算例显示了2种优化方法都具有很好的收敛性,且结果吻合,同时由这2种方法优化的阻尼控制器能够很好地阻尼系统的功率振荡,且比传统的基于线性系统设计的辅助控制器在不同的系统运行方式和振荡模式下具有更好的鲁棒性。A fuzzy supplementary controller is designed for UPFC to damp tie line power oscillation in interconnected power systems. The parameters of FDC（Fuzzy Damping Controller） are optimized using the gradient descent training method and the genetic algorithm respectively. The UPFC power frequency model with its main and supplementary control block diagrams and the formulation for FDC parameter optimization are presented. Time simulation on a multi-machine inter-connected power system shows that the two optimization methods converge very well and yield similar FDC parameters,with which the designed FDC performs well in damping the tie line power oscillation under large disturbances. Computer simulation results also show that the FDC designed with these two methods has better robustness under different system operation conditions than that designed with linear system.

关 键 词：统一潮流控制器 辅助控制器 模糊逻辑 梯度下降法 遗传算法 

分 类 号：TM714.3[电气工程—电力系统及自动化]