混合滤波系统中无源滤波器的均匀设计多目标优化方法  

Uniform design based multi-objective evolutionary design approach for passive power filters in hybrid power filter system

作　　者：赵曙光[1] 赵明英[1] 王军宁[2] 王宇平[3] 

机构地区：[1]西安电子科技大学电子工程学院,陕西西安710071 [2]西安电子科技大学通信工程学院,陕西西安710071 [3]西安电子科技大学理学院,陕西西安710071

出　　处：《电力自动化设备》2005年第1期7-11,共5页Electric Power Automation Equipment

基　　金：国家自然科学基金重点项目(60133010);国家自然科学基金资助项目(60374063)

摘　　要：针对在很大程度上决定混合滤波系统整体性能的无源电力滤波器参数优化问题,提出一种基于均匀设计的多目标自适应遗传算法,其核心是基于元件标称值的高效编/解码方案,基于权值均匀生成、动态调整和Pspice仿真的多目标合成与评估方法,可提高后代质量并减少运算量的多父本均匀交叉算子,以及兼顾种群多样性和遗传进程的遗传概率自适应策略。实验结果表明,该方法仅需单次运行即可获得一组经过优化且反映多目标间均衡关系的实用化设计结果。A UDT(Uniform Design Techniques) based multi-objective adaptive genetic algorithm is proposed and detailed for improving the optimization of PPF (Passive Power Filter),which determines the comprehensive performances of HPFS(Hybrid Power Filter Systems). It is characterized by an efficient and effective encoding-decoding scheme based on the standard series of component values,a mechanism of multi-objective integration and evaluation with dynamic adjusting of uniformly selected weight vectors and Pspice simulation,a UDT based multi-parent multi-offspring crossover operator for improving offspring's quality and decrease the computation cost,and an adaptation technique for genetic parameters to maintain the individual diversity and track the genetic process. Simulative results show that this approach can obtain a group of effective design results of PPFs through an execution,which meets the main optimizing objectives and reflects the impact of these objectives on each other. This project is supported by the National Natural Science Foundation Key Project of China(60133010) and the National Natural Science Foundation of China(60374063).

关 键 词：混合电力滤波系统 无源电力滤波器 自适应遗传算法 多目标优化 均匀设计 

分 类 号：TM711[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]