基于多项式混沌和进化算法的快速鲁棒优化算法及其在电磁场逆问题中的应用  

A Fast Robust Optimization Methodology Based on Polynomial Chaos and Evolutionary Algorithm for Inverse Problems

作　　者：吴烈[1] 杨仕友[1] 李玉玲[1] 

机构地区：[1]浙江大学电气工程学院,浙江省杭州市310027

出　　处：《中国电机工程学报》2013年第3期171-175,1,共5页Proceedings of the CSEE

基　　金：国家自然科学基金项目(50777054;51077114);浙江省重大科技专项(2011C11062)~~

摘　　要：为提高电磁场逆问题鲁棒优化设计的计算效率,提出一种基于多项式混沌和进化算法相结合的快速鲁棒优化方法。在该方法中,通过将多项式混沌展开作为鲁棒性能参数的随机响应面模型,提高了期望函数的计算效率和计算精度。此外,为进一步提高算法的计算效率,文章又提出了期望适值赋值的新机制,以及概率可行性模型处理约束函数鲁棒性的方法;引入同时搜索全局最优解和鲁棒最优解的迭代循环机制。通过典型电磁场逆问题的鲁棒优化设计对本文算法进行了数值验证。数值分析结果证明了该文方法的可行性和先进性。This paper explores the potential of polynomial chaos in robust design optimizations of inverse problems. To this end, a fast numerical methodology based on combinations of polynomial chaos expansion and evolutionary algorithm is proposed. In the proposed methodology, the polynomial chaos expansion is used as a stochastic response surface model for efficient computations of the expectancy metric 0fthe objective function. Additional enhancements, such as the introduction of new methodology for expected fitness assignment and probability feasibility model, a novel driving mechanism to bias the next iterative cycles to search for both global and robust optimal solutions, are introduced. The numerical results on two case studies are reported to illustrate the feasibility and advantage of the present work.

关 键 词：电磁场 逆问题 进化算法 多项式混沌 鲁棒优化 

分 类 号：TM155[电气工程—电工理论与新技术]