基于最小二乘支持向量机和自适应模拟退火算法的电磁场逆问题全局优化方法

分析了目前电磁场逆问题全局优化算法存在的收敛速度慢以及搜索时间长等问题的主要原因,并针对以上问题提出了基于最小二乘支持向量机和自适应模拟退火电磁场逆问题优化新算法,充分利用了自适应模拟退火算法中丢失的已搜索过点的信息,动态地建立和改进待求问题的数值模型,指导最优解的搜索过程,大大减少了求解电磁场正问题的求解次数,缩短了搜索到最优解的时间,通过仿真实验以及实际应用的对比,效果显著,提高了电磁场优化设计的实际应用能力。

A Global Optimization Algorithm Based on Least Squares Support Vector Machines and Adaptive Simulated Annealing Algorithm for Inverse Electromagnetic Problem

Main reasons for lower convergence speed and longer time consumption problems existed in the global optimization algorithm of the inverse electromagnetic problem are analyzed. In order to solve these problems, a new global optimization algorithm for the inverse electromagnetic problem is presented and it is based on the least squares Support Vector Machines (SVM) and the adaptive simulated annealing. In searching process of the adaptive simulated annealing algorithm, the solution information searched is fully taken to construct dynamically and improve the approximation mathematical model of optimization problem being solved by SVM. The model can be used in the searching process of adaptive simulated annealing algorithm to decrease the times of solving forward electromagnetic problem. And finally the time of solving inverse electromagnetic problem is greatly decreased. The comparison for the computational results shows that the new algorithm presented has better effect and the ability for practical application of electromagnetic optimization is enhanced greatly.