差异演化算法及其改进

Differential Evolution Algorithms and Modification

针对一类非凸、多峰及非线性函数优化难的问题,本文提供了一种新的求解策略——差异演化算法,该算法比其它演化算法稳健性强、收敛速度快。同时,为进一步提高差异演化算法的优化性能,该文也提出了一些改进措施,包括自适应线性变异以及迁徙操作,最后的仿真试验验证了本改进算法的优越性。

A new method called differential evolution algorithm for the optimization problem of nonconve, multi-modal (and non-linear) functions is provided in the paper. The algorithm is much robuster and quicker in convergence than other (evolution) algorithms. At the same time, some modifications including adaptive linear mutation and migration operation are proposed for differential evolution to improve it's optimized performance. The last test of simulated testing shows modified differential evolution's superiority.