反向学习全局和声搜索算法

提出一种反向学习全局和声搜索(OLGHS)算法.基于反向学习技术初始化和声记忆库,提高初始和声向量的质量;通过当前最差和声向当前最优和声学习进化,提高算法的全局搜索性能;通过其他和声向量之间不断回溯交互的随机学习策略,提高算法局部搜索性能;用由两种不同学习策略随机交叉动态产生的新和声与反向和声二者较优的个体更新和声记忆库,提高算法的搜索性能.将OLGHS算法与其他启发式优化算法以及目前文献中较优的改进HS算法进行性能测试,测试结果表明OLGHS算法具有较高的寻优精度和较快的收敛速度.

Opposition-based learning in global harmony search algorithm

This paper proposes an opposition-based learning global harmony search(OLGHS) algorithm. An opposition-based learning initialization technique is employed for initialize the harmony memory to enhance the quality of the initial harmony vector. The worst harmony learns from the best harmony, which can improve the global search performance of the algorithm. The local search performance of the algorithm is enhanced by means of random learning strategy of backtracking interaction among other harmony vectors. The new harmony is dynamically generated by means of random global crossover with two different learning strategies, and the harmony memory is updated by the optimal individual of the improvising harmony and its opposition harmony. Finally, a comparison test with other heuristic optimization algorithms and HS variants is carried out to test the optimization performance of the proposed algorithm. The simulation results demonstrate the OLGHS algorithm has higher convergence precision and convergence rate.