Abstract: Aiming at function optimization problem, by analyzing the influence of two key parameters(harmony pitch adjusting rate and harmony bandwidth) on search performance of Harmony Search(HS) algorithm, this paper proposes dynamic changing methods for harmony pitch adjusting rate and harmony bandwidth during the search, and obtains nine improved harmony search algorithms. Experimental results show that compared with M_IHS algorithm, the improved algorithms have higher optimization performance.

Key words: Harmony Search(HS) algorithm, optimization problem, continual optimization, evolutionary algorithm

摘要： 针对函数优化问题，通过分析和声搜索算法的2个关键参数(和声微调概率与和声微调幅度)对算法搜索性能的影响，提出和声微调概率与和声微调幅度随搜索过程的进行而动态适应变化的方法，从而得到9种改进的和声搜索算法。仿真实验表明，所得方法具有较好的优化性能，计算结果优于M_IHS算法。

关键词: 和声搜索算法, 优化问题, 连续优化, 进化算法

CLC Number: 

• TP18