基于混合优化策略的微分进化改进算法

微分进化算法具有控制参数少、鲁棒性强、易于使用等优点，并具有不同的优化策略．本文在对微分进化算法各优化策略性能进行分析的基础上，提出了基于混合优化策略的微分进化改进算法．改进算法的主要思想是将种群中的个体随机地分成两组，每组采用不同的优化策略．利用五个标准的优化算法测试函数对改进算法的收敛速度和搜索成功率进行了测试，并与动态微分进化算法和微粒群算法进行了比较．实验结果表明，本文提出的改进算法在保证算法搜索成功率的同时，大大提高了算法搜索效率．

A Modified Differential Evolution Algorithm with Hybrid Optimization Strategy

The differential evolution algorithm is robust,easy to use,requires few control parameters,and has various optimization strategies. Based on analysis of advantages and disadvantages of these optimization strategies, a modified differential evolution algorithm with hybrid optimization strategy is proposed. The main idea of the modified differential evolution algorithm is to divide all of the individuals into two groups randomly, and the two groups adopt different optimization strategies, The convergence speed and search succeed probability of the modified differential evolution are tested using five benchmark functions for optimization algorithm, and the results are compared with dynamic differential evolution and particle swarm optimization. From the simulation resuits, it is observed that the search efficiency of the modified differential evolution is significantly improved as well as the high search succeed probability is ensured.