摘要
针对普通遗传算法易出现早熟收敛和搜索效率低的缺陷,提出一种基于加权海明距离的自适应遗传算法.该算法综合考虑个体间加权海明距离和适应度值,自适应调整交叉概率和变异概率;采用精英保留法,保证最优个体不被破坏;使用双重停机准则,减少不必要的计算时间,提高遗传搜索效率.最后,运用经典测试函数对该算法进行了仿真实验.结果表明,该算法可以显著提高遗传优化的全局搜索能力,加快遗传算法的收敛速度.
To alleviate the defect of premature convergence and low-search efficiency of traditionalgenetic algo- rithms, an adaptive genetic algorithm based on weighted hamming distance is proposed. The algorithm considers the weighted hamming distaneeand the fitness value, adjusting crossover probability anti'mutation probability adaptively. It uses the method of elite preserving to ensure the best individual is not damaged. It also employs the criterion of dual stopping to reduce unnecessary computing time and improves the efficiency of genetic search. Finally, some simulation experiments are carried out with classical test functions in the Matlab platform. Experimental results show that the proposed algorithm can effectively improve the global search ability of genetic optimization, and speed up the convergence of genetic algorithm.
出处
《华南师范大学学报(自然科学版)》
CAS
北大核心
2015年第6期121-127,共7页
Journal of South China Normal University(Natural Science Edition)
基金
国家自然科学基金项目(61300107)
广州市科技计划基金项目(2013J4300055)
关键词
加权海明距离
遗传算法
自适应
收敛
weighted hamming distance
genetic algorithm
adaptation
convergence