改进交叉算子和变异算子抑制GA算法早熟

Improved Crossover Operators and Mutation Operators to Prevent Premature Convergence

分析了传统遗传算法早熟收敛的主要原因,提出了一类改进的遗传算法。通过引入个体相似度,改进传统的交叉算子,避免了近亲繁殖现象,采用二元变异算子替换传统变异算子。仿真结果表明该改进算法有效地提高了全局搜索性能和收敛速度。

Analyzed the primary reasons for premature algorithm was proposed. In order to avoid close breeding convergence of genetic algorithm, an improved genetic , introduced similarity to improve traditional crossover operators, binary mutation was designed to take place of traditional mutation. The result of simulation shows the effective improvement of the global searching performance and the convergent speed of this improved algorithm.