遗传算法的一种改进进化策略

An Improved Evolutionary Strategies of Genetic Algorithm

在现有文献研究的基础上,对传统遗传算法的进化策略又作了进一步研究,提出了一种改进的进化策略.进化策略克服了传统遗传算法中交又得到的优秀个体有可能在变异过程中遭到破坏而不能生存的不足.另外取消了遗传算法中难以确定的交叉、变异概率,使交叉产生的新个体数增多,这样可增大产生更优秀个体的可能性,因而可使遗传算法的性能得到更好的改善.通过4个测试函数的测试计算,结果表明,给出的改进进化策略比传统遗传算法进化策略的运算速度明显提高,迭代次数明显减少,从而验证了提出的改进进化策略的有效性.

The paper provides an improved evolutionary strategy of genetic algorithm on the basis of the existing literature, which makes further research in this field. The evolutionary strategies overcome the shortage of traditional genetic algorithm whose excellent individuals may not survive in the process of mutation. In addition, the crossover probability and mutation probability which is hard to determine is canceled too, At the same time, it increase the number of individual produced in process of crossover. This may increase possi- bility of producing excellent individuals, thus enable the performance of genetic algorithm to get a better improvement. The result shows that improved evolution strategies which were presented in this paper have faster calculation speed and the number of iterations signifi- cantly reduced than the traditional genetic algorithm by the trial calculation of the four test functions. Thus, validity of improved evolution strategies is powerfully illustrated.