自适应基因遗传算法及其在知识获取中的应用

Adaptive Genetic Algorithm and Its Application to Knowledge Acquisition

本文针对基因遗传算法中杂交率和变异率的难以选取问题,提出了一种自适应基因遗传算法.该方法利用降半Г分布函数对杂交率和变异率进行自适应调整,以保证群体的多样性和进化过程的稳定性,克服算法的未成熟收敛问题.最后以故障诊断知识获取为例,阐述该方法的有效性.

In this paper, an adaptive genetic algorithm is proposed to solve the problem of choosing the probabilities of crossover and mutation. A special function is defined to adptively adjust the above two parameters, which garantees the diversity and stability of genetic algorithm and avoids its premature convergence. Application examples of diagnostic knowlege acquisition de-mostrate that the proposed approach is more effective than standard genetic algorithm.