自调节基因移民遗传算法的研究

通过对遗传算法选择，交叉，变异3个算子的分析，揭示了早熟现象产生的原因。在此基础上，设计了一种变异概率可以自调节的基因移民遗传算法，提出用群体相异度指标来衡量群体的多样性。通过相异度可自动调节变异频率的大小，从而确保了群体的多样性，也避免了早熟现象的出现。最后，应用马尔科夫链数学工具，从理论上证明了该算法是收敛的，实验表明，该算法可操作性好，对选择的参数不敏感，具有一定的通用性，适合于实际的工程计算。

Study on Self-Adjusting of Gene Migration Genetic Algorithm

The selection operator, the crossover operator and the mutation operator of the genetic algorithm are analyzed in the paper. The reason of the premature phenomena is revealed. Then based on the analysis, a self-adjusting gene migration genetic algorithm with the self-change mutation rate is designed. And the guideline of the population differentia is put forward. The probability of mutation is automatically changed by the population differentia. So the population diversity is kept and the premature phenomena are avoided. At last, the convergence of the algorithm is proved by the Markov-chain. The experimental results indicate that it is a feasible method and not sensitive to the parameters. The method has general applicability and is suitable for the engineering calculation.