动态联盟伙伴选择的一种自适应遗传算法

针对动态联盟伙伴选择优化问题，提出一种自适应遗传算法用来求解此类问题。该算法设计了自适应交叉和变异概率，在遗传过程中可以根据适应度自动选择，从而使群体中每个个体对环境的变化具有自适应调节能力；所设计的自适应变异概率可以避免算法的早熟现象；遗传过程中，通过保持迭代过程中的最优解，加快了搜索速度，并保证了收敛于全局最优解。通过算例，证实了该算法的有效。

A Self-adaptive Genetic Algorithm for Partner Selection of Agile Virtual Ente rprise

Partner selection for Agile Virtual Enterprise (AVE) is a type of problem of resoruce combination during the building process of AVE, and is a hotspot in this field. A self-adaptive Genetic Algorithm (GA) is proposed for solving this problem. Self-adaptive probabilities of crossover and mutation are designed in this algorithm, and they can be selected automatically according to fitness. Thus, each individual of the group owns the ability of self-adaptation according to the variation of the environment. The self-adaptive probability of mutation designed can avoid the premature of the algorithm. And, the searching speed is improved by holding the optimum in iteration, which can guarantee the global optimal solution being found as well. This adaptive GA is proved valid by example.