直觉模糊小生境的自适应遗传算法求解旅行商问题

发布时间：2018-04-23 12:40

本文选题：直觉模糊集 + 小生境　；参考：《计算机科学》2016年12期

【摘要】：提出一种基于直觉模糊距离测度的小生境技术,结合模糊控制的自适应遗传算法求解旅行商问题。运用个体在遗传算法迭代寻优中的适应度值,通过直觉模糊集的距离测度确定个体之间的相似性,使用共享函数和惩罚函数对适应度低的个体进行惩罚和淘汰,维护了种群个体的多样性;建立模糊推理系统,以自适应调节遗传算法迭代中的交叉率和变异率,使遗传算法能在局部寻优和全局寻优之间达到平衡,弥补遗传算法易早熟收敛和后期寻优能力差的缺陷;通过求解TSPLIB中的多组实例并进行对比,结果表明所提算法的收敛速度、优化精度、效率均具有明显优势。
[Abstract]:A niche technology based on intuitionistic fuzzy distance measure is proposed, which combines the adaptive genetic algorithm of fuzzy control to solve the traveling salesman problem. By using the fitness value of the individual in the genetic algorithm iterative optimization, the similarity between individuals is determined by the distance measure of the intuitionistic fuzzy set, and the fitness is low by using the shared function and the penalty function. The individual is punished and eliminated, and the diversity of the individual is maintained, and a fuzzy inference system is established to adjust the cross rate and mutation rate in the iteration of the genetic algorithm, so that the genetic algorithm can balance between the local optimization and the global optimization, and make up the defects of the early convergence and the poor optimization ability of the genetic algorithm. Several examples of TSPLIB are compared and the results show that the proposed algorithm has obvious advantages in terms of convergence speed, optimization accuracy and efficiency.

【作者单位】：空军工程大学防空反导学院;
【基金】：国家自然科学基金(61402517) 中国博士后基金(2013M542331) 陕西省自然科学基金(2013JQ8035)资助
【分类号】：TP18

【相似文献】