基于混合并行遗传算法的多目标约束优化技术研究  

Research of Multi-objectives Constraints Optimization Technology Based on Hybrid Parallel Genetic Algorithm

作　　者：焦翠珍[1] 戴文华[1] 

机构地区：[1]咸宁学院计算机系,湖北咸宁437005

出　　处：《沈阳农业大学学报》2006年第1期125-127,共3页Journal of Shenyang Agricultural University

摘　　要：多目标约束优化问题属于NP问题。并行遗传算法是解决该类问题的常用算法,它具有较强的全局搜索能力和并行性,但局部搜索能力差,禁忌搜索算法则比较适合于局部搜索。提出了一种基于混合并行遗传算法的多目标约束优化方法,该方法综合了并行遗传算法和禁忌搜索算法的优势,改进了并行遗传算法的性能,能有效避免局部最优解。The multi-objectives constraints optimization question belongs to the NP question, The parallel genetic algorithm is a usual way to solve this kind of question. It has the strong overall situation search ability and parallelism, but the partial search ability is bad. The tabu search algorithm is quite suitably in partial search. The multi-objectives constraints optimization arithmetic based on hybrid parallel genetic algorithm is proposed in this paper. This method synthesizes the superiority of the parallel genetic algorithm and the tabu search algorithm, improves the performance of the parallel genetic algorithm, can effectively avoid the partial optimal solution.

关 键 词：并行遗传算法 禁忌搜索算法 多目标 约束 优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]