论文部分内容阅读
针对局部搜索类NSGA2算法计算量大的问题,提出一种基于密度的局部搜索NSGA2算法(NSGA2-DLS).使用解的密度衡量解的稀疏度,并将当前非支配解中稀疏度最小的解定义为稀疏解,每次遗传过程在稀疏解周围进行局部搜索.在局部搜索过程中,同时采用极限优化策略和随机搜索策略以提高解的质量和收敛速度.对ZDT系列函数和DTLZ系列函数进行仿真实验并与标准NSGA2算法、一种局部随机搜索算法和一种定向搜索算法进行比较,结果表明,NSGA2-DLS在消耗计算量和优化效果方面均优于对比方法.
In order to solve the problem of large computational load of NSGA2 algorithm, a density-based NSGA2-based local search algorithm (NSGA2-DLS) is proposed. The density of solution is used to measure the sparsity and the sparsity of the current non-dominated solution Defined as a sparse solution, each local genetic process sparse solution around the local search.In the local search process, both limit optimization strategy and random search strategy to improve the quality of the solution and convergence speed of ZDT series functions and DTLZ series of functions The simulation experiments are compared with the standard NSGA2 algorithm, a local random search algorithm and a directional search algorithm. The results show that NSGA2-DLS outperforms the comparative method in terms of consumption and optimization.