In order to optimize the coverage of wireless sensor network (WSN),for the MOEA/D had two deficiencies,such as the lack of preserving generation of high quality individuals and individuals rarely in optimal solutions,this paper proposed MOEA/D-PSO(multi-objective evolutionary algorithm based on decomposition particle swarm optimization).By preserving the generation of high-quality individuals population,and improving local optimization solution set in evolutionary search direction and the search for progress,it made up for the shortcomings of the original MOEA/D.Simulation experimental results show that com-pared with MOEA/D and NSGA-Ⅱ(non-dominated sorting genetic algorithm-Ⅱ),the non-dominated solutions of MOEA/D-PSO is closer to Pareto optimal surface,uniformity and diversity of solution set distribution perform better.Coverage of WSN is more widely,and consumes less energy.%为了优化无线传感器网络(WSN)的覆盖方法,针对MOEA/D中缺少对本代优质个体的保存和最优解集中个体极少的两个问题,提出了粒子群优化的基于分解的多目标进化算法(MOEA/D-PSO)。通过保留种群本代优质个体,改进本地优化解集在进化过程中的搜索方向和搜索进度,弥补了MOEA/D的不足。仿真实验证明,相对于MOEA/D和非支配排序遗传算法(NSGA-Ⅱ),MOEA/D-PSO所得非支配解更接近Pareto最优曲面,解集分布的均匀性和多样性表现更佳,WSN的覆盖范围更广,能量消耗更少。
展开▼