摘要
针对粒子群优化(PSO)算法局部搜索能力不足的问题,提出一种基于Pareto邻域交叉算子的多目标粒子群优化算法(MPSOP)。该算法利用粒子群优化算法和Pareto邻域交叉算子相结合的策略产生新种群,并利用尺度因子在线调节粒子群优化算法和Pareto邻域交叉算子的贡献量。数值实验选取6个常用测试函数并对NSGA-Ⅱ、SPEA2、MOPSO三个多目标算法进行比较,数值实验结果表明MPSOP算法的有效性。
A multi-objective particle swarm optimization algorithm with Pareto neighborhood crossover operation（MPSOP） was proposed to solve the defect of local search in particle swarm optimization algorithm.MPSOP combined particle swarm optimization algorithm and Pareto neighborhood crossover operation to generate a new population.A scaling factor was used to balance contributions of particle swarm optimization algorithm and Pareto neighborhood crossover operation.Numerical experiments were conducted ti compared MOSOP with NSGA-Ⅱ,AND SPEA2 on six benchmark problems.The numerical results show the effectiveness of MPSOP.
出处
《计算机应用》
CSCD
北大核心
2011年第7期1789-1792,共4页
journal of Computer Applications
基金
国家自然科学基金资助项目(60962006)
关键词
多目标优化
粒子群优化算法
Pareto邻域交叉算子
尺度因子
multi-objective optimization
Particle Swarm Optimization（PSO）algorithm
Pareto neighborhood crossover operation
scaling factor
