摘要
研究分布估计算法可以解决难优化问题,且具有很好的全局搜索能力,但存在局部搜索能力差以及因种群多样性容易丧失从而导致的早熟收敛问题。针对上述问题对分布估计算法进行改进,将优势解集克隆,对优势个体进行搜索,从而增强局部搜索能力,并对概率模型进行修正以改善种群多样性损失问题,通过对多维背包问题的标准问题进行测试比较,结果表明了改进的有效性,改进后的算法增加了局部搜索能力、有效保持了种群多样性,获得好的优化结果。
An improved algorithm based on estimation of distribution algorithm was proposed to solve muhidimen- sional knapsack problem(MKP). The estimation of distribution algorithm is good in global search, but poor in local search, and may suffer from premature convergence beacause of diversity loss. The proposed algorithm cloned and searched the dominance solutions to strengthen the local searching ability and corrected the probability model to im- prove the diversity loss. Numerical simulation was carried out based on the benchmark instances, and the results were compared.
出处
《计算机仿真》
CSCD
北大核心
2014年第10期286-290,共5页
Computer Simulation
基金
国家自然科学基金(61271143)
国家自然科学基金(60871080)
关键词
分布估计算法
优势克隆
概率模型修正
背包问题
Estimation of distribution algorithm
Dominance clone
Probability model correction
Knapsack problem