摘要
针对传统人工蜂群算法早熟收敛问题,基于模糊化处理和蜂群寻优的特点,提出一种模糊人工蜂群算法。将模糊输入输出机制引入到算法中来保持蜜源访问概率的动态更新。根据算法计算过程中的不同阶段对蜜源访问概率有效调整,避免算法陷入局部极值。通过对多选择多维背包问题的仿真实验和与其他算法的比较,表明本算法可行有效,有良好的鲁棒性。
Aiming at the premature convergence problem in traditional artificial bees colony algorithm, fuzzy artificial bees colony algorithm is proposed, which is ba.sed on the principles of fuzzy processing and bees colony behavior. Fuzzy inputs and fuzzy outputs are introduced into the algorithm to maintain dynamic updates of the nectar access probability. According to effective adjustment on nectar access probability during the different stages of algorithm calculation, the algorithm avoids local optima. Simulated tests of multi-choice multidimensional knapsack problem and comparisons with other algorithms show the algorithm is feasible and effective and the algorithm has strong global optimization ability.
出处
《运筹与管理》
CSSCI
CSCD
北大核心
2013年第5期98-103,共6页
Operations Research and Management Science
基金
国家自然科学基金资助项目(70871081)
上海市重点学科建设资助项目(S30504)
上海市研究生创新基金资助项目(JWCXSL1201)
关键词
智能优化算法
模糊规则
模糊人工蜂群算法
多选择多维背包问题
intelligent optimization algorithm
fuzzy rules
fuzzy artificial bees colony algorithm
multi-choice multidimensional knapsack problem
