摘要
通过分析用户需求偏好和资源灰色关联度,建立资源模糊相似矩阵并构建资源模糊聚类树。以复合值模糊聚类树为基础,以资源类内聚合度、资源类间分离度和用户需求满意度为目标,构建资源聚类优化数学模型,然后运用遗传算法进行优化。算法采用了交叉概率和变异概率自适应的重构策略和保优操作,避免了算法的早熟,增强了算法的寻优能力和搜索效率。通过实例仿真验证了算法的有效性。
After analyzing customer demands and the grey relation degree of manufacturing resources, a fuzzy similarity matrix and fuzzy cluster tree were constructed. An optimization model of resource clustering evaluated by internal genus degree, external detached degree and satisfaction degree of customer demands was proposed, which was based on fuzzy clustering tree. Then, genetic algorithm was applied to optimize the clustering results. To prevent the premature convergence problem and improve the optimization capability, the adaptive crossover and mutation probabilities method and elitist selection were employed. The simulation results indicate the effectiveness of the proposed new algorithm.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2011年第23期2828-2833,共6页
China Mechanical Engineering
基金
国家自然科学基金资助项目(61104171)
关键词
制造资源
用户需求
模糊聚类
遗传算法
manufacturing resource
customer demand
fuzzy clustering
genetic algorithm
