一种求解冰壶比赛对阵多约束问题的逐层优化算法

发布时间：2018-01-10 18:09

本文关键词：一种求解冰壶比赛对阵多约束问题的逐层优化算法　出处：《电子学报》2017年03期 　论文类型：期刊论文

【摘要】：冰壶比赛对阵编排问题是一个难于收敛的多约束优化问题.为此提出一种求解此类问题的逐层优化的单亲遗传算法.首先将待求解问题的多个约束进行分层;其次设计了靶向自交叉算子进行第一层优化以提高搜索效率,设计了定点-随机自交叉算子进行第二层优化以保持种群的多样性;最后,将改进的算法用于解决冰壶比赛对阵编排的多约束优化问题,构建了该问题的适应度函数.仿真实验表明,与粒子群算法和经典遗传算法相比,所提算法能够有效求解冰壶比赛对阵编排的多约束优化问题.
[Abstract]:The curling match scheduling problem is a multi-constraint optimization problem which is difficult to converge. For this reason, a single-parent genetic algorithm is proposed to solve this problem. Firstly, the multi-constraints of the problem to be solved are stratified. Secondly, the target self-crossover operator is designed to optimize the first layer to improve the search efficiency, and the fixed-point random self-crossover operator is designed to carry out the second layer optimization to maintain the diversity of the population. Finally, the improved algorithm is used to solve the multi-constraint optimization problem of curling match scheduling, and the fitness function of the problem is constructed. The simulation results show that compared with the particle swarm optimization algorithm and the classical genetic algorithm. The proposed algorithm can effectively solve the multi-constraint optimization problem of curling match scheduling.
【作者单位】：哈尔滨工程大学计算机科学与技术学院;牡丹江师范学院计算机与信息技术学院;
【基金】：国家自然科学基金资助项目(No.61472095,No.61272186) 黑龙江省教育厅智能教育与信息工程重点实验室开放基金支持牡丹江师范学院青年项目(No.QY2014003,No.QN201603)
【分类号】：TP18
【正文快照】： 1引言冰壶比赛对阵编排问题是一个多约束条件下的组合优化问题[1].目前国内外对冰壶比赛对阵方案的设计还没有实现自动化,也没有相关成果发表,比赛的对阵方案仍旧由裁判专家手工完成.为实现冰壶比赛对阵编排的自动化,我们已尝试使用多种方法:使用贪心算法[2]进行冰壶比赛对阵

【相似文献】