首页|基于改进麻雀搜索算法的矩形件排样优化研究

基于改进麻雀搜索算法的矩形件排样优化研究

扫码查看
基于改进麻雀搜索算法的矩形件二维排样优化研究有助于提高目前排样算法的排样效率,为实际工程应用提供理论基础。文章基于标准麻雀算法,引入帐篷Tent混沌映射和麦尔特洛夫Metropolis准则确定矩形件的排样顺序,并利用最低水平线算法排布矩形件放置位置。结果表明:采用Tent混沌映射初始化麻雀种群,会使种群分布更加均匀,增强了算法在矩形件二维排样过程中的全局搜索能力;采用Metropolis准则更新麻雀个体位置,增强了算法寻优过程中跳出局部最优解的能力;改进的麻雀搜索算法提升了 3。5%的原料利用率、排样时间减少了 55。5 s、消耗原料长度减少了 355。5 mm,验证了最低水平线法和改进的麻雀搜索算法在矩形件二维排样问题中组合优化的可行性。
Research on rectangular layout optimization based on improved sparrow search algorithm
The research on the two-dimensional nesting optimization of rectangular pieces,based on an improved sparrow search algorithm,can enhance the efficiency of current nesting algorithms and provide a theoretical foundation for practical engineering applications.The paper deviates from standard sparrow algorithms,introducing the Tent chaotic mapping and the Metropolis criterion criterion for probability to determine the order of rectangular pieces.The Lowest Horizontal Line Algorithm is then utilized to arrange the placement positions of these rectangular pieces.The results indicate that initializing the sparrow population with Tent chaotic mapping leads to a more uniform distribution of the population,enhancing the algorithm's global search capability during the two-dimensional nesting process of rectangular pieces.The use of the Metropolis criterion for updating the positions of sparrow individuals strengthens the algorithm's ability to avoid local optima during the optimization process.The improved sparrow search algorithm increases raw material utilization by 3.5%,reduces nesting time by 55 seconds,and decreases raw material length consumption by 355.5 mm.These validate the feasibility of the combined optimization of the Lowest Horizontal Line Algorithm and the improved sparrow search algorithm for two-dimensional nesting problems of rectangular pieces.

rectangular pieces2D layoutsparrow search algorithmcombinatorial optimization

张新、李虎啸、梁敏

展开 >

山东建筑大学 土木工程学院,山东 济南 250101

山东省国土空间数据和遥感技术研究院,山东 济南 250002

矩形件 二维排样 麻雀搜索算法 组合优化

2024

山东建筑大学学报
山东建筑大学

山东建筑大学学报

影响因子:0.576
ISSN:1673-7644
年,卷(期):2024.39(6)