摘要
基于粒子群优化算法提出一种机电产品三维管路自动敷设算法,该算法以预处理和优化搜索为基本框架。在预处理阶段,利用混沌技术,建立混沌栅格预处理模型。三维管路敷设空间经过混沌栅格预处理模型处理,可有效地降低管路自动敷设算法的求解复杂度。在优化搜索阶段,通过分析粒子群算法和人口迁移的特点,提出改进粒子群算法。改进粒子群算法针对粒子群算法收敛速度慢、容易陷入局部极小的缺点,提出子空间局部搜索、解空间全局搜索和补空间开拓搜索。基于混沌栅格预处理模型和改进粒子群算法,给出高效的管路自动敷设算法流程图。对比试验和改进粒子群算法的理论分析,表明该优化搜索算法具有较好的性能,并证明改进粒子群算法的收敛性。实例验证表明该算法的有效性和实用性。
Based on particle swarm optimization （PSO） algorithm, a new automatic pipe-routing algorithm is proposed for electromechanical products in 3D space. The algorithm consists of pre-processing phase and searching optimization phase. The pre-processing phase is aimed at making implement the searching optimization phase easily by reducing the searching range efficiently. By the chaos technique, chaos grid pre-processing model （CGPM） is constructed in the pre-processing phase. The main objective of the searching optimization phase is to find an approximate optimal trajectory in the available solution space. In the searching optimization phase, a modified PSO algorithm is designed to search for an optimal pipe trajectory by analyzing PSO algorithm and migration characters of people. Modified PSO gives evolution strategies for overcoming premature convergence of PSO. By CGPM and modified PSO, the automatic pipe-routing algorithm flow chart is put forward. By contrastive experiment and theory analysis, it shows that modified POS algorithm effectively increases diversity of particles and has well performance. The convergence of the modified PSO algorithm is given. The example illustrates effectiveness and practicability of the automatic pipe-routing algorithm.
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2007年第11期194-199,共6页
Journal of Mechanical Engineering
基金
黑龙江省留学归国基金(20010701018)。
关键词
管路自动敷设算法
混沌栅格预处理模型
改进粒子群优化
Automatic pipe-routing algorithm Chaos grid pre-processing model Modified particle swarm optimization
