基于萤火虫算法的管路系统布局序列优化技术  

Optimization technology of pipeline system layout sequence based on firefly algorithm

作　　者：吴宏超[1] 刘检华[1] 唐承统[1] 徐联杰 刘佳顺[1] 

机构地区：[1]北京理工大学机械与车辆学院数字化制造研究所,北京100081

出　　处：《计算机集成制造系统》2016年第8期1837-1848,共12页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(51275047);国防基础科研资助项目(A0420132503);总装预先研究资助项目(51318010102)~~

摘　　要：针对复杂机电产品中的多根管路布局设计与优化问题,提出一种基于萤火虫算法的管路布局序列优化方法。该方法以长度、折弯数和流阻为综合优化目标,结合布局过程的约束条件建立了优化模型;对萤火虫算法进行离散化,重新定义萤火虫的间距和个体更新机制,使其适用于序列规划问题的求解;提出一种确定性和随机性相结合的方法生成初始种群,以含免疫记忆的A*算法求得路径,同时采用混合种群迭代更新策略,通过更新精英解集获得全局最优解。设计并开发了原型系统,以旅行商问题为例对算法效率进行了测试,并将所提方法应用到某产品的液压管路布局上,验证了所提方法的可行性。To solve the multi-pipe layout design and optimization problem in mechanical and electrical products, a pipe layout sequence optimization method based on firefly algorithm was proposed. An optimization model was construc-ted by constraints and optimization objective, which consisted of pipe length, number of elbows and flow resistance. Furthermore, a firefly algorithm was modified to redefine the distance and flight mechanism among fireflies, which adapted to the solution of sequence planning problem. On this basis, a combination of deterministic and stochastic method was used to generate the initial population, the path was planned by A＊ algorithm with immune memory. Meanwhile, a mixed population iterative method was applied to update the population, eventually the global optimal solution could be obtained by updating the elite solution set. A prototype system was developed, and the efficiency of the algorithm was tested by TSP. The proposed method was applied to the hydraulic pipe system to verify the feasibility of proposed method.

关 键 词：管路布局 序列优化 萤火虫算法 多管路布局 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]