面向复杂钣金件折弯的高效工序规划算法  

Efficient process planning algorithm for complex sheet metal bending

作　　者：刘飞 魏娜 吕万林 游有鹏[1] LIU Fei;WEI Na;LYU Wanlin;YOU Youpeng(College of Mechanical Engineering,Nanjing University of Aeronautics&Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学机电学院,江苏南京210016

出　　处：《计算机集成制造系统》2023年第10期3331-3339,共9页Computer Integrated Manufacturing Systems

基　　金：国家重点研发计划资助项目(2018YFB1309203)。

摘　　要：为了解决复杂钣金件工序规划难、效率低的问题,提出一种基于反向状态传递的高效工序规划算法。算法提出折弯可行性状态概念以及状态在先后折弯工序中的传递机制,在将状态数据作为经验充分利用的同时避免丢失,以减少传统工序规划中普遍存在的大量折弯可行性重复测试;考虑到钣金件的形状复杂度对折弯可行性的影响不同,采用从最终成形至零件毛坯的逆向推理方式减少无效的折弯可行性检测;为规避随折弯数增多导致解空间阶乘增长,采用经验的组合表示以避免重复的折弯可行性检测。实验结果表明,所提算法可以高效获得复杂钣金件的所有可行折弯工序,为最终通过评估、优选获得最优折弯工序奠定了基础。To solve the problem of difficult and inefficient process planning for complex sheet metal parts,an efficient process planning algorithm based on reverse state transfer was proposed.To minimize the number of bend feasibility tests,the concept of bending feasibility state and the transfer mechanism of state in successive bending processes was proposed,and the state data was used as experience to fully utilize and avoid loss,so as to reduce a large number of repeated detection of bending feasibility commonly found in traditional process planning.Considering the different influence of the shape complexity of sheet metal parts on the bending feasibility,the reverse reasoning method from final forming to part blank was adopted to reduce the invalid bending feasibility detection.To escape the influence of the factorial growth of the solution space caused by the increase of the number of bends,the combination representation of experience was used to avoid repeated bending feasibility detection.The experimental results showed that the proposed algorithm could efficiently obtain all feasible bending processes of complex sheet metal parts,which laid a foundation for obtaining the optimal bending process through evaluation and optimization.

关 键 词：复杂钣金件 工序规划 状态传递 逆向推理 组合表示 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]