面向复杂零件网络化制造的资源优化配置方法  

Resources optimization deployment approach for complicated parts in networked manufacturing

作　　者：姚倡锋[1] 张定华[1] 彭文利[1] 李山[1] 

机构地区：[1]西北工业大学现代设计与集成制造教育部重点实验室,陕西西安710072

出　　处：《计算机集成制造系统》2006年第7期1060-1067,共8页Computer Integrated Manufacturing Systems

基　　金：国家863/CIMS主题资助项目(2001AA412020;2003AA414320);西北工业大学博士论文创新基金资助项目(CX200412);航空科学基金资助项目(04H53064)。~~

摘　　要：为实现复杂零件网络化制造的资源优化配置,提出一种以零件工艺流程为核心的制造资源优化配置方法。该方法依据零件工艺流程,设计逻辑制造单元和逻辑加工路线,进行基于逻辑制造单元和逻辑加工路线的制造任务分解和描述。为了对逻辑加工路线配置最优的可执行加工路线,以运行成本、时间和加工质量为目标,建立多目标优化模型;在用遗传算法求解模型的过程中,每一代种群,基于影响度因子将运行成本、时间和加工质量的目标特征值矩阵转换为相对隶属度矩阵,进而以每个染色体的距优海明加权距离平方与距劣海明加权距离平方之和最小为准则,求取基于相对隶属度的适应度值。最后,以具体实例证明了该方法解决复杂零件网络化制造资源优化配置问题的可行性与有效性。To solve resources optimization deployment problem for complicated parts in networked manufacturing, an approach taking the process flow of part as kernel was proposed. In this approach, the design of Logical Manufacturing Process （LMP） and Logical Manufacturing Unit （LMU） was studied according to the process flow of part, and manufacturing tasks were decomposed and described based on LMP and LMU. To obtain the best Executive Manufacturing Process （EMP） for LMP, the multi--objective manufacturing resources optimization deployment problem was modeled considering the operation Cost, Time and machining Quality （CTQ）. During solving the model by genetic algorithm, the eigenvalue matrix of CTQ was transformed into the relative membership degree matrix based on the influence degree coefficient in every generation, and the fitness value was calculated by taking the minimal sum of the square of hamming weighted distance of an individual to the ideal excellent individual and the square of hamming weighted distance of an individual to the ideal inferior individual as optimization rule. Finally, the validity and practicability of the approach were verified by an example of engine blade in networked manufacturing.

关 键 词：资源优化配置 物理制造单元 逻辑制造单元 逻辑加工路线 可执行加工路线 遗传算法 相对隶属度 

分 类 号：TH165[机械工程—机械制造及自动化]