基于免疫算法的拆卸序列规划方法研究

发布时间：2018-09-01 13:21

【摘要】：产品拆卸是绿色制造的重要环节,拆卸序列规划的目的是寻找拆卸成本最低,耗费时间最短的拆卸路径。为了使拆卸序列的评价更加符合实际,本文着重研究了拆卸序列的稳定性评价方法。为了提高拆卸过程的稳定性,减少拆卸时间,降低拆卸成本,本研究将紧固件看作独立的零件,分析螺纹面在重力支撑关系中的作用,然后根据重力的逐级传递关系建立重力支撑矩阵。然后,根据该矩阵判断并计算子装配体中不稳定零件的数量,对拆卸序列的稳定性进行评价。为了实现有效的拆卸序列规划,本文提出一种基于免疫算法的拆卸序列规划方法,针对拆卸序列规划的需求,结合免疫算法的特点,对现有算法进行改进,用单亲变异方法取代双亲交叉算子,以降低算法对初始种群的依赖;同时使用两种方法提取疫苗,一种来自于经验知识,另一种提取于进化过程中的种群,以提高算法的搜索速度;用精英代谢代替传统的新陈代谢,以提高算法的收敛性能。最后,将改进的免疫算法应用于装配体实例,通过试验的方法,分别从基准件疫苗,疫苗提取间隔代数,疫苗接种概率和精英代谢率等几个方面对算法性能进行验证,得到一个较佳的参数设置,以证明算法的收敛效率。同时,对拆卸过程中,子装配体稳定性评估和零件稳定性评估进行比较,证明使用零件的稳定性对拆卸过程进行评估的方法比较直接、合理。
[Abstract]:Product disassembly is an important part of green manufacturing. The purpose of disassembly sequence planning is to find the disassembly path with the lowest cost and the shortest time. In order to make the evaluation of disassembly sequence more practical, the stability evaluation method of disassembly sequence is studied in this paper. In order to improve the stability of the disassembly process, reduce the disassembly time and reduce the disassembly cost, the fastener is regarded as an independent part in this study, and the role of the thread surface in the gravity support relationship is analyzed. Then the gravity support matrix is established according to the stepwise transfer relation of gravity. Then, the stability of disassembly sequence is evaluated by judging and calculating the number of unstable parts in subassembly according to the matrix. In order to realize effective disassembly sequence planning, this paper presents a disassembly sequence planning method based on immune algorithm. According to the requirement of disassembly sequence planning and the characteristics of immune algorithm, the existing algorithms are improved. The parent crossover operator is replaced by a single parent mutation method to reduce the dependence of the algorithm on the initial population, and two methods are used to extract the vaccine, one from empirical knowledge and the other from evolutionary populations. In order to improve the search speed of the algorithm and to replace the traditional metabolism with elite metabolism, the convergence performance of the algorithm can be improved. Finally, the improved immune algorithm is applied to the assembly case, and the performance of the algorithm is verified from several aspects, such as benchmark vaccine, vaccine extraction interval algebra, vaccination probability and elite metabolic rate, etc. A better parameter setting is obtained to prove the convergence efficiency of the algorithm. At the same time, the stability evaluation of subassemblies and parts is compared in the process of disassembly, which proves that the method of evaluating the disassembly process by using the stability of parts is more direct and reasonable.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH186;TP18

【相似文献】