不确定知识化制造环境下航空发动机装配车间滚动自进化

发布时间：2018-11-03 14:18

【摘要】：针对产品装配次数不确定且装配组的调整时间与工序间排序相关的航空发动机装配车间,对不确定环境下知识化制造系统(Knowledgeable manufacturing system,KMS)的自进化问题进行研究。采用事件和周期混合驱动型的自进化机制,结合滚动时域方法实现航空发动机装配车间自进化,并给出一种可行的滚动规则。建立系统在各个决策时刻的静态决策子问题的数学模型,并针对该模型提出一种具有双层结构的遗传算法进行求解。在下层的混合型遗传算法中,给出一种直接解码算法,并引入了变邻域搜索算法,以增强局部搜索的能力。通过仿真对算法的性能进行了测试,此外,试验数据表明执行自进化操作的系统具有较好的生产性能。尤其是对于更敏感于自身调整的系统,自进化操作发挥的作用更大。
[Abstract]:The self-evolution of the knowledge-based manufacturing system (Knowledgeable manufacturing system,KMS) in uncertain environments is studied for the aeroengine assembly workshop where the product assembly times are uncertain and the adjustment time of assembly groups is related to the sequencing among processes. An event and periodic hybrid driven self-evolution mechanism is used to realize the self-evolution of the aeroengine assembly workshop with the rolling time domain method and a feasible rolling rule is given. The mathematical model of the static decision subproblem of the system at each decision time is established, and a genetic algorithm with two-layer structure is proposed to solve the model. In order to enhance the ability of local search, a direct decoding algorithm and a variable neighborhood search algorithm are presented in the lower layer hybrid genetic algorithm. The performance of the algorithm is tested by simulation. In addition, the experimental data show that the system that performs the self-evolution operation has good production performance. In particular, for systems that are more sensitive to self-adjustment, self-evolutionary operations play a greater role.
【作者单位】：东南大学自动化学院;东南大学复杂工程系统测量与控制教育部重点实验室;
【基金】：国家自然科学基金(60934008) 中央高校基本科研业务费专项资金(2242014K10031) 江苏高校优势学科建设工程资助项目
【分类号】：V263.2;TP18

【参考文献】