面向航空发动机装配线的知识化制造自适应优化调度

摘要/Abstract

摘要：

为了解决不确定生产环境下的航空发动机装配调度问题，设计了一种面向航空发动机装配线的知识化制造自适应优化调度算法。算法采用强化学习和过程仿真相结合的调度策略求解方式，以最小化提前期惩罚费用和完工时间成本为调度目标，给出了航空发动机装配的Q学习自适应调度模型；针对装配调度问题定义了四个新的调度规则，定义了航空发动机装配的四个状态特征用于对系统状态进行描述，并针对调度目标设计了合理的回报函数。仿真实验结果表明，在调度过程中，采用提出的Q学习方法在多数情况下都远优于其他规则，尤其在装配任务到达频繁的情况下，总体上表现出更好的优势，显示了良好的自适应性能。

关键词: 调度, 航空发动机装配, 知识化制造自适应, Q学习

Abstract:

To solve the problem of aircraft engine assembly scheduling in an uncertain production environment, an adaptive optimization scheduling algorithm of a knowledgeable manufacture oriented to an aircraft engine assembly lines was proposed, where a scheduling-policy solved mechanism combining Q learning and process simulation was used. A Q-learning adaptive scheduling model of aircraft engine assembly was built on the objective function of minimizing earliness penalty and completion time cost. Then four new scheduling rules were provided for assembly scheduling problem, four state features of aircraft engine assembly were defined for describing system states, and the proper reward function was designed for the objective function. Some simulation experiments indicate that the proposed algorithm outperforms other scheduling rules much in most cases, especially, better results are generally achieved with the frequently changes of task arrival rates to show good adaptive performance.

Key words: scheduling, aircraft engine assembly, knowledgeable manufacturing adaptation(KMA), Q-learning

中图分类号:

TH165

汪浩祥, 严洪森, 汪峥, . 面向航空发动机装配线的知识化制造自适应优化调度[J]. 中国机械工程, 2014, 25(23): 3180-3187,3194.

Wang Haoxiang, Yan Hongsen, Wang Zheng, . Adaptive Optimization Scheduling of Knowledgeable Manufacture Oriented to Aircraft Engine Assembly Lines[J]. China Mechanical Engineering, 2014, 25(23): 3180-3187,3194.

参考文献

[1]王成恩，于宏，张闻雷,等. 面向对象的航空发动机装配模型[J]. 计算机集成制造系统, 2010, 16(5): 942-948.
Wang Chenen, Yu Hong, Zhang Wenlei, et al. Object-oriented Aero-engine Assembly Models[J]. Computer Integrated Manufacturing Systems, 2010, 16(5): 942-948.
[2]汤新民，钟诗胜. 基于离散时间最优控制的航空发动机装配序列规划[J]. 控制与决策, 2008, 23(11): 1221-1230.
Tang Xinmin, Zhong Shisheng. Aero-engine Assembly Sequence Planning Based on Discrete-time Pontryagin’s Minimum Principl[J]. Control and Decision, 2008, 23(11): 1221-1230.
[3]Pathumnakula S, Egbelu P J. An Algorithm for Minimizing Weighted Earliness Penalty in Assembly Job Shops[J]. International Journal of Production Economics, 2006, 103: 230-245.
[4]Chan F T S, Wong T C, Chan L Y. Lot Streaming for Product Assembly in Job Shop Environment[J]. Robotics and Computer Integrated Manufacturing, 2008, 24(3): 321-331.
[5]Chen J C, Chen K H, Wu J J, et al. A Study of the Flexible Job Shop Scheduling Problem with Parallel Machines and Reentrant Process[J]. International Journal of Advanced Manufacturing Technology, 2008, 39(3/4): 344-354.
[6]谭宜勇, 王锐, 范玉顺,等. 分布式工作流中的自适应实时任务调度方法[J]. 计算机集成制造系统, 2009, 15(10): 1956-1959.
Tan Yiyong, Wang Rui, Fan Yushun, et al. Adaptive Scheduling Method for Real-time Tasks in Distributed Workflow[J]. Computer Integrated Manufacturing Systems,2009, 15(10): 1956-1959.
[7]蔡政英，肖人彬，谭勇,等. 不确定条件下循环供应链模糊自适应生产计划调度[J]. 控制与决策, 2008, 23(5): 524-529.
Cai Zhengying, Xiao Renbin, Tan Yong, et al. Fuzzy Adaptive Production Plan Dispatching of Cycle Supply Chain under Uncertainty Conditions[J]. Control and Decision, 2008, 23(5): 524-529.
[8]Noroozi A, Hadi M, Abadi I N K. Research on Computational Intelligence Algorithms with Adaptive Learning Approach for Scheduling Problems with Batch Processing Machines[J]. Neurocomputing, 2013, 101: 190-203.
[9]Behnamian J, Ghomi S M T F. The Heterogeneous Multi-factory Production Network Scheduling with Adaptive Communication Policy and Parallel Machine[J]. Information Sciences, 2013, 219: 181-196.
[10]Susan K H, Jonathan F B, Douglas J M. A GRASP for Simultaneously Assigning and Sequencing Product Families on Flexible Assembly Lines[J]. Annals of Operations Research, 2013, 203: 295-323.
[11]Nie L, Shao X Y, Gao L, et al. Evolving Scheduling Rules with Gene Expression Programming for Dynamic Single-machine Scheduling Problems[J]. International Journal of Advanced Manufacturing Technology, 2008, 50(5/8): 729-747.
[12]Nguyen S, Zhang M J, Johnston M, et al. Learning Iterative Dispatching Rules for Job Shop Scheduling with Genetic Programming[J]. International Journal of Advanced Manufacturing Technology, 2013, 67(1/4): 85-100.
[13]Sutton R S, Barto A G. Reinforcement Learning: an Introduction[M]. Cambridge, Massachusetts: MIT Press, 1998.
[14]Wei Y Z, Zhao M Y. A Reinforcement Learning-based Approach to Dynamic Job Shop Scheduling[J]. Acta Automatica Sinica, 2005, 31(5): 765-771.
[15]王国磊,林琳,钟诗胜.基于聚类状态隶属度的动态调度Q-学习[J]. 高技术通讯, 2009, 19(4): 428-433.
Wang Guolei, Lin Lin, Zhong Shisheng. Clustering State Membership-based Q-learning for Dynamic Scheduling[J]. Chinese High Technology Letters, 2009,19(4): 428-433.
[16]李文超, 严洪森. 一类流水型知识化制造单元调度的自进化算法[J]. 中国机械工程, 2011, 22(7), 830-835.
Li Wenchao, Yan Hongsen.A Self-evolution Algorithm for Scheduling a Flow-shop-like Knowledgeable Manufacturing Cell[J]. China Mechanical Engineering, 2011, 22(7), 830-835.
[17]Watkins C, Dayan P. Q-learning[J]. Machine Learning, 1992, 8(3/4): 279-292.
[18]Aydin M E, ztemel E. Dynamic Job-shop Scheduling Using Reinforcement Learning Agents[J]. Robotics and Autonomous Systems, 2000, 33(2): 169-178.
[19]Wang Y C, Usher J M. Application of Reinforcement Learning for Agent Based Production Scheduling[J]. Engineering Applications of Artificial Intelligence, 2005, 18(1): 73-82.
[20]Zhang Z C, Zheng L, Li N. Minimizing Mean Weighted Tardiness in Unrelated Parallel Machine Scheduling with Reinforcement Learning[J]. Computers & Operations Research, 2012, 39(7): 1315-1324.
[21]严洪森, 刘飞. 知识化制造系统——新一代先进制造系统[J]. 计算机集成制造系统, 2001, 7(8): 7-11.
Yan Hongsen, Liu Fei. Knowledgeable Manufacturing System-a New Kind of Advanced Manufacturing System[J]. Computer Integrated Manufacturing Systems, 2001, 7(8): 7-11.