Abstract: Aiming at the task disturbance problem represented by order disturbance appears in the production process of workshop,a multi-objective dynamic scheduling model with completion time,machine waiting time,total delay time and workpiece deviation time was constructed.An adaptive scheduling strategy based on tolerance degree of workpiece delay was proposed,and an improved gray wolf algorithm was designed to solve it.According to the characteristics of the workshop scheduling,the strategy based on process encoding and decoding was adopted to propose the reverse strategy of plane mirror imaging for generating the initial population.The population diversity was enriched by improving the catfish effect strategy,and the linear convergence factor and crossover mutation operator was introduced to balance the search ability of the algorithm.In the simulation environment,a comparative experiment was carried out for the case of order disturbance,which verified the effectiveness of the proposed algorithm.It was showed that the adaptive scheduling strategy and improved gray wolf algorithm could enhance the stability of scheduling system.

Key words: order perturbation, flexible job shop, dynamic scheduling, gray wolf algorithm

摘要： 针对车间生产过程中出现以订单扰动为代表的任务扰动问题,建立了以完工时间、机器等待时间、总拖期时间和工件偏差时间为优化目标的多目标动态调度模型,提出一种基于工件延迟容忍度的自适应调度策略,并设计了一种改进灰狼算法进行求解。在该算法中,根据车间调度的特点,采用了基于工序编码与解码的策略方法,提出了平面镜成像的反向学习策略以生成初始种群;通过改进鲶鱼效应策略,丰富了种群多样性;引入非线性收敛因子、交叉变异算子,以平衡算法搜索能力。在仿真环境下,针对订单扰动情形进行了对比实验,验证了所提算法的有效性。实验结果表明所提出的自适应调度策略及改进灰狼算法在一定程度上增强了调度系统的稳定性。

关键词: 订单扰动, 柔性作业车间, 动态调度, 灰狼算法