面向节能的导弹结构件混线生产作业车间多目标调度研究

Energy-saving Oriented Multi-objective Shop Floor Scheduling for Mixed-line Production of Missile Components

针对导弹结构件混线生产过程中具有的型号多、工艺复杂、生产能耗大、交货期紧的特性,以能量消耗和完工时间为目标,建立了基于设备-能耗曲线的柔性作业车间混线生产系统的数学调度模型。提出了一种双元混合的改进遗传算法对该调度模型进行求解,具体包括:引入粒子群算法的信息共享机制,对遗传算法的交叉算子进行改进,提高算法的寻优能力;用Hill函数构建传统模拟退火的温度更新函数,替代遗传算法的变异部分,以弥补遗传算法容易陷入早熟收敛的不足。采用多指标加权灰靶决策模型从得到的一组Pareto解集中选择最满意调度方案。分别用完全柔性和部分柔性的作业实例对算法进行验证,证明了改进算法的有效性。最后,将算法用于上海航天精密机械研究所结构件生产车间的生产实例,取得了较好生产的指导效果。

Aimed at the characteristics of multi-project production, complex process, high energy consumption and tight delivery in the mix-line production of missile structural components, an optimization model for a flexible job-shop scheduling problem considering energy consumption and makespan is developed based on equipment state-energy-consumption curve. A binary hybrid improved genetic algorithm(BH-GA) is proposed to solve the established optimization problem. To improve the searching ability of the algorithm, an information-sharing mechanism based on particle swarm optimization(PSO) is introduced to design the crossover operation of genetic algorithm(GA). In order to avoid falling into local optical solution, a novel temperature update function of simulated annealing algorithm(SA) based on Hill function is used to replace the mutation operation of GA. In addition, a weighted multi-attribute grey target decision model is adopted to select the most satisfactory schedule scheme. The effectiveness of the proposed algorithm is verified by the completely and partially flexible scheduling problems. Finally, the proposed algorithm is applied to the production workshop of missile structural components at an aerospace institute in Shanghai, and good effect is gained.