基于改进NSGA-Ⅱ算法的多目标生产智能调度

Multi-objective Intelligent Production Optimal Scheduling Based on Improved NSGA-Ⅱ Algorithm

在制造业自动化、智能化生产模式的需求日益增加的趋势下,针对生产制造过程中生产工序安排不合理造成的生产效率低、资源浪费严重等问题,构建以最大完成时间和最大生产成本的智能优化调度模型,使用一种改进的NSGA-Ⅱ算法进行研究。通过MSOS染色体编码方案,将个体基因分成机器和工序两部分分别编码。种群初始化通过适当扩大种群的方式,提高算法的全局搜索能力。采用动态拥挤度计算和精英保留的方式选择个体,保证解得多样性的同时保留优秀个体。采用动态混合交叉算法,确保种群更快地向最优方向进化。利用动态自适应变异概率来提高种群后期的多样性。通过对西装上衣缝制环节生产环节编排调度的仿真实验,验证了改进算法在多目标生产智能优化调度中的可行性和有效性。

With the increasing demand for automated and intelligent production models in the manufacturing industry,in view of the low production efficiency and serious waste of resources caused by unreasonable production process arrangements in the manufacturing process,an intelligent optimization scheduling model with the maximum completion time and maximum production cost is constructed.An improved NSGA-Ⅱ algorithm is used for research.Through the MSOS chromosome coding scheme,the individual gene is divided into the machine and the process to code respectively.The population initialization improves the global search ability of the algorithm by appropriately expanding the population.The dynamic crowding calculation and elite retention are used to select individuals to ensure the diversity of solutions while retaining outstanding individuals.The dynamic hybrid crossover algorithm is adopted to ensure that the population evolves to the optimal direction faster.The dynamic adaptive mutation probability is to increase the diversity of the population in the later stage.Through the simulation experiment of the scheduling and scheduling of the production link in the sewing of suits,the feasibility and effectiveness of the improved algorithm in the intelligent scheduling of multi-objective production are verified.