基于改进NSGA-II的柔性工艺路线多目标优化  

Multi-Objective Optimization of Flexible Process Route Based on Improved NSGA-II

作　　者：杨方晴 李仁旺[1] 叶晓蕾 

机构地区：[1]浙江理工大学机械工程学院,浙江 杭州

出　　处：《建模与仿真》2023年第2期786-798,共13页Modeling and Simulation

摘　　要：针对机械零件加工的柔性工艺路线规划的低碳高效多目标优化问题,根据零件加工过程碳排放和完工时间的影响因素,建立了以最少碳排放和最短完工时间为目标的工艺路线优化模型。采用三段式编码方式对柔性工艺路线进行描述,为了提高算法的局部搜索能力,将成绩标量函数值作为评价标准,采用模拟退火操作以改进NSGA-II (Non-dominated sorting genetic algorithm II)进行优化求解;同时将约束矩阵引入算法中,保证生成的工艺路线满足特征约束。以某型号导向轴支撑座的加工工艺为例,验证了所提模型和优化方法的可行性和有效性。结果表明,所提改进算法与传统NSGA-II算法相比,平均排放量和完工时间分别降低了4.3%和3.6%,该研究可对工艺路线的低碳高效多目标优化问题提供一定的参考。Aiming at the low carbon and high efficiency multi-objective optimization problem of flexible pro-cess route planning for mechanical parts processing, a process route optimization model aiming at the minimum carbon emission and the shortest completion time is established according to the factors affecting the carbon emission and completion time of the parts processing process. Three segment coding method is used to describe the flexible process route. In order to improve the local search ability of the algorithm, the score scalar function value is used as the evaluation standard, and simulated annealing operation is used to improve NSGA-II (Non dominated sorting genetic al-gorithm II) to optimize the solution;At the same time, the constraint matrix is introduced into the algorithm to ensure that the generated routing meets the feature constraints. Taking the machining process of a certain type of guide shaft support as an example, the feasibility and effectiveness of the proposed model and optimization method are verified. The results show that compared with the traditional NSGA-II algorithm, the average emissions and completion time of the improved algo-rithm are reduced by 4.3% and 3.6%, respectively. This study can provide some reference for the low-carbon and high-efficiency multi-objective optimization of process routes.

关 键 词：局部搜索能力 约束矩阵 特征约束 完工时间 多目标优化 工艺路线 机械零件加工 标量函数 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]