多工序组合气囊抛光策略智能优化  

Intelligent Optimization of Multi Process Combination Bonnet Polishing Strategy

作　　者：邱家军 

机构地区：[1]上海理工大学,机械工程学院,上海

出　　处：《建模与仿真》2023年第2期1388-1399,共12页Modeling and Simulation

摘　　要：针对进一步提高多工序气囊抛光效率和质量的技术需要,开展基于智能算法的多工序组合气囊抛光策略优化研究。以黄铜H62轴套为实验样件进行气囊粗抛与精抛实验,分析总结样件表面粗糙度变化规律;构建BP神经网络(Back propagation neural network, BP)模型,利用实验数据训练获得不同抛光工艺参数下表面粗糙度及抛光时间的映射关系;以获得最优表面粗糙度和抛光时间为目标,采用NSGA-II (Non-dominated sorting genetic algorithm-II, NSGA-II)多目标优化的方法优化多工序组合抛光中各工序的工艺参数,预测获得最佳转换工序的时间节点及最优多工序组合抛光参数。实验结果表明,当采用优化的多工序组合抛光策略进行抛光时,相比于优化后的单工序精抛所需总时长缩短了25%的时间,由16分钟下降至12分钟,而表面粗糙度Ra相比于优化后的单工序粗抛下降了0.317 μm。该研究针对多工序组合气囊抛光工艺优化需要,设计了基于NSGA-II多目标优化算法的多工序工艺参数优化策略,可以实现轴套零件多工序气囊抛光加工效率和质量的同步提升。In order to further improve the efficiency and quality of multi-process bonnet polishing, the opti-mization research of multi-process combination airbag polishing strategy based on intelligent algo-rithm is carried out. The brass H62 shaft sleeve was used as the experimental sample to carry out the air bag rough polishing and fine polishing experiments, and the change rules of the surface roughness of the sample were analyzed and summarized;The Back propagation neural network (BP) model is constructed, and the mapping relationship between surface roughness and polishing time under different polishing process parameters is obtained by training the experimental data;In order to obtain the optimal surface roughness and polishing time, the Non-dominated sorting genetic algorithm-II (NSGA-II) multi-objective optimization method is used to optimize the process parameters of each process in multi-process combination polishing, and predict the time node of the optimal conversion process and the optimal multi-process combination polishing parameters. Results: The experimental results show that when the optimized multi-process combination pol-ishing strategy is used for polishing, the total time required for the optimized single-process pol-ishing is reduced by 25%, from 16 minutes to 12 minutes, and the surface roughness Ra is reduced by 0.317 μm compared with the optimized single-process polishing. Aiming at the needs of mul-ti-process combination airbag polishing process optimization, this study designed a multi-process process parameter optimization strategy based on NSGA-II multi-objective optimization algorithm, which can achieve the synchronous improvement of multi-process airbag polishing efficiency and quality of shaft sleeve parts.

关 键 词：工序组合 抛光时间 单工序 气囊抛光 智能优化 BP神经网络 多目标优化 智能算法 

分 类 号：TG1[金属学及工艺—金属学]