基于MOGWO的45#钢表面激光抛光工艺参数多目标优化  

Multi-objective Optimization of Laser Polishing Process Parameters for the Surface of 45#Steel Based on MOGWO

作　　者：梁强 徐永航 李永亮 王敬 杜彦斌[1,2] LIANG Qiang;XU Yonghang;LI Yongliang;WANG Jing;DU Yanbin(School of Mechanic Engineering,Chongqing Technology and Business University,Chongqing 400067,China;Chongqing Key Laboratory of Green Design and Manufacturing of Intelligent Equipment,Chongqing Technology and Business University,Chongqing 400067,China)

机构地区：[1]重庆工商大学机械工程学院,重庆400067 [2]重庆工商大学智能装备绿色设计与制造重庆市重点实验室,重庆400067

出　　处：《表面技术》2024年第10期173-182,共10页Surface Technology

基　　金：重庆市自然科学基金面上项目(CSTB2022NSCQ-MSX0473);重庆市高校创新研究群体资助项目(CXQT21024);制造装备机构设计与控制重庆市重点实验室开放课题(KFJJ2019078);重庆工商大学研究生创新型科研资助项目(YJSCXX2023-211-54)。

摘　　要：目的提高45#钢表面激光抛光后的成形质量,提出一种激光抛光工艺参数多目标优化方法。方法构建基于功率、扫描速度、搭接距离的三因素三水平激光抛光试验,并分别应用粗糙度测量仪、显微硬度计和超景深三维显微镜测试抛光层的粗糙度、显微硬度和抛光层深度。基于试验数据,分别应用指数模型和二阶响应面模型构建抛光工艺参数与表面粗糙度、显微硬度、抛光深度的回归预测模型,并对2种模型的预测精度进行对比分析。采用多目标灰狼优化算法(MOGWO)结合优劣解距离法(TOPSIS)-CRITIC综合评价决策体系对抛光工艺参数进行寻优和多属性决策。结果二阶响应面模型具有更高的预测精度,能够更好地反映激光抛光工艺参数与各响应目标之间的映射关系。当功率为113W、扫描速度为3m/min、搭接距离为0.13 mm时,粗糙度值Ra从11.563μm降至5.713μm,降幅为50.59%,显微硬度从185.9HV0.5升至364.7HV0.5,升幅为96.18%,此时的抛光深度为0.051 mm,最大相对误差为7.84%。结论此方法可以为其他金属材料表面激光抛光质量预测模型的构建及工艺参数寻优提供借鉴。To improve the forming quality of 45 steel surfaces after laser polishing,the work aims to propose a multi-objective optimization method of laser polishing process parameters.Laser power,scanning speed,and overlap distance were taken as process parameters,and surface roughness,microhardness,and polishing depth were taken as evaluation indexes to construct a 3-factor and 3-level laser polishing experiment.Before the experiment,the plate was ground flat and processed with a ball milling cutter to produce a texture and then ultrasonically cleaned with anhydrous ethanol and dried,and the zigzag scanning trajectory was used to carry out the laser polishing experiment.A roughness meter was used to measure the surface roughness of the polished surface before and after laser polishing,a microhardness tester was used to measure the microhardness of the polished layer of the material before and after laser polishing,and a super depth-of-field 3D microscope was used to measure the polishing depth after laser polishing.Based on the experimental data,the exponential model and the second-order response surface model were used to construct the regression prediction models of the laser polishing process parameters and the surface roughness,microhardness,and polishing depth regarding the construction method of the prediction model of the geometrical characteristics of the laser cladding layer.By comparing the correlation coefficient R,determination coefficient R2,and determination adjustment coefficient 2 Radj with the significance test of the two models,as well as comparing the correlation between the experimental values and the predicted values of the two models,it was obtained that the second-order response surface model had a higher prediction accuracy,and it could better reflect the mapping relationship between the laser polishing process parameters and the response targets.The main effect analysis was used to study the effect law of each process parameter of laser polishing on the surface roughness,microhardness,and polishi

关 键 词：激光抛光 二阶响应面模型 MOGWO算法 TOPSIS-CRITIC 多目标优化 

分 类 号：TN249[电子电信—物理电子学] TG356.28[金属学及工艺—金属压力加工]