超硬材料工程2024，Vol.36Issue(4) ：15-19,35.

Al2O3陶瓷基板切割用金刚石划片刀选型方法研究——基于高斯过程回归模型

Research on the selection method of diamond dicing knife for Al2O3 ceramic substrate cutting——Regression model based on Gaussian process

佘凤芹 张迪 王战
超硬材料工程2024，Vol.36Issue(4) ：15-19,35.
✕
  • 万方数据

Al2O3陶瓷基板切割用金刚石划片刀选型方法研究——基于高斯过程回归模型

Research on the selection method of diamond dicing knife for Al2O3 ceramic substrate cutting——Regression model based on Gaussian process

佘凤芹 1张迪 2王战3
扫码查看

作者信息

  • 1. 郑州大学管理学院,河南郑州 450001;郑州磨料磨具磨削研究所有限公司,河南郑州 450001
  • 2. 郑州宝驰迪科技有限公司,河南郑州 450001
  • 3. 郑州磨料磨具磨削研究所有限公司,河南郑州 450001
  • 折叠

摘要

为改善Al2O3陶瓷基板切割过程中的背崩问题,通过选取金刚石划片刀划切过程中会对背崩产生显著影响的磨料粒度、磨料浓度、结合剂硬度和冷水槽数量等4个特征参数作为关键因子,将陶瓷基板背崩尺寸作为响应变量,利用正交试验设计在因子可行域内选择样本数据,并采用高斯过程回归模型进行建模,最后用粒子群算法进行寻优.结果表明:该方法对电镀结合剂金刚石划片刀选型有较好的指导意义,验证结果与算法寻优结果仅相差0.5 μm.

Abstract

To solve the back-chipping in the cutting process of Al2O3 ceramic substrates,four characteristic parameters with significant impact on back chipping during the cutting process with diamond blade were selected as key factors,including abrasive particle size,abrasive concentration,binder hardness and the number of cold water tanks.The back chipping size of ceramic substrates was taken as the response variable.Sample data were selected within the feasible domain of factors by orthogonal experimental design.Gaussian process regression model was used for modeling and particle swarm optimization algorithm was used for optimization.The results show that this method has good guiding significance for the selection of diamond blade with electroplated binder,and the difference between the verification results and the optimization results of the algorithm is only 0.5 µm.

关键词

LED封装/氧化铝陶瓷基板/电镀结合剂/正交试验设计/粒子群优化算法

Key words

LED packaging/Alumina Ceramic substrate/electroplated binder/orthogonal experimental design/particle swarm optimization algorithm

引用本文复制引用

出版年

2024
超硬材料工程
桂林矿产地质研究院

超硬材料工程

影响因子：0.201
ISSN：1673-1433
段落导航相关论文