[1]胡子翔,何 旭,朱凯鹏.基于 Boosting-MKELM 的回流焊过程质量预测研究[J].机械与电子,2023,41(05):12-18.
　HU Zixiang,HE Xu,et al.Quality Prediction Research of Reflow Soldering Process Based on Boosting-MKELM[J].Machinery & Electronics,2023,41(05):12-18.
点击复制

基于 Boosting-MKELM 的回流焊过程质量预测研究()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
41
期数:
2023年05期
页码:
12-18
栏目:
设计与研究
出版日期:
2023-05-25

文章信息/Info

Title:
Quality Prediction Research of Reflow Soldering Process Based on Boosting-MKELM
文章编号:
1001-2257 ( 2023 ) 05-0012-07
作者:
胡子翔 1 ; 2 ; 何 旭 1 ; 朱凯鹏 1
1. 中国电子科技集团公司第三十八研究所,安徽 合肥 230088 ; 2. 国家级工业设计中心(中电 38 所),安徽 合肥 230088
Author(s):
HU Zixiang1 ;  2 ;  HE Xu1 ;  ZHU Kaipeng1
( 1.No.38 Research Institute of CETC , Hefei 230088 , China ; 2.National Industrial Design Center ( CETC38 ), Hefei 230088 , China )
关键词:
回流焊; 钽电容; 质量预测; Boosting-MKELM
Keywords:
reflow soldering ;  tantalum capacitors ;  quality prediction ;  Boosting-MKELM
分类号:
TN405
文献标志码:
A
摘要:
通过建立基于钽电容仿真模型的代理模型,实现对其回流焊过程的质量预测。以锡膏尺寸中的最小末端连接宽度、最小侧面连接长度、最小填充高度和焊料填充厚度作为钽电容回流焊过程中的关键质量指标,运用 Boosting-MKELM 算法建立质量指标预测模型;并以 ERMS 和 R2?作为模型性能评估指标,通过与传统代理模型算法对比,验证算法的有效性及性能。结果表明,相比 PRS 模型、 SVM 模型及 RBF 模型,Boosting-MKELM所预测的 4 个质量指标都 拥有最小的 ERMS 和 R2 ,说 明 所 建 立的 Boosting-MKELM 预测模型具有更高的精度,可更好地表征输入变量与质量指标之间的映射关系。
Abstract:
By establishing a surrogate model based on tantalum capacitance simulation model , the quality prediction of the reflow process is realized.The minimum end connection width , minimum side connection length , minimum filling height and solder filling thickness in the solder paste size are taken as the key quality indicators in the process of tantalum capacitor reflow soldering , and a quality indicator prediction model?is established by using Boosting-MKELM algorithm ;?ERMS?and?R2?are taken as the performance evaluation indicators of the model , and the effectiveness and performance of the algorithm are verified by comparing with the traditional agent model algorithm.The results show that , compared with PRS model , SVM model and RBF model , the four quality indicators predicted by Boosting-MKELM all have the smallest?ERMS?and?R2 , indicating that the established Boosting-MKELM prediction model has higher accuracy and can better characterize the mapping relationship between input variables and quality indicators.

参考文献/References:

[ 1 ] 李荣茂 . 浅谈电子封装技术的重要性及发展趋势[ J ] .科技传播,2010 ( 23 ): 253 , 251.

[ 2 ] 张宝林 . 低温烧结热稳定 BaTiO3 基陶瓷介质材料研究[ D ] . 天津:天津大学,2013.
[ 3 ] 吕利叶 . 先进代理模型方法与应用研究[ D ] . 大连:大连理工大学,2020.
[ 4 ] HUANG G B , CHEN L , SIEW C K.Universal approximation using incremental constructive feedforward networks with random hidden nodes [ J ] .IEEE Transactions on neural networks , 2006 , 17 ( 4 ): 879-892.
[ 5 ] 董志奎 . 基于深度学习算法的短期交易电价预测[ D ] .广州:华南理工大学,2020.
[ 6 ] 连克强 . 基于 Boosting 的集成树算法研究与分析[ D ] .北京:中国地质大学(北京),2018.
[ 7 ] 孙曙光,唐尧,王景芹,等 . 基于多信号特征融合的断路器操作机构寿命预测[ J ] . 高电压技术, 2022 , 48 ( 11 ):4455-4468.
[ 8 ] TIBSHIRANI R.Regression shrinkage and selection via the Lasso [ J ] .Journal of the royal statistical society: series B( methodological ),1996 , 58 ( 1 ): 267-288.
[ 9 ] 粟华 . 飞行器高拟真度多学科设计优化技术研究[ D ] .西安:西北工业大学,2014.
[ 10 ] LOEPPKY J L , SACKS J , WELCH W J .Choosing the sample size of a computer experiment : a practical guide [ J ] .Technometrics , 2009 , 51 ( 4 ): 366-376.
[ 11 ] 王学智 . 基于深度学习的土壤温湿度预测相关技术研究[ D ] . 吉林:吉林大学,2022.

相似文献/References:

[1]范子铭,田富君,胡子翔,等.基于机器学习的回流焊焊点形貌预测[J].机械与电子,2023,41(01):53.
　FAN Ziming,TIAN Fujun,HU Zixiang,et al.Topography Prediction of Reflow Solder Joints Based on Machine Learning[J].Machinery & Electronics,2023,41(05):53.

备注/Memo

备注/Memo:
收稿日期: 2022-09-29
基金项目:国防基础科研项目( JCKY2019210B006 , JCKY2020210B007 )
作者简介:胡子翔 ( 1985- ),男,湖北十堰人,博士,高级工程师,研究方向为电子设备数字化设计与制造。
更新日期/Last Update: 2023-05-23