基于有限元分析和机器学习的跌落所致封装结构力学行为预测

doi:10.16257/j.cnki.1681-1070.2021.0201

电子与封装 ›› 2021, Vol. 21 ›› Issue (2): 020204 . doi: 10.16257/j.cnki.1681-1070.2021.0201

所属专题：人工智能

基于有限元分析和机器学习的跌落所致封装结构力学行为预测

张筱迪¹,毛明晖¹,卢昶衡²,王文武¹,贾冯睿³,龙旭²

1. 辽宁石油化工大学土木工程学院、辽宁抚顺　113001；2. 西北工业大学力学与土木建筑学院，先进电子封装材料与结构研究中心，西安　710021；3. 浙江清华长三角研究院，浙江嘉兴　314006

收稿日期:2020-11-20 出版日期:2021-02-24 发布日期:2021-01-07
作者简介:张筱迪（1995—），女，陕西商洛人，研究生在读，研究方向为结构动态力学性能及其可靠性。

Prediction ofMechanical Behavior of Package Structure Subjected to Drop Impact Based onFinite Element Analysis and Machine Learning

ZHANG Xiaodi¹, MAO Minghui¹, LU Changheng², WANG Wenwu¹,

JIA Fengrui3,LONG Xu2

Received:2020-11-20 Online:2021-02-24 Published:2021-01-07

摘要/Abstract

摘要： 目前电子封装行业中，在进行封装结构力学可靠性研究时需要开展大量的有限元仿真分析，存在仿真模型建模流程复杂且计算过程漫长的常见问题。鉴于此方面的技术瓶颈，首先使用ABAQUS有限元软件对封装结构跌落过程动力响应进行数值模拟并获取特征候选值，建立了4×3的以抗跌落可靠性评估的关键特征变量为输入特征值和以应力和等效塑性应变为输出特征值的数据组。在有限元分析结果基础上，应用相关性驱动神经网络的机器学习方法对数据组进行训练，进而得到相应预测模型。最后，通过与数值模拟结果进行对比，验证了所提出神经网络预测结果与有限元模拟结果具有很好的吻合度。结果表明，基于有限元仿真模型的机器学习方法，在预测封装结构复杂工况下力学性能可靠性方面具有巨大的潜力。

关键词: 封装结构, 力学可靠性, 有限元模拟, 机器学习, 动力响应

Abstract: In the electronic packaging industry, the evaluation of mechanical reliability of the packaging structures is usually studied by means of finite element simulations, which is however considerably time-consuming due to modeling creation and iterative analysis. To eliminate these technical limitations, the dynamic response of the packaging structure during the drop process was firstly simulated by ABAQUS in the present study, and the values of feature candidates were obtained to construct a 4×3 data set. In fact, the main feature variables of the drop resistance evaluation is taken as the input features, while the stress and equivalent plastic strain are taken as the output feature values. Consequently, the data set was trained by the correlation-driven neural network to obtain the corresponding machine-learning based prediction model. Finally, compared with the finite element simulation, the predicted results of machine-learning based neural network are similar to those of finite element simulation results. The results of this paper show that the machine learning method based on finite element simulations has great potential in predicting the reliability of the mechanical performance of the packaging structure under complicated working conditions.

Key words: packagingstructure, mechanicalreliability, finiteelementmethod, machinelearning, dynamicresponse

中图分类号:

TN305.94

张筱迪,毛明晖,卢昶衡,王文武,贾冯睿,龙旭. 基于有限元分析和机器学习的跌落所致封装结构力学行为预测[J]. 电子与封装, 2021, 21(2): 020204 .

ZHANG Xiaodi, MAO Minghui, LU Changheng, WANG Wenwu, JIA Fengrui, LONG Xu. Prediction ofMechanical Behavior of Package Structure Subjected to Drop Impact Based onFinite Element Analysis and Machine Learning[J]. Electronics & Packaging, 2021, 21(2): 020204 .

参考文献

[1] LIM C T , TEO Y M , SHIM V P W . Numerical simulation of the drop impact response of a portable electronic product[J]. IEEE Transactions on Components and Packaging Technologies, 2002, 25(3):478-485.
[2] WONG E H , MAI Y W , SEAH S K W . Board Level Drop Impact—Fundamental and Parametric Analysis[J]. Journal of Electronic Packaging, 2005, 127(4):496-502.
[3] VARGHESE J , DASGUPTA A . Test methodology for durability estimation of surface mount interconnects under drop testing conditions[J]. Microelectronics Reliability, 2007, 47(1):93-103.
[4] TEE T Y , NG H S , LIM C T , et al. Impact life prediction modeling of TFBGA packages under board level drop test[J]. Microelectronics Reliability, 2004, 44(7):1131-1142.
[5] CREMER J , KONSTANTELOS I , STRBAC G . From optimization-based machine learning to interpretable security rules for operation[J]. IEEE Transactions on Power Systems, 2019, 34(5):3826-3836.
[6] LABED I, LABED D. Extreme learning machine-based alleviation for overloaded power system[J]. IET Generation, Transmission & Distribution, 2019, 13(22):5058-5070.
[7] POZO B, GARATE J I, Ferreiro S, et al. Supercapacitor electro-mathematical and machine learning modelling for low power applications[J]. Electronics, 2018, 7(4):44.
[8] ZHANG D, HAN X, DENG C. Review on the research and practice of deep learning and reinforcement learning in smart grids[J]. CSEE Journal of Power and Energy Systems, 2018, 4(3):362-370.
[9] SAMAVATIAN V, FOTUHI-FIRUZABAD M, SAMAVATIAN M, et al. Correlation-driven machine learning for accelerated reliability assessment of solder joints in electronics[J]. Scientific reports, 2020, 10(1):1-14.
[10] DING Z, PASCAL E, DE GRAEF M. Indexing of electron back-scatter diffraction patterns using a convolutional neural network[J]. Acta Materialia, 2020, 199:370-382.
[11] YI S, JONES R. Machine learning framework for predicting reliability of solder joints[J]. Soldering & Surface Mount Technology, 2019 19(1):28-33.
[12] SHIRAIWA T, BRIFFOD F, MIYAZAWA Y, et al. Fatigue performance prediction of structural materials by multi-scale modeling and machine learning[C]//Proceedings of the 4th World Congress on Integrated Computational Materials Engineering (ICME 2017). Springer, Cham, 2017:317-326.
[13] LONG X, HE X, YAO Y. An improved unified creep-plasticity model for SnAgCu solder under a wide range of strain rates[J]. Journal of Materials Science, 2017, 52(10):6120-6137.
[14] CHENG H C, CHENG H K, LU S T, et al. Drop impact reliability analysis of 3-D chip-on-chip packaging: Numerical modeling and experimental validation[J]. IEEE Transactions on Device and Materials Reliability, 2013, 14(1): 499-511.

中国半导体行业协会封装分会会刊

中国电子学会电子制造与封装技术分会会刊

基于有限元分析和机器学习的跌落所致封装结构力学行为预测

Prediction ofMechanical Behavior of Package Structure Subjected to Drop Impact Based onFinite Element Analysis and Machine Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	梁达平,赵玉祥,张进兵. 基于SECS/GEM协议的芯片烘箱设备智能故障诊断算法设计^*[J]. 电子与封装, 2024, 24(7): 70501-.
[2]	梁达平;赵利民;王鸿斌. 基于贝叶斯学习的PID温控算法在芯片烘箱中的应用^*[J]. 电子与封装, 2022, 22(8): 80202-.
[3]	王美玉, 胡伟波, 孙晓冬, 汪青, 于洪宇. 功率电子封装关键材料和结构设计的研究进展^*[J]. 电子与封装, 2021, 21(10): 100109-.
[4]	丁荣峥¹，马国荣²，张玲玲³，邵康⁴. 一种半导体探测器气密性封装结构和工艺优化[J]. 电子与封装, 2017, 17(12): 1-4.