近年來隨著電子科技技術發展快速，電子產品逐漸趨向於輕、薄、短小等方便攜帶的趨勢，且消費者對於電子產品的功能要求也越來越嚴苛，為了因應此趨勢，除了IC設計上的開發外，電子封裝技術也從傳統的DIP(Dual In-line Packaging)逐漸發展成擁有高密度I/O的覆晶(Flip Chip)結構、晶片級封裝(Chip Scale Packaging)、晶圓級封裝(Wafer Level Package)、扇出型(Fan out)、3D堆疊(3D Stacked)、系統式封裝(System in Package)等結構。
熱循環負載實驗為確保電子元件量產品質的重要測試，然而實際熱循環負載實驗需要花費許多時間與成本，導致產品週期過長或成本過高，因此現今在設計階段多採用有限單元分析模擬取代大部分的實驗以提高140產品競爭力。但在有限元素分析模擬中，不同研發人員所得到的結果可能會有所差異，且模擬仍然會耗費大量的計算時間。隨著電腦硬體設備的蓬勃發展，快速的運算速度與龐大的儲存空間，促使機器學習、人工智慧越來越普及於日常生活中。
本研究目的為透過可信賴的有限分析模擬步驟，採用Coffin-Manson壽命預估公式，對於WLCSP封裝結構，建立可靠的訓練資料庫。當材料固定、負載固定情形下，給定不同的幾何結構參數值，透過支援向量回歸(Support Vector Regression)模型，即可立即獲得預測壽命，並且有一定的準確度，藉此減少有限分析模擬次數，或給予前端設計階段一個參考依據，此外，依據此資料庫嘗試改進SVR與RF的預測表現

Today, electronic devices are developing rapidly, moving toward lighter, thinner and smaller. However, consumers want to have a versatile electronic product. In order to meet these needs. Not only new developments in IC design, but also electronic packaging technology is also a solution. Electronic packaging technology has evolved from traditional packaging technologies such as dual in-line package (DIP) to advanced technologies with high I/O density, such as Flip Chip structures, Chip Scale packages, Wafer Level packages, Fan-out structures, 3D stacking structure and System in Package structure, etc.
Thermal cycling test is one of the most important tests to ensure the quality of electronic components. However, the thermal cycling test will take a lot of time and cost. It will lead to a long time to market and high cost of electronic products. Therefore, many companies use finite element simulation to reduce the number of TCT real experiments, which will increase product competitiveness. However, in finite element analysis, the results of different researchers may be different, and it still takes some time to run the simulation. With the development of computer equipment, high computing speed and numerous storage spaces make machine learning and artificial intelligence more and more common in daily life.
The purpose of this study was to establish a database based on reliable finite analysis simulation steps and to use the Coffin-Manson life prediction model. Then train the SVR prediction model. For the WLCSP structure, when we fix the material properties and loading conditions, enter different geometric parameter values, we can obtain the reliability immediately through the support vector regression model, and the difference between the simulation and the SVR must be within an acceptable range. I hope to reduce the number of finite element simulations with SVR, or give the front-end designer a reference. In addition, based on this database, try to improve the predictive performance of SVR and RF

摘要 I
Abstract II
目錄 IV
圖目錄 VII
表目錄 XIV
第一章 緒論 1
1.1 研究動機 1
1.2 文獻回顧 2
1.3 研究目標 5
第二章 基礎理論 6
2.1 錫球外型預測 6
2.2 有限單元法基礎理論 9
2.2.1 線彈性有限單元理論 9
2.2.2 材料非線性理論 13
2.2.3 數值方法及收斂準則 15
2.3 硬化法則 17
2.3.1 等向硬化法則 17
2.3.2 動態硬化法則 18
2.4 Chaboche 模型 19
2.5 封裝結構可靠度之預測方法 21
2.5.1 Coffin-Manson應變法 21
2.5.2 Darveaux 能量密度法 22
2.5.3 修正型能量密度法 22
2.6 機器學習 23
2.6.1 監督式學習(Supervised Learning) 24
2.6.2 非監督式學習(Supervised Learning) 26
2.6.3 強化學習(Reinforcement Learning) 27
2.6.4 資料前處理(Data Preprocessing) 28
2.6.5 人工類神經網路(Artificial Neural Network) 30
2.6.6 CART決策樹(Classification and Regression Tree) 33
2.6.7 支援向量機(Support Vector Machine) 35
2.6.8 支援向量回歸(Support Vector Regression) 50
2.7 回歸模型效能評估 55
第三章 有限單元模型之建立與驗證 57
3.1 有限單元模型基本假設 58
3.2 材料參數之設定 63
3.3 邊界條件設定 64
3.4 溫度負載設定 64
3.5 ANSYS模擬之驗證 65
第四章 結果驗證與討論 68
4.1 支援向量回歸 69
4.1.1 支援向量回歸之參數 69
4.1.2 集成學習(Ensemble Learning) 80
4.1.3 資料分群 90
4.1.4 多核心之支援向量回歸 94
4.2 決策樹與隨機森林 115
4.2.1 隨機森林參數探討 115
4.2.2 決策樹與隨機森林之預測表現 117
4.2.3 特徵重要度 119
第五章 結論與未來工作 129
參考文獻 131
附錄 135

[1] C.M. Liu, C.C. Lee and K.N. Chiang, “Enhancing the reliability of wafer level packaging by using solder joints layout design”, IEEE Transactions on Component and Packaging Technologies, vol. 29, pp. 877-885, Dec. 2006.
[2] L. F. Coffin, "A study of the effects of cyclic thermal stresses on a ductile metal”, Tarnsactions-American Society of Mechanical Engineers Journal of Electronic Packaging, vol. 76, pp. 931-950, 1954.
[3] S. S. Manson, “Behavior of materials under conditions of thermal stress”, National Advisory Committee for Aeronautics-Techncal Note 2933, pp. 317-350, 1953.
[4] R. Darveaux, K. Banerji, A Mawer, and G. Dody, “Reliability of plastic ball grid array assembly”, Ball Grid Array Technology, New York, 1995.
[5] R. Darveaux, “Effect of simulation methodology on solder joint crack growth correlation”, 2000 Proceedings. 50th Electronic Components and Technology Conference (Cat. No.00CH37070), Las Vegas, NV, USA, pp. 1048-1058, May 21-24, 2000
[6] 吳凱強, “先進封裝錫球接點於不同溫度循環負載速率之下之可靠度評估”, 碩士論文, 國立清華大學動力機械工程學系, 2016
[7] C. Y. Tsou, T. N Chang, K. C. Wu, P. L. Wu and K.N. Chiang, “Reliability Assessment using Modified energy based model for WLCSP Solder Joints”, ICEP2017, Yamagata, Japan, April 19-22, 2017.
[8] K. N. Chiang, H. C. Cheng, and W. H. Chen, “Large-Scaled 3-D area array electronic packaging analysis”, Journal of Computer Modeling and Simulation in Engineering, Vol. 4, No.1, pp. 4-11, 1999.
[9] J.M. Weaver, "A course in Design for Quality", Frontiers in Education Conference 1997. 27th Annual Conference. Teaching and Learning in an Era of Change. Proceedings., vol. 2, pp. 968-971 vol.2, Nov. 5-8 1997
[10] W. S. McCulloch, and W. Pitts, “A logical calculus of the ideas immanent in nervous activity”, The bulletin of mathematical biophysics, Vol.5, Issue 4, pp 115–133, Dec., 1943.
[11] 周佩勳, “以人工神經網路回歸模型評估晶圓級封裝之可靠度”, 碩士論文, 國立清華大學動力機械工程學系, 2019
[12] J. R. Quinlan, “Induction of Decision Trees”, Machine Learning, Vol.1, Issue 1, pp. 81–106, 1986.
[13] T. K. Ho, “Random Decision Forest”, Proceedings of 3rd International Conference on Document Analysis and Recognition, Montreal, Que., pp. 278-282 vol.1, 1995
[14] 蕭翔云, “以隨機森林回歸模型評估晶圓級封裝之可靠度”, 碩士論文, 國立清華大學動力機械工程學系, 2019
[15] C. Cortes, and V. Vapnik, “Support-Vector Networks”, Machine Learning, Vol. 20, Issue 3, pp. 273–297, 1995.
[16] T. Benkedjouh, K. Medjaher, N. Zerhouni, and S. Rechak, “Health assessment and life prediction of cutting tools based on support vector regression”, Journal of Intelligent Manufacturing, Volume 26, Issue 2, pp. 213-223, April, 2015
[17] A. Aljouie, and U. Roshan, “Prediction of Continuous Phenotypes in Mouse, Fly, and Rice Genome Wide Association Studies with Support Vector Regression SNPs and Ridge Regression Classifier”, 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA), Miami, FL, pp. 1246-1250, Dec. 09-11, 2015
[18] A. Rabe, S. van der Linden and P. Hostert, "Simplifying Support Vector Machines for Regression analysis of hyperspectral imagery," 2009 First Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, Grenoble, pp. 1-4, Aug. 26-28, 2009
[19] Y. Xia, Y. Liu and Z. Chen, "Support Vector Regression for prediction of stock trend," 2013 6th International Conference on Information Management, Innovation Management and Industrial Engineering, Xi'an, China, pp. 123-126, Nov. 23-24, 2013
[20] L. S. Goldmann, “Geometry optimization of controlled collapse interconnections”, IBM Journal of Research and Development, Vol. 13, pp. 251-265, May, 1969.
[21] S. M. Heinrich, M. Schaefer, S. A. Schroeder, and P. S. Lee, “Peidiction of solder joint geomertry on array-type interconnections”, American Society of Mechanical Engineers Journal of Electronic Packaging, Vol. 118, pp. 114-121, Sep. 01, 1996.
[22] K. A. Brakke, “Surface Evolver Manual”, version 2.01 Minneapolis, MN: The Geometry Center, 1996.
[23] K. J. Bathe, Finite Element Procedures in Engineering Analysis: Prentice Hall, 1982
[24] S. Timoshenko, Theory of Elasticity: Mcgraw-Hill College, 1970.
[25] W. N. Findley, J. S. Lai, and K. Onaran, “Creep and relaxation of nonlinear viscoelastic materials”, with an introduction to linear viscoelasticity: Amsterdam ; New York : North-Holland Pub. Co. : sole distributors for the U.S.A. and Canada, Elsevier/North Holland, 1976.
[26] R. D. Cook, D. S. Malkus, M. E. Plesha, and R. J. Witt, “Concepts and Applications of Finite Element Analysis”, Wiley; 4th edition, 2001.
[27] J. Chakrabarty, “Theory of Plasticity”, Butterworth-Heinemann, 3th edition, 2006.
[28] N. E. Dowling, “Mechanical Behavior of Materials: Engineering Methods for Deformation”, Fracture, and Fatigue, Upper Saddle River, New Jersey: Prentice-Hall, Inc, 1999.
[29] J. L. Chaboche, “Constitutive Equations for Cyclic Plasticity and Cyclic Viscoplasticity”, International Journal of Plasticity, Vol. 5, pp. 247-302, 1989.
[30] J. L. Chaboche, “On Some Modifications of Kinematic Hardening to Improve the Description of Ratchetting Effect”, International Journal of Plasticity, Vol. 7, pp. 661-678, Jan. 01, 1991.
[31] L. F. Coffin, “A study of the effects of cyclic thermal stress on a ductile metal”, Transactions ASME, Vol. 76, pp. 931-950, 1954.
[32] C. Y. Yeh, C. W. Huang, and S. J. Lee, “A Multiple-Kernel Support Vector Regression Approach for Stock Market Price Forecasting”, Expert Systems with Applications, Vol.38, Issue 3, pp. 2177-2186, Mar. 01, 2011.
[33] G. R. G. Lanckriet, N. Cristianini, P. Bartlett, L. E. Ghaoui, and M. I. Jordan, “Learning the kernel matrix with semidefinite programming”, Journal of Machine Learning Research, vol. 5, pp. 27-72, Jan. 27-72, 2004
[34] O. Chapelle, V. Vapnik, O. Bousquet, and S. Mukherjee, “Choosing Multiple Parameters for Support Vector Machines”, Machine Learning, Vol. 46, no. 1-3, pp. 131-159, Jan., 2002.
[35] V. Vapnik, The Nature of Statistical Learning Theory, New York, Springer, pp. 90, 1995
[36] M. Motalab, M. Mustafa, J. C. Suhling, J. Zhang, J. Evans, M. J. Bozack, and P. Lall, “Thermal Cycling Reliability Predictions for PBGA Assemblies That Include Aging Effects”, ASME 2013 International Technical Conference and Exhibition on Packaging and Integration of Electronic and Photonic Microsystems, San Francisco, USA, July 16-18, 2013
[37] V. Cherkassky, and Y. Ma, ” Practical selection of SVM parameters and noise estimation for SVM regression”, Neural Networks, Volume 17, Issue 1, pp. 113-126, 2004
[38] H. Drucker, “Improving Regressors using Boosting Techniques”, ICML, Jul 8, 1997
[39] S. Qiu and T. Lane, “A Framework for Multiple Kernel Support Vector Regression and Its Applications to siRNA Efficacy Prediction”, IEEE/ACM Transactions on Computational Biology and Bioinformatics, Volume 6, no. 2, pp. 190-199, April-June, 2009.
[40] M. Gönen and E. Alpaydın, “Multiple Kernel Learning Algorithms”, The Journal of Machine Learning Research, Volume 12, pp. 2211-2268, Feb. 01, 2011