跳到主要內容

臺灣博碩士論文加值系統

(44.210.85.190) 您好!臺灣時間:2022/12/10 13:18
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:陳文平
研究生(外文):Chen Wenping
論文名稱:以計算流體力學方法建立類神經網路預測多熱源晶片溫度分佈之研究
論文名稱(外文):Integrating Computational Fluid Dynamics and Neural Networks to Predict Temperature Distribution of the Semiconductor Chip with Multiple Heat Sources
指導教授:連信仲連信仲引用關係鄒宏基
指導教授(外文):Hsin-Chung LienHong-Kee Tzou
學位類別:碩士
校院名稱:北台科學技術學院
系所名稱:機電整合研究所
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2006
畢業學年度:94
語文別:中文
論文頁數:120
中文關鍵詞:倒傳遞類神經網路計算流體力學熱管理半導體晶片
外文關鍵詞:Back-propagation Neural NetworkIntegrating Computational Fluid Dynamicshot managementsemiconductorchip
相關次數:
  • 被引用被引用:1
  • 點閱點閱:246
  • 評分評分:
  • 下載下載:62
  • 收藏至我的研究室書目清單書目收藏:3
本論文是以結合計算流體力學(Integrating Computational Fluid Dynamics, CFD)及倒傳遞類神經網路(Back-propagation Neural Network, BNN)預測多熱源晶片溫度分佈。
本文藉由CFD建立100組之訓練資料,每一筆資料包含70組數據,前6項數據為晶片內部3個熱源座標點,同時並作為BNN之輸入,後64項資料為晶片上規劃64個溫度監控點,此資料做為BNN之輸出,再藉由倒傳遞類神經網路的學習能力,建立輸入座標資料與輸出溫度值的網路對應關係,以產生倒傳遞類神經網路預測多熱源晶片分佈之數學模型。最後再建立16組CFD分析模型數據資料庫作為測試數學模型數據資料庫,實驗結果顯示訓練完成BNN數學模型可以由97%準確的預估不同的座標位置,所產生出溫度分佈以及最高溫度,值同時BNN預估熱源晶片約為CFD之 倍效能,可有效的提供業界快速預測與最佳化分析應用。
This thesis describes the combination of Integrating Computational Fluid Dynamics (CFD) with Back-propagation Neural Network (BNN) to predict multiple heat source chip temperature profiles.
This text presents 100 groups of training data by CFD. Every data included 70 groups of statistic, and the first 6 groups are 3 heat source seat punctuation marks within the chip as the input of BNN at the same time. On the other hand, the last 64 groups are the 64 temperature control points on the chip as the output of BNN. According to the learning ability of the BNN, the network corresponding relation between coordinates data input and temperature value output is established. It caused the mathematics model by BNN predicting that multiple heat source chips distribution. Finally, set up 16 groups of statistic database analyzed by CFD as the database testing the mathematics model. The result shows that BNN mathematics model with training finishes can be produced the temperature distribution and maximum temperature by estimating different coordinates position with 97% accuracy. The value of the heat source chip estimated by BNN is more efficient than CFD about times. It can offer industries effectively that speedy prediction and the best analytic application.
摘中文要 ………………………………………………………………… i
摘英文要 ………………………………………………………………… ii
目 錄 ………………………………………………………………… iii
圖 目 錄 ………………………………………………………………… v
表 目 錄 ………………………………………………………………… vi
第一章 緒論…………………………………………………………… 1
1-1 研究動機與目的……………………………………………… 1
1-2 文獻回顧……………………………………………………… 3
1-3 本文架構……………………………………………………… 5
第二章 FC-PBGA封裝簡介…………………………………………… 6
2-1 FC-PBGA簡介………………………………………………… 6
2-2 FC-PBGA優點………………………………………………… 7
第三章 倒傳遞神經網路及計算流體力學簡介……………………… 9
3-1 類神經網路…………………………………………………… 9
3-1-1 倒傳遞神經網路……………………………………………… 10
3-2 計算流體力學………………………………………………… 13
3-2-1 熱傳遞學……………………………………………………… 13
3-2-2 分析軟體介紹………………………………………………… 14
第四章 計算流體力學分析及倒傳遞網路預測實驗………………… 15
4-1 分析實驗流程………………………………………………… 15
4-2 建立計算流體力學模型……………………………………… 16
4-2-1 建立熱源座標位置…………………………………………… 17
4-2-3 FC-PBGA模型分析及溫度監控取樣………………………… 19
4-3 倒傳遞神經網路架構………………………………………… 26
4-3-1 倒傳遞神經網路數學模型建立……………………………… 26
4-3-2 倒傳遞神經網路訓練過程…………………………………… 28
4-3-3 倒傳遞神經網路訓練結果…………………………………… 30
4-4 倒傳遞神經網路測試實驗…………………………………… 33
4-4-1 倒傳遞神經網路測試過程…………………………………… 35
4-4-2 倒傳遞神經網路測試結果…………………………………… 37
4-5 倒傳遞神接網路與計算流體力學數據比較………………… 40
第五章 結論與未來展望……………………………………………… 52
5-1 結論…………………………………………………………… 52
5-2 未來展望……………………………………………………… 53
參考文獻 ………………………………………………………………… 54
附錄A 訓練組計算流體力學分析資料……………………………… 62
附錄B 訓練組倒傳遞神經網路預測資料…………………………… 72
附錄C BNN與ICEPAK訓練組溫度比較圖………………………… 82
附錄D BNN與ICEPAK測試組溫度比較圖………………………… 95
附錄E 計算流體力學及倒傳遞神經網路訓練組3D溫度分佈圖…… 98
附錄F 計算流體力學及倒傳遞神經網路測試組3D溫度分佈圖…… 111

圖目錄
圖1.1 本研究實驗架構圖…………………………………………… 5
圖3.1 倒傳遞神經網路架構圖……………………………………… 11
圖4.1 計算流體力學分析模型……………………………………… 17
圖4.2 定義晶片內熱源位置起點…………………………………… 18
圖4.3 第1組訓練組晶片內熱源位置座標………………………… 19
圖4.4 分析模型熱傳遞係數………………………………………… 20
圖4.5 第1組座標資料經ICEPAK分析熱溫度分佈……………… 20
圖4.6 晶片表面溫度監控溫度規劃圖……………………………… 21
圖4.7 取樣溫度座標定義…………………………………………… 22
圖4.8 倒傳遞神經網路架構圖……………………………………… 27
圖4.9 倒傳遞神經網路架構6輸入64輸出………………………… 27
圖4.10 倒傳遞神經網路訓練流程…………………………………… 28
圖4.11 溫度收斂曲線圖……………………………………………… 30
圖4.12 訓練組平均溫度誤差比較圖………………………………… 32
圖4.13 倒傳遞神經網路測試流程…………………………………… 35
圖4.14 測試組平均溫度誤差比較圖………………………………… 39
圖4.15 組訓練組最高溫度誤差比較圖……………………………… 40
圖4.16 訓練組第1組BNN及ICEPAK溫度比較圖………………… 41
圖4.17 訓練組第81組BNN及ICEPAK溫度比較圖……………… 41
圖4.18 組訓練組最高溫度誤差比較圖……………………………… 42
圖4.19 測試組第1組BNN及ICEPAK溫度比較圖………………… 43
圖4.20 測試組第5組BNN及ICEPAK溫度比較圖………………… 44
圖4.21 測試組第13組BNN及ICEPAK溫度比較圖………………… 44
圖4.22 測試組第15組BNN及ICEPAK溫度比較圖………………… 45
圖4.23 第1組訓練組3D溫度誤差百分比分布圖…………………… 45
圖4.24 第1組訓練組的BNN及ICEPAK 3D分布圖………………… 46
圖4.25 第81組訓練組的BNN及ICEPAK 3D分布圖……………… 47
圖4.26 第1組測試組的BNN及ICEPAK 3D溫度分佈圖…………… 48
圖4.27 第5組測試組的BNN及ICEPAK 3D溫度分佈圖…………… 48
圖4.28 第13組測試組的BNN及ICEPAK 3D溫度分佈圖………… 48
圖4.29 第15組測試組的BNN及ICEPAK 3D溫度分佈圖………… 49


表目錄
表4.1 第1組熱源座標資料………………………………………… 19
表4.2 封裝晶片表面溫度分隔監控區……………………………… 23
表4.3 第1組溫度監控點座標對應………………………………… 23
表4.4 訓練組座標資料及封裝晶片表面溫度資料………………… 25
表4.5 倒傳遞神經網路預測測試組溫度資料……………………… 31
表4.6 第1組測試組座標資料……………………………………… 33
表4.7 測試組座標資料及封裝晶片表面溫度資料………………… 34
表4.8 倒傳遞神經網路預測測試組溫度資料……………………… 38
表4.9 訓練及測試數據比較表……………………………………… 50
表4.10 CFD及BNN分析速度比較…………………………………… 51
1.The International Technology Roadmap for Semiconductors, (2001).
2.“http://www.spil.com.tw/”,Siliconware Precision Industries Co, 2000-2006.
3.JOHN H. LAU, ”BALL GRID ARRY TECHNOLOGY”, 鴻海精密科技股份有限公司 1997.
4."http://www.nchc.org.tw/”, 國家高速網路與計算中心, 2005.
5.章斐章,張麗秋,黃浩倫, 類神經網路理論與實務 ,東華書局,2003.
6.蘇木春,張孝德, 機器學習類神經網路、模糊系統以及基因演算法 則 ,全華科技圖書股份有限公司, 2003.
7.Soule, C.A., “Future Trends in Heat Sink Design,” Electronics Cooling, Vol. 7, No.1, 2001, pp. 18-27.
8.Kraus, A. and Bar-Cohen, A., Thermal Analysis and Control of Electronic Equipment, Hemisphere Publishing Corp. 1983.
9.Kraus, A.D., and Bar-Cohen, A., Design and Analysis of Heat Sinks, John Wiley & Sons, Inc., 1995.
10.Wong, H. and Lee, T. Y., “Thermal Evaluation of a PowerPC 620 Multi-Processor in a Multi-Processor Computer,” IEEE Transaction on Components, Packaging, and Manufacturing Technology – Part A, Vol. 19, No.4, Dec. 1996, pp. 469-477.
11.Chang. J. Y., Yu, C. W., and Webb, R. L., “Identification of Minimum Air Flow Design for a Desktop Computer Using CFD Modeling,” Journal of Electronic Packaging, Transactions of the ASME, Vol. 123, 2001, pp. 225-231.
12.Yu, C-W. and Webb, R. L., “Thermal Design of a Desktop System Using CFD Analysis,” Seventeenth IEEE SEMI-THERM Symposium 2001, pp. 18-26.
13.Goh, T. J., Seetharamu, K. N., Quadir, G. A., Zainal, Z. A. and Jeevan, K., “Prediction of Temperature in Silicon Chip with Non-uniform power: A Largranigan Interpolation Approach,” Microelectronics International, Vol. 21 No.2, 2004, pp. 29-35.
14.Goh, T. J., Seetharamu, K. N., Quadir, G. A., Zainal, Z. A. and Ganeshamoorthy, K. J., “Thermal Investigations of Microelectronic Chip with Non-Uniform Power Distribution: Temperature Prediction and Thermal Placement Design Optimization,” Microelectronics International, Vol. 21, No.3, 2004, pp. 29-43.
15.Cheng, H.-C., Chen, W.-H., and Chung I.-C., “Integration pf Simulation and Response Surface Methods for Thermal Design of Multichip Modules,” IEEE Transactions on Components and Packaging Technologies, Vol. 27, No.2, 2004
16.Lien, H.C., and Lee, S., “A Method of Feature Selection for Textile Yarn Grading Using the Effective Distance Between Clusters,” Textile Res. J. Vol. 72, No.10, 2002, pp. 870-878.
17.Lien H.C., and Lee S., “Applying Pattern Recognition Principles to Grading Textile Yarns,” Textile Res. J. Vol. 72, No.4, 2002, pp. 320-326.
18.Lien, H.C., and Lee, S., “Applications of Neural Networks for Grading Textile Yarns,” Neural Computing and Applications, Vol. 13, 2004 pp.185-193. Lien, H.C., and Lee, S., “Applications of Neural Networks for Grading Textile Yarns,” Neural Computing and Applications, Vol. 13, 2004, pp. 185-193.
19.Ludwig L, Sapozhnikova E, Lunin V, Rosenstiel W., “Error Classification and Yield Prediction of Chips in Semiconductor Industry Applications,” Neural Computing & Applications, Vol. 9, 2000, pp. 202-210.
20.Verikas, A., Malmqvist K, Bergman L, Signahl M., “Color Classification by Neural Networks in Graphic Arts,” Neural Computing & Applications, Vol. 7, 1998, pp. 52-64.
21.Kos, A., “Approach to Thermal Placement in Power Electronics using Neural Networks,” Proceedings - IEEE International Symposium on Circuits and Systems, Vol. 4, 1993, pp. 2427-2430.
22.Y. Kuan and H. Lien, “The Integration of the Neural Network and Computational Fluid Dynamics for the Heatsink Design, Lecture Notes in Computer Science: Advances in Neural Networks,” -ISNN2005, Vol. 3498, 2005,.
23.Y. Kuan,Y.Hsueh,H Line,W.Chen, “Integrating Computational Fluid Dynamics and Neural Networks to Predict Temperature Distribution of the Semiconductor Chip with Multiple Heat Sources,” Lecture Notes in Computer Science: Advances in Neural Networks -ISNN2006, 2006, Vol.
24.Marrs, R. C., “Recent Developments in Low Cost Plastic MCM’s” Proceedings of the 2nd international Conference and Exhibition on multichip Modules, April 1993, pp. 220-229.
25.Lau, J. H., Miremadi, J.,Gleason, J., haven, R., Ottoboni, S., and S. mimura, “No Clean Mass Reflow of large Over Molder Plastic Pad Array Carriers (OMPAC),”Proceedings of the IEEE International Electronic Manufacturing Technology Symposium, October 1993, pp. 63-75.
26.Lau, J. H., K. Gratalo, T.Baker, E. Schneider, T. Marmura, “Reliability of Ball Grid Array Solder Joints Under Bending, Twisting, and Vibration Conditions,” to be published in Ciircuit World, 1995.
27.Marrs, R. C., ”Thermal Performance of the 28X28 mm QFP powerQuan®,”1994 ITAP & Flip Chip Proceedings, February 16-18, 1994, San Jose, Calif, pp. 150-164.
28.A. Malhammer. “Heat Dissipation Limits For Components Cooled By The PCB surface,” 1991 I.E.P.S Conference, San Diego, Calif, pp. 304-311.
29.V. Manno, N. Kurita, K. Azar, “Experimental Characterization of Board Conduction Effects,” 1993 Ninth IEEE Semi-therm symposium, pp. 127-135.
30.Marrs, R. C., “Recent Developments in Low Cost Plastic MCM’s” 1993 Proceedings International Conference on Mulit-Chip modules,” April 14-16, 1993, Denver, Colo, pp. 220-229.
31.Marrs, R. C., “Advanced BGA Technology for MCM Application,” Nepcon West 1994, May 1-3, 1994, Anaheim, Calif ,pp. 2006-2111.
32.Marrs, R. C., “Trends & Drivers in Technology of Low Cost MCM’s,” Workshop Speech,January 1993 Semi-ISS Conference, Monterey, Calif.
33.Marrs, R. C., “low Cost Plastic MCM Technology Trends, ”Luncheon Speech, 5th International TAB & Advanced Packaging Conference, February 3,1993, San Jose, Calif.
34.Tummula R., “Multi-chip Technologies from Personal Computers to Mainframes and Supercomputers,” February 7-11, 1993, pp. 637-643.
35.Khadpe S., “A Global View of Technology and Market Trends in TAB/Advanced Packaging,” 5th International TAB/Advanced Packaging Symposium, February 2-5 1993, pp. 1-8.
36.P. A. Sandborn, H. Hashemi, and L. Bai, “Design of MCM’s for Insertion into Standard Surface Mount Packages,” Nepcon West 1993, February 7-11, 1993, pp. 651-660
37.A. Lin, “Low Cost Multi-Chip Modules,” Nepcon West 1993 February 1993, pp. 64-649.
38.Marrs, R. C., “Technology Trends in Low Cost Flip Chip,” Luncheon Speech,First International Flip Chip & 6th International TAB & Advanced Packaging Conference, February 16, 1994, San Jose, Calif.
39.McCulloch W. S. and Pitts W. “A logical Calculus of the Ideas Immanent in Nervous Activity. Bulletin of Mathematical Biophysics,” 1943 No5: pp. 115-133
40.Hebb D. O. “The Organization of Behavior A Neuropsyshological Theory,” New York:Wiley. 1949.
41.Minsky M. L. and Papert S. A. “Perceptons. Cambridge,” MA : MIT Press. 1969.
42.Rumelhart D. E. and McClelland J. L. Parallel Distributed Peocessing: “Explorations in the Microstructure of Cognition,” MA: MIT Press Vol. 1. Cambridge. 1986.
43.P. J. Werbos, “Beyond regression : New tools of prediction and analysis in the behavioral sciences,” Doctoral dissertation, Harvard University. 1974.
44.D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning internal representations by error propagation,” in D. E. Rumelhart, J.L. McMclelland, et al., eds., Parallel distributed Processing, Vol. 1 Chap. 8, Cambridge, MA : Mit Press, 1986
45.連信仲,”特徵選擇與決策理論在纱線分級之研究,”國立中央大壆機, 械工程研究所博士論文, 2003.
46.J.P.Holman, “HEAT Transfer,” Mc Gr.w Hill, 2000.
47.Icepak 4.1 User’s Guide, Fluent Inc., 2003.
48.“http://www.ilight.com,”
49.“http://www.cfdrc.com,” CFD Research Corporation, 2006.
50.“http://www.ansys.com,” ANSYS, Inc. All Right Reserved 2006.
51.“http://www.fluent.com,” Fluent Inc., 2006
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
1. 28.戴國良(2003),無店舖販賣的主流趨勢-電視購物,突破雜誌,第211期,頁73-75。
2. 28.戴國良(2003),無店舖販賣的主流趨勢-電視購物,突破雜誌,第211期,頁73-75。
3. 27.簡竹書、錢志勇、陳慧敏(2003),這種購物頻道簡直就像一條鈔票輸送帶,新新聞,第495期。
4. 27.簡竹書、錢志勇、陳慧敏(2003),這種購物頻道簡直就像一條鈔票輸送帶,新新聞,第495期。
5. 21.郭昭琪(2001),東森得易購-結合電視購物與電子商務,商業現代化,第48期,頁18-25。
6. 21.郭昭琪(2001),東森得易購-結合電視購物與電子商務,商業現代化,第48期,頁18-25。
7. 19.陳健麟(1997),有線電視購物消費者行為之研究,南開學報,第2期,頁205-222。
8. 19.陳健麟(1997),有線電視購物消費者行為之研究,南開學報,第2期,頁205-222。
9. 12.李海容(1997),大陸電視購物的特點與前景,廣告學研究,第9期,頁87-97。
10. 12.李海容(1997),大陸電視購物的特點與前景,廣告學研究,第9期,頁87-97。
11. 11.呂雅惠(2002),銷售天后利菁90分鐘創下八千多萬業績,突破雜誌。
12. 11.呂雅惠(2002),銷售天后利菁90分鐘創下八千多萬業績,突破雜誌。
13. 5.吳正治(1996),透過電視購物的玄機,錢雜誌,第112期,頁162-164。
14. 5.吳正治(1996),透過電視購物的玄機,錢雜誌,第112期,頁162-164。
15. 簡志青(2003). 淺談微生物系生物可降解塑膠。化工資訊與商情月刊,第五期,223頁。