臺灣博碩士論文加值系統

類神經網路已被廣泛的運用於各種領域，如工程系統診斷與監控、數位影像處理分析、語音辨識、聲納與地震訊號分類以及商業金融管理等。這是因其能處理大量複雜的資訊，建立輸入與輸出間的對應關係，並進行學習與修正，從而得到一辨識系統。
本研究藉由離散小波轉換與類神經網路理論，來找出帶有損傷之鋁合金懸臂樑與平板的破損位置與破損程度。在數值模擬上，本研究首先以有限元素模擬含有缺陷之懸臂樑與平板結構，藉由模態分析獲得結構物之模態振型，再以離散小波分解模態振型後，重構各層之高頻部份並取均方根值(RMS)計算獲得各層破損指標，以此作為類神經網路所需之訓練樣本，建立一套損傷識別系統。在實驗分析上，先以衝擊鎚給於懸臂樑結構一個外力衝擊，並以加速規量取結構物之振動訊號，隨後利用小波包轉換振動訊號得到實驗結構物的模態振型，最後將實驗所得之模態振型，經由離散小波分解與重構，並同樣地取RMS計算來求得實驗之破損指標。最後匯入已測試好的模擬類神經網路中，以期所建立的類神經網路的損傷識別系統，能找出懸臂樑之破損深度。

Neural networks have been widely used in various fields, such as diagnosis and monitoring of industrial engineering system, digital image processing and analysis, speech recognition, sonar and seismic signal classification, and financial management. The method can process complex information and establish the mapping relationship between input and output by learning and revision. Finally, an identification system is acquired.
The aim of this study is to find the depth of the damage of an aluminum bar and to detect the approximate length of the crack of a damaged plate by the discrete wavelet transform and the neural network method. In numerical analysis, the first mode shape of a damaged cantilever beam and plates are simulated by the finite element method. The obtained mode shapes of the damaged cantilever beam and plates are then transformed by the discrete wavelet transform. The damage index of the beam or plates can be obtained by reconstructed the high frequencies detail signals of each layer and used to be the neural networks training samples, and then the damage identification system for beams and plates can finally be established. In experiment analysis, several accelerometers are used to measure the vibration responses of the beam, and the first mode shapes of the beam is attained by using the wavelet packet node norm. Finally, the experimental damage index are obtained and used as inputs of the identification system to find the approximate depth of the damage of a cantilever beam. It is believed that the proposed crack identification system is feasible to identify the damage depths of cantilever beams.

目錄
摘要.......................................................I
Abstract ................................................II
誌謝.....................................................III
目錄.....................................................IV
表目錄..................................................VIII
圖目錄....................................................X
符號說明.................................................XIV
第一章 緒論................................................1
1.1 研究動機............................................1
1.2 文獻回顧............................................2
1.3 本文架構............................................6
第二章 振動與模態分析理論....................................7
2.1 前言................................................7
2.2 離散系統的自由振動....................................7
2.3 懸臂樑之振動分析....................................10
2.4 含損傷板之振動模態推導...............................13
2.5 實驗模態分析........................................17
2.5.1 頻率響應函數...........................18
2.5.2 振型分析...............................19
2.5.3 自然頻率分析...........................19
2.6 實驗模態分析流程..............................20
第三章 訊號分析及小波簡介.............................22
3.1 前言.........................................22
3.2 傅立葉轉換....................................22
3.3 短時傅立葉轉換................................25
3.4 小波轉換.....................................26
3.4.1 小波函數...............................27
3.4.2 Symlet（SymN）小波系...................30
3.4.3 連續小波轉換............................32
3.5 離散小波轉換..................................35
3.5.1 近似空間與細節空間.......................36
3.5.2 Mallat運算法...........................38
3.5.3 正交基底...............................40
3.6 小波包分析....................................41
3.6.1 小波包的構造............................42
3.6.2 小波包的元素............................43
3.6.3 小波包節點範數..........................44
3.7 二維離散小波轉換..............................45
3.8 傅立葉轉換與小波轉換之比較.................. ...47
第四章 類神經網路....................................49
4.1 類神經網簡介................................49
4.1.1 生物神經元模型.........................50
4.1.2 類神經元模型...........................52
4.2 類神經網路的架構..............................55
4.3 倒傳遞類神經網路..........................58
4.3.1 前饋傳遞過程...........................59
4.3.2 倒傳遞過程.............................61
4.3.3 倒傳遞網路演算法........................65
4.4 倒傳遞網路的參數..............................66
4.5 類神經網路的運行方式...........................68
第五章 數值模擬分析..................................70
5.1 前言........................................70
5.2 有限元素的劃分................................70
5.3 懸臂樑.......................................75
5.3.1 裂縫寬度10mm之懸臂樑....................75
5.3.2 裂縫寬度3mm之懸臂樑......................83
5.3.3 不同破損位置之懸臂樑.....................87
5.4 平板.........................................95
5.4.1 單破損深度之平板.........................95
5.4.2 單破損長度之平板.........................101
5.4.3 不同破損位置之平板.......................104
5.5 實驗誤差分析..................................112
第六章 實驗量測與結果.................................114
6.1 實驗流程.....................................114
6.2 訊號之取樣頻率................................115
6.3 實驗材料尺寸及振動訊號之量測....................117
6.4 實驗分析......................................120
6.4.1 離散小波包轉換...........................120
6.4.2 實驗模態振型分析.........................122
第七章 結論...........................................131
7.1 結論..........................................131
7.2 未來研究方向與建議..............................133
參考文獻..............................................134

參考文獻
[1]Cawley, P., and Adams, R.D., The Location of Defects in Structures from Measurement of Natural Frequencies., Journal of Strain Analysis, 14, 1979.
[2]Choubey, A. , D.K. Sehgal and N. Tandon ,“Finite element analysis of vessels to study changes in natural frequencies due to cracks, International Journal of Pressure Vessels and Piping, Vol. 83, Issue 3, pp.181-187 , 2006.
[3]Chen, Wenyuan , Lei Zhao and Guotang Bi,“ Application of neural network and wavelet analysis in monitoring multiple structural damage, International Conference on Electronic Measurement and Instruments, pp. 4463-4467, 2007.
[4]Daubechies, I., “Ten Lectures on Wavelets, SIAM, Philadelphia, 1992.
[5]Deng, X., Wang, Q., “Crack Detection Using Spatial Measurements and Wavelet, International Journal of Fracture, Vol. 91, 1998.
[6]Douka, E., Loutridis, S., and Trochidis, A., “Crack Identification in Doube-cracked Beams Using Wavelet Analysis, Journal of Sound and Vibration, 277, pp.1025-1039, 2004.
[7]Diao, Yan-Song, Hua-Jun Li and Yan Wang, “A two-step structural damage detection approach based on wavelet packet analysis and neural network, Proceedings of the 2006 International Conference on Machine Learning and Cybernetics, pp.3128-3133, 2006.
[8]Fuchs, H.O., Metal Fatigue in Engineering, John-Wiley ＆ Sons Inc, 1980.
[9]Kudva, J.N. , N. Munir and P.W. Tan, “Damage detection in smart structures using neural networks and finite-element analyses, Smart Materials and Structures, Vol.1, NO.2, pp. 108-112, 1992.
[10]Liew, K. M., Wang, Q., “Application of Wavelet Theory for Crack Identication in Structures, Journal of Engineering Mechanics, Vol.124(2), pp.152-157, 1998.
[11]Loutridis, S., Douka, E., and Trochidis, A., “Crack Identification in Beams Using Wavelet Analysis, International Journal of Solids and Structures , Vol.40, pp.3557-3569, 2003.
[12]Mallat, S.G., “A Theory for Multiresolution Signal Decomposition, The Wave Representation, Comm. Pure Appl. Math, 41, pp.674-693, 1988.
[13]Mallat, S., “A Wavelet Tour of Signal Processing, Academic Press, USA, 1998.
[14]Messina, A., Jones, I.A., “Damage Detection and Localization Using Natural Frequency Change, 14th International Modal Analysis Conference, Orlando, pp.67-76, 1996.
[15]Messina, A., Williams, E.J., Contursi, T., “Structural Damage Detection by Sensitivity and Statistical-Based Method, Journal of Sound and Vibration, 216, pp.791-808, 1998.
[16]Meyer, Y., Ondettes et Functions Splines, Lectures given at the University of Torino, Italy, 1986.
[17]Meyer, Y., Ondettes, Functions Splines et Analyses Graduees, Seminaire EDP, Ecole Polytechnique, Paris, France, 1986.
[18]Morlet, J., Arens, G., Fourgeau, I., and Giard, D., “Wave Propagation and Sampling Theory, Geophysics, 47, pp.203-236, 1982.
[19]Morlet, J., “Sampling Theory and Wave Propagation, NATO ASI Series, Issues in Acoustic Signal/Image Processing and Recognition , Vol. I, pp.233-261, 1983.
[20]Narkis, Y., Identification of Crack Location in Vibrating Simply Supported Beams, Journal of Sound and Vibration, 172, pp.549-558, 1994.
[21]Pandey, A.K., Biswas, M., Samman, M.M., “Damage Detection from Changes in Curvature Mode Shapes, Journal of Sound and Vibration, 145, pp. 321-332, 1991.
[22]Peng, Z.K., Chu, F.L., “Application of the Wavelet Transform in Machine Condition monitoring and Fault Diagnostics: a Review with Bibliography, Mechanical Systems and Signal Processing, 18, pp.199–221, 2004.
[23]Ratcliffe, C.P., “Damage Detection Using a Modified Laplacian Operator on Mode Shape Data, Journal of Sound and Vibration, 204, pp.505-517, 1997.
[24]Sampaio, R.P.C., Maia, N.M.M., Silva, J.M.M., “Damage Detection Using the Frequency-Response-Function Curvature Method, Journal of Sound and Vibration, 226, pp. 1029-1042, 1999.
[25]Stubbs, N., Kim, J.T., “Damage Localization in Structures Without Baseline Modal Parameters, American Institute of Aeronautics and Astronautics Journal, 34, pp. 1644-1649, 1996.
[26]Surace, C., and Ruotolo, R., “Crack Detection of a Beam Using the Wavelet Transform, Proceedings of the 12th International Modal Analysis Conference, Honolulu, pp.1141-1147, 1994.
[27]Wang, Q., Deng, X., “Damage Detection with Spatial Wavelets, International Journal of Solids and Structures, Vol.36, pp.3443-3468, 1999.
[28]WERNER SOEDEL, “VIBRATIONS OF SHELLS AND PLATES, Marcel Dekker,America,1993.
[29]Yam, L.H. , Y.J. Yan and J.S. Jiang, “Vibration-based damage detection for composite structures using wavelet transform and neural network identification, Composite Structures, Vol. 60, Issue 4, pp. 403-412, 2003.
[30]Yam, L.H. , Y.J. Yan , L. Cheng and J.S. Jiang, “Identification of complex crack damage for honeycomb sandwich plate using wavelet analysis and neural networks, Smart Materials and Structures, Vol.12, NO.5, pp. 661-671, 2003.
[31]Yang, J.M., C.N. Hwang and B. L. Yang, Crack Identification in Beams and Plates by Discrete Wavelet Transform Method, Journal of Ship Mechanics,Vol.3,NO. 3,pp.464-472,2008.
[32]Zhong, Shuncong, Olutunde Oyadiji, S., “Crack Detection in Simply Supported Beams Without Baseline Modal Parameters by Stationary Wavelet Transform, Mechanical Systems and Signal Processing, 21, pp.1853-1884, 2007.
[33]Zang, C. and M. Imregun, “Structural damage detection using artificial neural networks and measured FRF data reduced via principal component projection, Journal of Sound and Vibration, Vol. 242, NO. 5, pp. 813-827, 2001.
[34]王?村, “實驗模態分析, 課程講義, 台灣, 1996.
[35]王?村, “振動學, 全華科技, 台灣, 1996.
[36]長思科技產品研發中心,“MATLAB6.5輔助小波分析與應 用,電子工業出版社,北京,2003.
[37]周鵬程,“類神經網路入門-活用Matlab,全華科技圖書股份有限公司,2002.
[38]周志豪,“應用類神經網路於主動振動控制之研究,南台科技大學機械工程系研究所,碩士論文,中華民國九十五年六月.
[39]林子超,“類神經網路在結構損傷偵測分析的應用,國立成功大學航空太空工程研究所,碩士論文,中華民國八十六年六月.
[40]胡昌華等,“基於MATLAB的系統分析與設計：小波分析, 西安電子科技大學出版社,中國大陸,1999.
[41]洪維恩,“ Matlab 7 程式設計, 旗標出版社,台灣,2006.
[42]許有村, “小波理論在機械結構破壞點偵測上的應用, 國立台灣大學機械工程研究所, 碩士論文, 中華民國八十七年六月.
[43]陳精一,蔡國忠, “電腦輔助工程分析ANSYS使用指南, 全華科技股份有限公司, 台灣, 2000.
[44]惠汝生, “自動量測系統-LabVIEW, 全華出版社, 台灣, 2002
[45]張智星, “MATLAB程式設計與應用, 清蔚科技, 台灣, 2000.
[46]張智傑,“小波訊號處理於結構破損偵測之應用,國立成功大學機械工程學系,博士論文,中華民國九十三年六月.
[47]單維彰, “凌波初步, 全華科技, 台灣, 2000.
[48]楊澤民,廖翊宏, 嵇泓鈞,“小波理論應用於樑的破損偵測之究,中國造船暨輪機工程學刊,Vol.23,No.4,pp.189-197,2004.
[49]楊澤民,黃正能,柯志宏,曾韋鳴, “二維離散小波轉換應用於帶有加強材之平板破損偵測, 中國造船暨輪機工程學刊,Vol.28,No.1,pp.29-40,2009.
[50]楊澤民,楊正瑋,“小波轉換應用於板架結構的損傷偵測, 中國造船暨輪機工程學刊,Vol.28,No.4,pp.221-235,2009.
[51]潘威成,“以模態實驗方法研究懸臂樑的振動行為 , 國立成功大學土木工程研究所,中華民國八十九年七月.
[52]蔡展榮,“小波基礎理論與應用課程講義, 國立成功大學空間及測量工程學系研究所, 2002.
[53]蔡哲緯,“應用類神經網路於結構損傷定位之研究,國立成功大學航空太空工程研究所,碩士論文,中華民國九十七年六月.
[54]劉晉奇,褚晴暉,“有限元素分析與ANSYS的工程應用, 滄海書局, 台灣, 2006.
[55]蕭子健,“LabVIEW分析篇, 高立出版社, 台灣, 2000.
[56]蕭子健,“LabVIEW進階篇, 高立出版社, 台灣, 2000.
[57]鐘啟聞,“應用類神經網路與自然頻率於結構物之損壞預測,國立成功大學造船暨船舶機械工程研究所,碩士論文,中華民國九十一年六月.
[58]羅華強,“類神經網路-Matlab的應用,高立圖書有限公司, 台灣,2008.