本文探討利用類神經網路對於輸入模態訊息作形式辨別的最佳化過程，進行結構損傷定位。藉由模態曲率對於振型變化十分靈敏之特性，針對含雜訊之模態資料情況下進行結構損傷定位分析。因模態資料中含些許雜訊時，使得模態曲率產生極大的變化，造成識別不良的情形。為此，吾人提出平均模態曲率變化之方法，降低雜訊對損傷定位的干擾，突顯結構損傷位置；再利用類神經網路有效的標定損傷位置。最後經由數值模擬，驗證本文所探討的方法之可行性。

In this thesis, structural damage detection analysis is studied with consideration of the modal data under the noisy conditions using the artificial neural networks. In the process of damage detection, a method in combination with system identification technique is employed. We proposed a modification to the conventional method of modal curvature, so that possibly damaged elements in a structure can be located more reliably. Numerical simulation results show that the proposed method is accurate and robust using the modal data under noisy conditions.

[1]Young, J. K. and On, F. J., “Mathematical Modeling via Direct Use of Vibration Data”, National Aeronautic and Space Engineering and Automotive Engineers, Los Angeles, California, Reprint No 690615, October 1969.

[2]Berman, A. and Flannely, W. G., “Theory of Incomplete Models of Dynamic Structures”, AIAA Journal, Vol.9 (8), August 1971.

[3]Berman, A., “System Identification of Structural Dynamic Models - Theoretical and Practical Bounds”, AIAA paper 84-0929, 1984.

[4]Adams , R. D., Walton, D., Flitcroft, J. E., and Short, D., ”Vibration Testing as a Non-destructive Test Tool for Composite Materials”, Composite Reliability, ASTM STP 580, pp. 159-175, 1975.

[5]Adams ,R.D., Cawley, P., Pye, C. J., and Stone, B. J., ”Vibration Techniques For Non-destructively Assessing the Integrity of Structures”, Journal of Mechanical Engineering Science, Vol. 20, no. 2, pp. 93-100, 1978.

[6]Kudva, J. N., Munir, N., and Tan, P. W., “Damage Detection in Smart Structure Using Neural Network and Finite Element Analysis”, Smart Materials and Structures (UK)., Vol. 1, no. 2, pp. 108-112, June. 1992.

[7]Wu, X., Ghaboussi, J., and Garrett, J. H., “Use of Neural Network In Detection of Structural Damage”, Computer & Structures, Vol. 42, No. 4, pp. 649-659, 1992.

[8]葉怡誠, “機器學習在土木工程專家系統應用之研究”,博士論文, 國立成功大學土木工程學研究所, 1992.

[9]Pandey, P. C. and Barai, S.V., “Multilayer Perceptron In Damage Detection Of Bridge Structural”, Computer & Structures, Vol. 54, No. 4, pp. 597-608, 1995.

[10]林子超, “類神經網路在結構損傷偵測分析的應用“,碩士論文, 國立成功大學航空太空工程學研究所, 1997.

[11]Aygen, Z. E., Seker, S., Bagnyanik, M., Bagnyanik, F. G., and Ayaz, E., “Fault Section Estimation in Electrical Power Systems Using Artificial Neural Network Approach”, Transmission and Distribution Conference of IEEE, Vol. 2, pp. 466 - 469, 1999.

[12]Fanf, X., Luo, H., and Tang, J. “Structural Damage Detection Using Neural Network with Learning Rate Improvement”, Computer & Structures, Vol. 83, pp. 2150-2161, 2005.

[13]Pandey, A. K., Biswas, M., and Samman, M. M., ”Damage Detection From Changes in Curvature Mode Shapes” , Journal of Sound and Vibration, Vol. 2, pp.321-332, 1990.

[14]Rosenblatt, F.,”A Probabilistic Modal for Information Storage and Organization in the Brain”, Psychology Review, Vol. 65, pp. 386-408, 1958.

[15]Webos, P. J., ”New Tools for Prediction and Analysis in the Behavioral Science”, Beyond Regression: Doctoral Dissertation, Appl. Math., Harvard University, Mass.

[16]Rumelhart, D. E., Hinton, G. E., and Williams, R. J., ”Learning Internal Representation by Error Propagation” ,in parallel Distributed Processing, Vol. 1, pp.318-362, 1986.

[17]Abrahart, R., See, L., and Kneale, P. E., “Using Network Pruning and Model Breeding Algorithms to Discover Optimum Inputs and Architectures”, In Proceeding of the 3 International Conference on Geocomputation, University of Bristol, 1998.

[18]Dawson, C. W. and Wilby, R. L., “Hydrological Modeling Using Artificial Neural Networks”, Progress in Physical Geography, Vol. 25, No. 1, pp. 80-108, 1995.

[19]Liqun, Ren. and Zhiye, Zhao.,“ An Optimal Neural Networks and Concrete Strength Modeling”, Advances in Engineering Software, Vol. 33, pp. 117-130, 2002.

[20]葉怡誠, “類神經網路-模式應用與實作”, 儒林圖書有限公司, 1995.