Fault classification in cylinders using multi-layer perceptrons, support vector machines and Gaussian mixture models

• Autorzy: Marwala T. , Mahola U. , Chakraverty S.
• Konferencja: Neural Networks and Soft Computing/International Symposium (30.06-02.07.2005 ; Cracow, Poland)
• Języki publikacji: EN

Abstrakty

EN

Gaussian mixture models (GMM) and support vector machines (SVM) are introduced to classify faults in a population of cylindrical shells. The proposed procedures are tested on a population of 20 cylindrical shells and their performance is compared to the procedure, which uses multi-layer perceptrons (MLP). The modal properties extracted from vibration data are reduced into low dimension using the principal component analysis and are then used to train the GMM, SVM and MLP. It is observed that the GMM gives 98% classification accuracy, SVM gives 94% classification accuracy while the MLP gives 88% classification accuracy. Furthermore, GMM is found to be more computationally efficient than MLP which is in turn more computationally efficient than SVM.

Słowa kluczowe

EN

Gaussian mixture models; support vector machine; cylinder; neural network

PL

model Gaussa; wektor; walec; sieci neuronowe

• Wydawca: Instytut Podstawowych Problemów Techniki PAN
• Czasopismo: Computer Assisted Mechanics and Engineering Sciences
• Rocznik: 2007
• Tom: Vol. 14, No. 2
• Strony: 307--316
• Opis fizyczny: Bibliogr. 27 poz., rys., tab.

Twórcy

autor

Marwala T.

autor

Mahola U.

autor

Chakraverty S.

• School of Electrical and Information Engineering, University of the Witwatersrand Private Bag X 3, Wits 2050, South Africa

Bibliografia

• [1] R. Bellman. Adaptive Control Processes: A Guided Tour. Princeton University Press, New Jersey, USA, 1961.
• [2] CM. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, Oxford, UK, 1995.
• [3] C.B. Bishop. Pattern Recognition and Machine Intelligence. Springer, Berlin, 2006.
• [4] CA. Burges. A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 2: 121-167, 1998.
• [5] A. Dempster, N. Laird, D. Rubin. Maximum likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society, 39: 1-38, 1977.
• [6] S.W. Doebling, CR. Farrar, M.B. Prime, D.W. Shevitz. Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: a literature review. Los Alamos Technical Report LA-13070-MS, Los Alamos National Laboratory, New Mexico, USA, 1996.
• [7] D.J. Ewins. Modal Testing: Theory and Practice, Research Studies Press, Letchworth, UK, 1995.
• [8] S. Geman, E. Bienenstock, R.R. Doursat. Neural networks and the bias/variance dilemma. Neural Computation, 4: 1-58, 1992.
• [9] E. Habtemariam, T. Marwala, M. Lagazio. Artificial intelligence for conflict management. Proceedings of the IEEE International Joint Conference on Neural Networks, Montreal, Canada, pp. 2583-2588, 2005.
• [10] S. Haykin. Neural networks. Prentice-Hall, New York, USA, 1995.
• [11] G.E. Hinton, Learning translation invariant recognition in massively parallel networks. In: J.W. de Bakker, A.J. Nijman, P.C. Treleaven, eds., Proceedings PA RLE Conference on Parallel Architectures and Languages Europe, pp. 1-13. Springer-Verlag, Berlin, 1987.
• [12] A. Iwasaki, A Todoroki, Y. Shimamura, et al.. An unsupervised statistical damage detection method for structural health monitoring (applied to detection of delamination of a composite beam). Smart Materials and Structures, 13: 80-85, 2004.
• [13] J. Joachims. Making large-scale SVM learning practical. In: B. Schölkopf, C.J.C. Burges, A.J. Smola, eds.,Advances in Kernel Methods - Support Vector Learning, pp. 169-184. MIT Press, Cambridge, MA, 1999.
• [14] I.T. Jollife. Principal Component Analysis. Springer-Verlag , New York, USA, 1986.
• [15] N.M.M. Maia, J.M.M. Silva. Theoretical and Experimental Modal Analysis. Research Studies Press, Letchworth, U.K, 1997.
• [16] T. Marwala. Fault identification using pseudomodal energies and modal properties. AIAA Journal, 39: 1608-1618, 2001.
• [17] T. Marwala. Probabilistic fault identification using a committee of neural networks and vibration data. AIAA, Journal of Aircraft, 38: 138-146, 2001.
• [18] T. Marwala. Fault Identification Using Neural Networks and Vibration Data. University of Cambridge Ph.D.Thesis, Cambridge, UK, 2001.
• [19] L.M. Meng, G. Prei, I. Osorio, G. Strang, T.Q. Nguyen. Gaussian mixture models of ECoG signal features for improved detection of epileptic seizures, Medical Engineering and Physics, 26: 379-393, 2004.
• [20] M. M0ller. A scaled conjugate gradient algorithm for fast-supervised learning. Neural Networks, 6: 525-533, 1993.
• [21] K.R. Muller, S. Mika, G. Ratsch, K. Tsuda, B. Schölkopf. An introduction to kernel-based learning algorithms.IEEE Transactions on Neural Networks, 12: 181-201, 2001.
• [22] F.V. Nelwamondo, T. Marwala, U. Mahola. Early classifications of bearing faults using hidden Markov models,Gaussian mixture models, Mel-frequency cepstral coefficients and fractals. International Journal of Innovative Computing, Information and Control, 2: 1281-1299, 2006.
• [23] M.M. Pires, T. Marwala. American option pricing using multi-layer perception and support vector machines. Proceedings of the IEEE Conference in Systems, Man and Cybernetics, The Hague, pp. 1279-1285, 2004.
• [24] D.A. Reynolds, T.F. Quatieri, R.B. Dunn. Speaker verification using adapted Gaussian mixture models. Digital Signal Processing, 10: 19-41, 2000.
• [25] B. Schölkopf, A.J. Smola. A short introduction to learning with kernels. Proceedings of the Machine Learning Summer School, pp. 41-64. Springer, Berlin, 2003.
• [26] V. Vapnik. The Nature of Statistical Learning Theory. Springer Verlag, New York, Berlin, 1995.
• [27] K. Worden, A.J. Lane. Damage identification using support vector machines. Smart Materials and Structures, 10: 540-547, 2001