AUTOMATIC CLASSIFYING METHOD FOR WEDGE TIGHTNESS BY SUPPORT VECTOR MACHINE AND ARTIFICIAL NEURAL NETWORK
Keywords:
Wedge tightness signal, support vector machine, artificial neural network, classificationAbstract
This study proposed an automatic classifying system for the tightness of a generator’s wedge. It consists of 4 processes called data collection, preprocessing, feature extraction, and classification. The aim of this study is to classify and verify the tightness of a generator’s wedge for the Electricity Generating Authority of Thailand by using 2 machine learning algorithms called support vector machine (SVM) and artificial neural network (ANN). The linear function and radial basis function (RBF) are selected for the SVM classifier. The evaluation of the SVM classifier is completed by using a 10-fold cross validation technique to give high accuracy and a low number of false negatives (FN). From the simulation results, the efficiencies are above 90% in the time domain and the frequency domain, which are satisfactory for classification. By comparison, the signals extracted in the frequency domain have fewer FN than the time domain. The ANN gives the best performance among the classifiers in the time domain (93.75% for the ANN, 92.34% for the SVM with the linear function, and 91.38% for the SVM with the RBF) and the frequency domain (100% for the ANN, 100% for the SVM with the linear function, and 99.47% for the SVM with the RBF).
References
Ahmad, A.R., Khalid, M., Gaudin, C.V., and Poisson, E. (2004). Comparison of support vector machine and neural network in character level discriminant training for online word recognition. Unpublished data from the UNITEN Students’ Conference on Research and Development; December 9-10, 2004; Kuala Lumpur, Malaysia.
Avci, E. (2009). A new intelligent diagnosis system for the heart valve disease by using genetic-SVM classifier. Expert Syst. Appl., 36(7):10618-10626.
Chatunapalak, I., Phatiwuttipat, P., Kongprawechnon, W., and Tungpimolrut, K. (2012). Childhood musical murmur classification with support vector machine technique. Proceedings of the 9th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON 2012); May 16-18, 2012; Hua Hin, Thailand, p. 1-4.
Delen, D., Walker, G., and Kadam, A. (2005). Predicting breast cancer survivability: a comparison of three data mining methods. Artif. Intell. Med., 34(2): 113-127.
George, F.D., Mark, W.F., and Harry, L.S., inventors; Westinghouse Electric Corporation, assignee. February 27, 1996. System and Method for Impact Testing Wedge Tightness in an Electrical Generator. U.S. Patent No. 5493894.
Kim, J.S., Kim, I.B., and Yoo, K.S. (2012). An experimental comparison of ultrasonic pulse velocity and frequency domain signal of concrete. Unpublished data from the 5th International Conference of the Asian Concrete Federation; October 24-26, 2012; Pattaya, Thailand.
Klempner, G. and Kerszenbaum, I., (2004). Operation and Maintenance of Large Turbo-generators. John Wiley & Sons, Inc, Hoboken, New Jersey, p. 485-487.
Phatiwuttipat, P., Kongprawechnon, W., and Tungpimolrut, K. (2011). Cardiac auscultation analysis system with Neural Network and SVM technique. Unpublished data from the Proceedings of the 8th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology; May 17-19, 2011; Khon Kaen, Thailand, p. 1027-1030.
Saechai, S., Kongprawechnon, W., and Sahamitmongkol, R. (2012). Test system for defect detection in construction material with ultrasonic waves by support vector machine and neural network. Proceedings of the Joint 6th International Conference on Soft Computing and Intelligent Systems (SCIS) and the 13th International Symposium on Advanced Intelligent Systems (ISIS); November 20-24, 2012; Kobe, Japan, p. 1034-1039.
Tohka, J. (2013). SGN-2,506: Introduction to Pattern Recognition. Department of Signal Processing, Tampere University of Technology, Tampere, Finland, 80p.
Vejjanugraha, P., Kongprawechnon, W., Kondo, T., Tungpimolrut, K., Kaneko, H., and Sintuwong, S. (2013). An automatic screening method for detecting glaucoma using multi-class support vector machine. Proceedings of the International Conference on Information and Communication Technology for Embedded Systems (ICICTES2014), [CD-Rom]; January 23-25, 2014; Ayutthaya, Thailand.
