Abstract: The bearing is an essential part of all rotating machinery. The early detection of faults in bearing by using vibration signals saves a significant amount of financial loss. Many approaches have been used to overcome this problem in the past, but they succeeded to only some extent to identify the faults occurring on the outer race, inner race, and rollers/balls. None of them is capable of diagnosing these faults accurately. Machine Learning-based Data-driven methods have shown better results as compared to signal processing-based techniques. In this paper we are presenting a Multinomial Logistic Regression (MLR) method, which is capable of detecting faults with an accuracy of 99.80%. The results are shown in terms of Area Under Curve (AUC) of the precision-recall curve. A comparison with the Support Vector Classifier (SVC) is also shown.
Key words: Support Vector Classifier (SVC), Multinomial Logistic Regression (MLR), Area Under Curve (AUC)
[1]. Samanta, B., Al-Balushi, K.R. and Al-Araimi, S.A., Artificial Neural Networks and Support Vector Machines with Genetic Algorithm for Bearing Fault Detection, Engineering Applications of Artificial Intelligence, Vol. 16(7-8), 2003, 657-665.
[2]. Abbasion, S., Rafsanjani, A., Farshidianfar, A. and Irani, N., Rolling Element Bearings Multi-Fault Classification Based on the Wavelet Denoising and Support Vector Machine, Mechanical Systems, and Signal Processing, Vol. 21(7), 2007, 2933-2945.
[3]. Bordoloi, D.J., and Tiwari, R., Optimum Multi-Fault Classification of Gears with Integration of Evolutionary and SVM Algorithms, Mechanism and Machine Theory, Vol. 73, 2014 (a), 49-60.
[4]. Gangsar, P. and Tiwari, R., Multiclass Fault Taxonomy in Rolling Bearings at Interpolated and Extrapolated Speeds Based on Time Domain Vibration Data by SVM Algorithms, Journal of Failure Analysis and Prevention, Vol. 14(6), 2014, 826837.
[5]. Liu, Z., Cao, H., Chen, X., He, Z. and Shen, Z., Multi-Fault classification based on Wavelet SVM with PSO Algorithm to Analyze Vibration Signals from Rolling Element Bearings, Neurocomputing, Vol. 99(1), 2013, 399-410.
