Open Access

ARTICLE

Vibration Based Fault Diagnosis of a Hydraulic Brake System using Variational Mode Decomposition (VMD)

R. Jegadeeshwaran¹, V. Sugumaran², K.P. Soman³

1 School of Mechanical and Building Sciences, VIT University Chennai Campus, Chennai, Tamil Nadu, India – 600 127. E-mail: krjegadeeshwaran@gmail.com; jegadeeshwaran.r@vit.ac.in
2 School of Mechanical and Building Sciences, VIT University Chennai Campus, Chennai, Tamil Nadu, India – 600 127. E-mail: v_sugu@yahoo.com
3 Center of excellence in computational engineering and networking, Amrita Vishwa Vidyapeetham, Coimbatore, Tamil Nadu, India – 641 112. E-mail: kp_soman@amrita.edu.

Structural Durability & Health Monitoring 2014, 10(1), 81-97. https://doi.org/10.3970/sdhm.2014.010.081

Abstract

In automobile, brake system is an essential part responsible for control of the vehicle. Vibration signals of a rotating machine contain the dynamic information about its health condition. Many research papers have reported the suitability of vibration signals for fault diagnosis applications. Many of them are based on (Fast Fourier Transform) FFT, which have their own drawback with nonstationary signals. Hence, there is a need for development of new methodologies to infer diagnostic information from such non stationary signals. This paper uses vibration signals acquired from a hydraulic brake system under good and simulated faulty conditions for the purpose of fault diagnosis. A new approach called Variational mode decomposition (VMD) was used in this study. VMD decomposes the signal into various modes by identifying a compact frequency support around its central frequency, such that adding all the modes reconstructs the original signal. VMD finds intrinsic mode functions on central frequencies using alternating direction multiplier method (ADMM). Descriptive statistical features were extracted from VMD processed signals and classified using a machine learning algorithm. For classification J48 decision tree algorithm was used. The results were compared with the statistical features extracted from raw signal using decision tree classifier.

---

Keywords

---