International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 05 Issue: 06 | June-2018 www.irjet.net p-ISSN: 2395-0072 © 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 748 FAULT DETECTION AND PREDICTION OF FAILURE USING VIBRATION ANALYSIS R.Megala 1 , Dr. V.Eswaramoorthy 2 1 PG Scholar, Dept of Computer Science and Engineering, Maharaja Engineering College, Avinashi -641 654 2 Assistant Professor, Dept of Computer Science and Engineering, Maharaja Engineering College, Avinashi -641 654 ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - In Industrial applications, the uptimes of machines are enhanced through equipment monitoring. This is minimized the risks of unpredicted failures and consequent plant outages. Since, all failure modes can cause an increase in machine vibrations, monitoring this area is the predominant and most widely used method to determine equipment condition, and to predict failures. The objective of this paper is to detect faults in rotating equipment with the use of vibration analysis. A motor condition monitoring experiment is set up, and the motor’s operational speed is controlled by an AC motor drive. The vibration of the motor is measured and monitored and analyzed using spectrum analysis. The overall vibration level is monitored; the vibration severity is compared with the standard severity table and is used to determine the condition of the motor. The specific natural frequency corresponds with which kind of fault or failure mode is identified. This information is provided insight on the condition of the machine. In proposed approach, vibration signal is decomposed into several intrinsic mode functions (IMF). Subsequently, the frequency marginal of the Gabor representation is calculated to obtain the spectral content of the IMF in the frequency domain. An extended version of the STFT (Short-time Fourier transform) is the time frequency representation of Gabor, which uses a Gaussian window type and a Fourier Transform (FT) to achieve the time–frequency analysis. The extracted spectral content are fed into the classifier like support vector machine (SVM) or Random Forest classifiers(RF) to predict which type of failure occurred. Before prediction using classifier, the classifier is trained with number of sample. Three common faults in motors are analyzed in this paper: unbalance condition (UNB), bearing faults (BDF), and broken rotor bars (BRB). Key Words: IMF, Short-time Fourier transform, support vector machine, Random Forest classifiers 1. INTRODUCTION 1.1 Data Mining It is the process of extracting patterns from data. In general, it is the search for hidden patterns that may exist in large databases. Data Mining scans through a large volume of data to discover patterns and correlations between patterns. The tools that can be included are statistical models, mathematical algorithms, and machine learning methods (algorithms that improve their performance automatically through experience, such as neural networks or decision trees). Consequently, data mining consists of more than collecting and managing data, it also includes analysis and prediction. Data mining can be performed on data represented in quantitative, textual, or multimedia forms. Data mining applications can use a variety of parameters to examine the data. They include association (patterns where one event is connected to another event, such as purchasing a pen and purchasing paper), sequence or path analysis (patterns where one event leads to another event, such as the birth of a child and purchasing diapers), classification (identification of new patterns, such as coincidences between duct tape purchases and plastic sheeting purchases), clustering (finding and visually documenting groups of previously unknown facts, such as geographic location and brand preferences), and forecasting (discovering patterns from which one can make reasonable predictions regarding future activities, such as the prediction that people who join an athletic club may take exercise classes). This process can be defined by using the following six basic steps:  Defining the Problem  Preparing Data  Exploring Data  Building Models  Exploring and Validating Models  Deploying and Updating Models The following diagram describes the relationships between each step in the process Fig -1: Data Mining Process