Review on engine vibration fault analysis based on data mining

Through equipment monitoring, the uptimes of machines are enhanced in industrial applications. The unpredicted failures risks minimized by proper monitoring. machine vibrations increased caused failure modes. vibration data requires effective analysis accurate assessment equipment. For fault feature selection and detection faults rotating equipment, empirical knowledge is required. Low efficiency methods motor speed control main drawbacks existing techniques. So basic aim this paper utilizing analysis. analyzed monitored using spectrum spectral content extracted fed into classifier like k-Nearest neighbors (KNN), back-propagation neural network BPNN, Sparse Representation Classifier (SRC), Support vector (SVM) Random Forest (RF) for type prediction analyze unbalance condition (UNB), bearing (BDF), broken rotor bars (BRB) faults. RF better as compared to other classifiers terms accuracy, precision recalls values approximately 10.92 %, 11.03 % 20.13 respectively.