The use of features selection and nearest neighbors rule for faults diagnostic in induction motors
Engineering Applications of Artificial Intelligence
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
Hi-index | 0.00 |
This paper introduces a technique for diagnosing mechanical faults of induction motors by using support vector machine (SVM) and genetic algorithm (GA). Features are extracted from the vibration time signals and selected by using GA with a distance evaluation fitness function. All SVM parameters are also obtained simultaneously by the same GA. The SVM is studied with two types of kernel functions, the radial basis function and the polynomial function. Four motor conditions are investigated with the chosen SVM classifiers. The classification results have high accuracy for the chosen feature set and SVM parameters.