Investigation of wavelets and radial basis function neural network for incipient fault diagnosis in induction motors

Abstract

One of the most important tasks in industries is considered to be the condition monitoring of induction motors. However, most of the traditional methods employed for this purpose suffer from certain limitations in accomplishing this task. A successful implementation of condition monitoring requires the development of a simple but reliable detector of various faults. This paper investigates the performance of Discrete Wavelet Transform and Radial Basis Function based Neural Network for incipient stator fault diagnosis in induction motors. Wavelet analysis helps in the extraction of important features from the faulty signal and neural network classifies the fault type depending on the nature of fault. A mean absolute percentage error of 1.4236% between the actual values and the predicted values by the neural network show the effectiveness of the proposed approach.

A novel neuro-classifier using Multiscale Permutation Entropy for motor fault diagnosis

“…(ii) Root Mean Squared Error (RMSE): Another statistical index called RMSE is mathematically computed by the program according to (9). A low value of MSE shows a good performance and vice-versa.…”

“…Similar trend is observed for RMSE which has a maximum value of only 0.1317 for the validation part. The average accuracy of prediction with the proposed methodology has been compared with a wavelet based RBF Neural network algorithm [9] and Singular Value Decomposition based Exact Radial Basis Neural Network [17] in Table III. Clearly a greater accuracy in fault prediction is achieved if MPE features are used (RMSE as less as 0.06958) as compared to wavelet based RBF-NN with RMSE 0.6124 and SVD based Exact Radial basis NN with RMSE 0.3077.…”

“…The adaptive nature of artificial intelligence techniques and their ease of implementation make them preferable over the conventional fault diagnosis approaches [8]. An efficient fault detection algorithm has been presented in [9] which integrates Discrete Wavelet Transform and Radial Basis Function neural network to make near accurate fault classification. However, time-frequency analysis methods like Wavelet Transform and Short-Time Fourier Transform are computationally time consuming.…”

Nonlinear Control of an Active Magnetic Bearing System Achieved Using a Fuzzy Control with Radial Basis Function Neural Network

“…It consists of a fuzzy controller, a reference model, an RBFNN, and an adjusting mechanism. The operating principle of this controller is based on AMB dynamics and control knowledge can be incorporated into an NFC design based on RBFNN identification [14,31,32]. A detailed description of these is shown in Figure 9, where * and are a step reference and a reference model, respectively.…”

“…This paper proposes a method for controlling the position of the rotor by using the neural fuzzy controller (NFC) approach. The method employs a fuzzy controller system with radial basis function neural network (RBFNN) rotation to identify the AMB system by Jacobian transformation [10][11][12][13][14]. The parameters of fuzzy logic controller (FLC) can be optimally tuned to solve the problem of unbalanced vibration in the AMB system by applying the gradient descent method [15] and the real time values according to the AMB system information.…”