Arezki Menacer

Fast Fourier and discrete wavelet transforms applied to sensorless vector control induction motor for rotor bar faults diagnosis

This paper presents new techniques to evaluate faults in case of broken rotor bars of induction motors. Procedures are applied with closed-loop control. Electrical and mechanical variables are treated using fast Fourier transform (FFT), and discrete wavelet transform (DWT) at start-up and steady state. The wavelet transform has proven to be an excellent mathematical tool for the detection of the faults particularly broken rotor bars type. As a performance, DWT can provide a local representation of the non-stationary current signals for the healthy machine and with fault. For sensorless control, a Luenberger observer is applied; the estimation rotor speed is analyzed; the effect of the faults in the speed pulsation is compensated; a quadratic current appears and used for fault detection.

Detection of broken rotor bar faults in induction motor at low load using neural network.

The knowledge of the broken rotor bars characteristic frequencies and amplitudes has a great importance for all related diagnostic methods. The monitoring of motor faults requires a high resolution spectrum to separate different frequency components. The Discrete Fourier Transform (DFT) has been widely used to achieve these requirements. However, at low slip this technique cannot give good results. As a solution for these problems, this paper proposes an efficient technique based on a neural network approach and Hilbert transform (HT) for broken rotor bar diagnosis in induction machines at low load. The Hilbert transform is used to extract the stator current envelope (SCE). Two features are selected from the (SCE) spectrum (the amplitude and frequency of the harmonic). These features will be used as input for neural network. The results obtained are astonishing and it is capable to detect the correct number of broken rotor bars under different load conditions.

Approach Signal for Rotor Fault Detection in Induction Motors

In this paper, two approach signals are used for broken rotor bar fault diagnosis. One is based on the spectrum analysis, such as the fast Fourier transform, which utilizes the steady-state spectral components of the stator quantities. The accuracy of this technique depends on the loading conditions and constant speed of the machine. The second approach is based on the discrete wavelet transform which is considered an ideal tool for this purpose due to its suitability for the analysis of signals, the frequency spectrum of which is variable in time. These two approaches are tested in simulation and validated experimentally.

Application of the fourier and the wavelet transform for the fault detection in induction motors at the startup electromagnetic torque

In this paper, a method for the diagnosis of rotor bar failures for the induction machines has been presented. It based on the analysis of the stator current, using the Fourier transform and discrete wavelet transform (DWT) at the start-up electromagnetic torque. Using the simplified dynamic model of the squirrel cage induction motor taking account the broken rotor bars and the discrete wavelet transform (DWT) in order to extract the different harmonic components of the stator currents. The performance presented by using of the DWT: its ability to provide a local representation of the non stationary current signals for the healthy machine and with fault (two adjacent broken bars). The results are compared with those obtained using the Fourier transform.

Mixed eccentricity fault diagnosis in the sensorless field oriented control induction motor using DWT technique

This paper deals with the mixed eccentricity fault detection in the closed loop of the induction motor driven by Field-Oriented Control. In this paper, in order to analysis, the unbalanced signals caused by the fault, two techniques are used for stator and quadratic currents component, such as FFT and wavelet transform. The result obtained by wavelet is more suitable for emergency signal analysis. This method is effective for stationary signal processing as well as non-stationary signal processing. Wavelet analysis was introduced to overcome the shortcomings of Fourier analysis.

A novel method for induction motors stator inter-turn short circuit fault diagnosis based on wavelet energy and neural network

This paper presents a wavelet neural network technique for the inter turn short-circuit fault detection and location of the induction machine at non stationary state. This technique is used in order to remedy the problem from using the classical signal-processing (FFT). This method is based from using the discrete wavelet energy (DWE) as the input for the neural network (NN). The fault detection and location are achieved by a feed-forward multilayer-perceptron neural network trained by back propagation. Simulation results are presented to illustrate the effectiveness of the proposed method.

Stator inter turns fault detection using discrete wavelet transform

The main objective of this paper presents a discrete wavelet transform (DWT) to diagnoses the stator inter-turns faults in three induction motor. This technique is based on the analysis of stator current under healthy and faulty states. Through the energy stored in each decomposition obtained by wavelet analysis, it was possible to remedy the problem of confusion between the types of defects above the short-circuit defect.

DWT wavelet transform for the rotor bars faults detection in induction motor

The detection of broken rotor bars faults based on the analysis techniques, such as the fast Fourier transform (FFT), the accuracy of these technique depend on the loading conditions of the machine, and the ability to maintain a constant speed. To over come this problem, the analysis of the envelope of the transient starting-current waveform using the wavelet transform has been investigated. In this paper, a method for the diagnosis of rotor bar failures for the induction machine has been presented. It is based on the analysis of the stator current, using the discrete wavelet transform (DWT) at the start-up electromagnetic torque. Using the simplified dynamic model taking account the faults of the squirrel cage induction motor and the discrete wavelet transform (DWT), in order to extract the different harmonics components of the stator currents. The performance presented by using of the DWT: its ability to provide a local representation of the non stationary current signals for the healthy machine and with fault (two adjacent broken rotor bars).

Discrete wavelet transform for stator fault detection in induction motors

This paper present the stator inter-turn short circuit fault detection in induction machine from stator current analysis using two techniques, such as the fast Fourier transform (FFT), which utilize the steady-state spectral components of the stator quantities. The accuracy of these techniques depends on the load and speed of the machine and the discrete wavelet transform (DWT) witch considered an ideal tool for this purpose, due to its suitability for the analysis of signal whose frequency spectrum is variable in time. So, it will study the problem of stator inter-turn short circuit. Simulation and experimentally results are presented in order to compare the performances of these techniques for inter-turn short circuit fault diagnosis.

Parameters Identification of the Induction Machine Using a Non Linear Parametric Technique

Abstract The aim of this paper is to determine the electrical parameters of induction machine experimentally by using a non linear parametric identification technique. This technique is based on the output error method and the Levenberg Marquardt algorithm, which guaranties the stability of the algorithm far from the optimum and a good convergence speed near the optimum. To assure the identification algorithm convergence, a specific excitation is added on the voltages which feed the induction machine instead of the SBPA noise signal. This excitation should not disturb the normal running of the motor and must be sufficient to assure a good convergence of the identification algorithm.