Mechanical Faults In Induction Motors Research Articles

Imbalanced Diagnosis Scheme for Incipient Rotor Faults in Inverter-Fed Induction Motors

Recently, fault diagnosing supervised classifiers have been widely proposed to diagnose both electric and mechanical faults in induction motors (IM). However, many of them require a large amount of data, which implies a great effort required for processing fault-related features and building the training set. Furthermore, in real-world datasets, it is required to deal with highly skewed data distributions, also known as class imbalance, which is a limiting issue and can misguide the tuning of machine learning algorithms. Resampling techniques based on a synthetic generation of minority class observations aim to address this problem. Last but not least is the fact that inverter-fed IM introduces undesired harmonics in the monitoring signal altering the diagnosis patterns. This diagnosis scheme is evaluated on experimental imbalanced data oriented to deal with the diagnosis of a rotor in situations where it is fed with an inverter. The results show how this imbalanced approach determines the actual diagnosis performance on a small amount of data. The experimental results demonstrate that balanced training sets built with class balancing techniques improve the classifier and therefore its performance for diagnosing incipient rotor faults in inverter-fed IM with studied and interpretable features recently proposed in this field of study.

A CNN-Based Methodology for Identifying Mechanical Faults in Induction Motors Using Thermography

Infrared thermography (IRT) has become an interesting alternative for performing condition assessments of different types of induction motor (IM)-based equipment when it operates under harsh conditions. The reported results from state-of-the-art articles that have analyzed thermal images do not consider (1): the presence of more than one fault, and (2) the inevitable noise-corruption the images suffer. Bearing in mind these reasons, this paper presents a convolutional neural network (CNN)-based methodology that is specifically designed to deal with noise-corrupted images for detecting the failures that have the highest incidence rate: bearing and broken bar failures; moreover, rotor misalignment failure is also considered, as it can cause a further increase in electricity consumption. The presented results show that the proposal is effective in detecting healthy and failure states, as well as identifying the failure nature, as a 95% accuracy is achieved. These results allow considering the proposal as an interesting alternative for using IRT images obtained in hostile environments.

Comparison of acoustic signal–based fault detection of mechanical faults in induction motors using image classification models

This study presents a novel deep neural network–based method for fault detection in induction motors. The focus was on identifying five types of mechanical cases: normal operation, shaft/load breakage, misalignment, mounting bolt looseness, and cooling fan problems. To increase the realism of the results, a laboratory-collected dataset of stereo microphone recordings was augmented with real factory noise. The audio data was transformed into image data using Mel-frequency cepstral coefficients as the feature extraction method and then processed with image-based classifiers. A comparison was made among 12 different networks in terms of accuracy and number of parameters, revealing that Mobilenet_v2, EfficientNetV2B0, and NASNetMobile had the best performance in terms of both network size and accuracy.

Rapid evaluation of the mechanical fault severity in induction motors using the model‐based diagnosis technique

The motor current signal analysis (MCSA) technique is widely used as a non-invasive method for detecting mechanical faults in induction motors by capturing characteristic components in the stator line current. However, the threshold of the characteristic component is not clear now, which makes it difficult for MCSA to judge whether the fault occurs or evaluates the mechanical fault severity. The existing model-based evaluation methods cannot meet the requirement of online condition monitoring because of their slow calculation. To solve these problems, a simplified dynamic motor model under any type of mechanical fault is established, and a formula for the amplitude relationship between the radial vibration of the rotor and fault-related component in the stator line current is derived. The radial vibration amplitude is related to mechanical fault severity. Using this formula, the MCSA technique can rapidly evaluate the mechanical fault severity according to the amplitude of the characteristic component in the collected stator current. The simulation study results demonstrate the accuracy of the simplified model and formula. The experimental results presented for condition monitoring in a real induction motor clearly validate the evaluation approach.

A PCA and Two-Stage Bayesian Sensor Fusion Approach for Diagnosing Electrical and Mechanical Faults in Induction Motors

Induction motors are widely used in industrial plants for critical operations. Stator faults, bearing faults, or rotor faults can lead to unplanned downtime with associated cost and safety implications. Different sensors may be used to monitor the health state of induction motors with each sensor typically being better suited for diagnosing different faults. Condition monitoring approaches that fuse data from multiple sensors have the potential to diagnose a greater number of faults. In this paper, a sensor fusion approach based on the combination of a two-stage Bayesian method and principal component analysis (PCA) is proposed for diagnosing both electrical and mechanical faults in induction motors. Acoustic, electric, and vibration signals are gathered from motors operating under different loading conditions and health states. The inclusion of the PCA step ensures robustness to varying loading conditions. The obtained results highlight that the proposed method performs better than the equivalent single-stage or feature-based Bayesian methods.

Machine Learning-Based Fault Diagnosis for Single- and Multi-Faults in Induction Motors Using Measured Stator Currents and Vibration Signals

In this paper, a practical machine learning-based fault diagnosis method is proposed for induction motors using experimental data. Various single- and multi-electrical and/or mechanical faults are applied to two identical induction motors in lab experiments. Stator currents and vibration signals of the motors are measured simultaneously during experiments and are used in developing the fault diagnosis method. Two signal processing techniques, matching pursuit, and discrete wavelet transform, are chosen for feature extraction. Three classification algorithms, support vector machine (SVM), K-nearest neighbors (KNN), and ensemble, with 17 different classifiers offered in MATLAB Classification Learner toolbox are used in the study to evaluate the performance and suitability of different classifiers for induction motor fault diagnosis. It is found that five classifiers (fine Gaussian SVM, fine KNN, weighted KNN, bagged trees, and subspace KNN) can provide near 100% classification accuracy for all faults applied to each motor, but the remaining 12 classifiers do not perform well. A novel curve fitting technique is developed to calculate features for the motors that stator currents or vibration signals under certain loadings are not tested for a particular fault. The proposed fault diagnosis method can accurately detect single- or multi-electrical and mechanical faults in induction motors.

A polytopic LPV approach to active fault tolerant control system design for three-phase induction motors

ABSTRACTThis paper presents a new active approach in order to design a model-based fault-tolerant control (FTC) system for three-phase induction motors (IMs) subjected to mechanical faults caused by stator and rotor failures. In this approach, a nominal controller achieves control objectives in fault-free case. The main part of this approach is related to design a fault/state observer for simultaneous estimation of additive faults which model the mechanical faults and axial fluxes in any operating conditions. For trade-off between robustness and performance in the presence of uncertainties and faults, this fault/state observer design problem is reduced to a well-known mixed optimisation problem by using polytopic linear parameter-varying approach. The FTC system consists of a nominal controller together with an observer which provides auxiliary inputs using the estimation of additive faults in order to compensate their undesirable effects on IM performance. The simulation results are shown to illustrate the effectiveness of the proposed approach to compensate the mechanical faults in IM.

Fault Detection of the Electrical Motors Based on Vibration Analysis

Abstract Detection and diagnosis of faults while the system is running can help to reduce all kind of losses. Mechanical faults, usages, slackness's, cause different noises and vibrations with different amplitude and frequency against the normal sound and movement of the equipment. Induction motors are present in any processes and systems and for this it is important to known the types of faults which can occur during its working. This paper presents briefly some electrical and mechanical faults in induction motors and how these faults influence the motor vibration. Each type of fault, electrical faults or mechanical faults, produces a vibration whit a specific frequency. Through monitoring and analyzing the vibration spectrum these specific frequencies can be detect and through these we can detect usages, slackness and forces which act and produce the fault. It is an undestroyed method to find out the functioning problems. If the amplitude of these vibrations reach to a certain level the fault can be detect and identify. The paper presents a possible two degree of freedom model for electrical motors. Motion equations and transfer function are deduced and through simulation can be followed how the vibration spectrum is changing for some important faults, like electrical imbalance, mechanical unbalance, looseness, bearings, external effects, base structure or resonance, which can be reduce to mass or spring variation in the system model. To obtain the behaviour of the system, the equations of the motion and Matlab simulations was used.

Detection of Mechanical Imbalances of Induction Motors with Instantaneous Power Signature Analysis

Mechanical imbalances are common mechanical faults in induction motors. Vibration monitoring techniques have been widely used for the diagnosis of mechanical faults in induction motors, but electrical detection methods have been preferred in recent years. For many years, researchers have concentrated on the Motor Current Signature Analysis (MCSA). This paper examines the effect of mechanical imbalances to induction machine electrical parameters. Instantaneous Power Signature Analysis (IPSA) technique used to detect these faults. In the paper, a full analysis of the proposed technique is presented, and experimental results for healthy and faulty motors have been shown and discussed.

Mechanical Load Fault Detection in Induction Motors by Stator Current Time-Frequency Analysis

This paper examines the detection of mechanical faults in induction motors by an original use of stator current time-frequency analysis. Mechanical faults lead generally to periodic load torque oscillations. The influence of the torque oscillations on the induction motor stator current is studied using an analytical approach. The mechanical fault results in a sinusoidal phase modulation of the stator current, which is equivalent to a time-varying frequency. Based on these assumptions, several signal processing methods suitable for stator current signature analysis are discussed: classical spectral analysis, instantaneous frequency estimation, and the Wigner distribution. Experimental and simulation results validate the theoretical approach in steady-state operating conditions