Diagnosis Of Electrical Machines Research Articles

Modeling, Control and Diagnosis of Electrical Machines and Devices

Nowadays, the increasing use of electrical machines and devices in more critical applications has driven the research in condition monitoring and fault tolerance [...]

The Edge Application of Machine Learning Techniques for Fault Diagnosis in Electrical Machines.

The advent of digitization has brought about new technologies that enable advanced condition monitoring and fault diagnosis under the Industry 4.0 paradigm. While vibration signal analysis is a commonly used method for fault detection in literature, it often involves the use of expensive equipment in difficult-to-reach locations. This paper presents a solution for fault diagnosis of electrical machines by utilizing machine learning techniques on the edge, classifying information coming from motor current signature analysis (MCSA) for broken rotor bar detection. The paper covers the process of feature extraction, classification, and model training and testing for three different machine learning methods using a public dataset to then export the results to diagnose a different machine. An edge computing approach is adopted for the data acquisition, signal processing and model implementation on an affordable platform, the Arduino. This makes it accessible for small and medium-sized companies, albeit with the limitations of a resource-constrained platform. The proposed solution has been tested on electrical machines in the Mining and Industrial Engineering School of Almadén (UCLM) with positive results.

Cover Image

The cover image is based on the Review Article Airgap and stray magnetic flux monitoring techniques for fault diagnosis of electrical machines: An overview by Jawad Faiz et al., https://doi.org/10.1049/elp2.12157.

Airgap and stray magnetic flux monitoring techniques for fault diagnosis of electrical machines: An overview

Flux-based fault diagnosis methods have attracted ever-increasing attention from technical communities because of their potential diagnostic capabilities and ability to overcome weaknesses of the traditional methods such as motor current signature analysis (MCSA). This study reviews recent literature on these methods and addresses their strengths and limitations. Different types of flux sensors that can be used for fault diagnosis applications are also discussed. Flux-based approaches are classified based on the type of utilised flux into internal (airgap flux) and external (stray flux) methods and compared with each other. The most common faults in both induction machines (IMs) and permanent magnet machines (PMMs), including inter-turn short-circuit (ITSC), airgap eccentricity (AGE), rotor broken bars (BRB), and irreversible demagnetisation (IDM) faults, are considered in this review. Finally, research gaps are identified, and some suggestions for future research in this area are provided.

Software Sensors for Order Tracking Applied to Permanent Magnet Synchronous Generator Diagnostics: A Comparative Study

The paper deals with software sensors which facilitates the diagnosis of electrical machines in non-stationary operating conditions. The technique targeted is order tracking for which different techniques exist to estimate the speed and angle of rotation. However, from a methodological point of view, this paper offers a comparison of several methods in order to evaluate their performance from tests on a test bench. In addition, to perform the tests, it is necessary to initialize the different methods to make them work correctly. In particular, an identification technique is proposed as well as a way to facilitate initialization. The example of this paper is that of a synchronous generator. Angular sampling allows the spectrum to be stationary and the interpretation of a possible defect. The realization of the angular sampling and the first diagnostic elements require the knowledge of two fundamental quantities: the speed of rotation and the angular position of the shaft. The estimation of the rotation speed as well as the estimation of the angular position of the shaft are carried out from the measurement of an electric current (or three electric currents and three voltages). Four methods are proposed and evaluated to realize software sensors: identification technique, PLL (Phase Locked Loop), Concordia transform and an observer. The four methods are evaluated on measurements carried out on a test bench. The results are discussed from the diagnosis of a mechanical fault.

Early Detection of Broken Rotor Bars in Inverter-Fed Induction Motors Using Speed Analysis of Startup Transients

The fault diagnosis of electrical machines during startup transients has received increasing attention regarding the possibility of detecting faults early. Induction motors are no exception, and motor current signature analysis has become one of the most popular techniques for determining the condition of various motor components. However, in the case of inverter powered systems, the condition of a motor is difficult to determine from the stator current because fault signatures could overlap with other signatures produced by the inverter, low-slip operation, load oscillations, and other non-stationary conditions. This paper presents a speed signature analysis methodology for a reliable broken rotor bar diagnosis in inverter-fed induction motors. The proposed fault detection is based on tracking the speed fault signature in the time-frequency domain. As a result, different fault severity levels and load oscillations can be identified. The promising results show that this technique can be a good complement to the classic analysis of current signature analysis and reveals a high potential to overcome some of its drawbacks.

An accurate inter-turn short circuit faults model dedicated to induction motors

Safety, disponibility and continuity of industrial systems are major issue in maintenance. In the last decades, these points are the important axes in the field of research. In fact, in many industrial processes research has picked up a fervent place and a particular importance in the area of fault diagnosis of electrical machines, in fact, a fault prognosis has become almost indispensable. The need of a mathematical model of three-phase induction machine, suitable for the simulation of machines behaviour under fault conditions, has received considerable attention. The paper presents a new practical and more precise model for induction motors after introducing inter turn short circuits faults. The proposed model is based on coupled magnetic circuit theory, capable to take into account any electrical asymmetry conditions. To verify the exactitude and the effectiveness of the model, simulation results for induction machine under interturn short circuit fault are presented. In spite of its simplicity, the proposed model is able to provide useful indications for diagnostic purposes. Experimental study is presented at the end of the paper to show that the proposed model predicts the induction machine behavior with a good accuracy.

An Enhanced Binary Particle Swarm Optimization for Optimal Feature Selection in Bearing Fault Diagnosis of Electrical Machines

This study proposes an effective bearing fault diagnosis model based on an optimized approach for feature selection. The measured signal of the electric motor is processed by envelope analysis and Hilbert-Huang transform techniques to extract the potential features. An enhancement of the binary particle swarm optimization algorithm through population initialization strategy based on feature weights, new updating mechanism, and the screening and replacing process create a new and effective feature selection method that improves classification accuracy and reduces data size. The optimal feature subset is provided separately for artificial neural networks, and support vector machine classifier for the final recognition task. In multiple case studies, the proposed feature selection method is evaluated against the benchmark data sets and shows performance comparable to that of other peer competitors. The effectiveness of the proposed bearing fault diagnosis model is verified on different testbeds and achieves high accuracy and robustness under noise conditions. In addition, experimental results are compared with existing fault diagnostic models, showing the high possibility of the proposed bearing fault diagnosis model.

Machinery Failure Approach and Spectral Analysis to Study the Reaction Time Dynamics over Consecutive Visual Stimuli: An Entropy-Based Model

The reaction times of individuals over consecutive visual stimuli have been studied using an entropy-based model and a failure machinery approach. The used tools include the fast Fourier transform and a spectral entropy analysis. The results indicate that the reaction times produced by the independently responding individuals to visual stimuli appear to be correlated. The spectral analysis and the entropy of the spectrum yield that there are features of similarity in the response times of each participant and among them. Furthermore, the analysis of the mistakes made by the participants during the reaction time experiments concluded that they follow a behavior which is consistent with the MTBF (Mean Time Between Failures) model, widely used in industry for the predictive diagnosis of electrical machines and equipment.

Artificial Intelligence-Based Fault Diagnosis for Condition Monitoring of Electric Motors

In the era of globalization, manufacturing industries are facing intense pressure to prevent unexpected breakdowns, reduce maintenance cost and increase plant availability. Induction motors are the most sought-after prime movers in modern-day industries due to their robustness. Recently, research has picked up a fervent pace in the area of fault diagnosis of electrical machines. This paper presents the application of Support Vector Machine (SVM) and Artificial Neural Network (ANN)-based system to diagnose the vibration and Instantaneous Power (IP)-based responses of rolling element bearings and broken rotor bars in an induction motor. The dimensionality of the extracted features was reduced using Principal Component Analysis (PCA) and thereafter the selected features were ranked in order of relevance using the Sequential Floating Forward Selection (SFFS) method for reducing the size of input features and finding the most optimal feature set. A comparative analysis of the effectiveness of SVM and ANN is carried out using statistical parameters extracted from vibration and IP signals. The highest accuracy of 92.5% and 98.2% was achieved for vibration and IP signatures, respectively, using the proposed SFFS-based feature selection technique and ANN classification method. The results reveal that ANN has better performance than SVM and the proposed strategy can be used for automatic recognition of machine faults. The use of this type of intelligent system helps in avoiding unwanted and unplanned system shutdowns due to the failure of the motor.

Application of unified microprocessor relay protection units in electrical machine diagnostics

The article concentrates on the topical issue connected with the detection of defects in the rotor winding of electric machines. The experimental sampling and digital processing of electrical signals from controlled windings is in the basis. The authors conducted the research using two experimental units for asynchronous motor and synchronous generator, respectively. A distinctive feature of the proposed research stands is in solving the problems of digital processing and data analysis on the basis of the application of microprocessor relay protection unit BMRZ developed in Russia. The authors applied method of wavelet decomposition to select the detailing component. They also presented the results of experiments for the breakage in the short-circuit rotor of an asynchronous motor and proved that the microprocessor-based BMRZ device is capable of digitizing at a sampling rate that meets the requirements, and in conjunction with the data processing algorithms that carry out the selective determination of hard-to-detect defects, is applicable in the diagnosis of electrical machines.

Higher-Order Spectral Analysis of Stray Flux Signals for Faults Detection in Induction Motors

Abstract This work is a review of current trends in the stray flux signal processing techniques applied to the diagnosis of electrical machines. Initially, a review of the most commonly used standard methods is performed in the diagnosis of failures in induction machines and using stray flux; and then specifically it is treated and performed the algorithms based on statistical analysis using cumulants and polyspectra. In addition, the theoretical foundations of the analyzed algorithms and examples applications are shown from the practical point of view where the benefits that processing can have using HOSA and its relationship with stray flux signal analysis, are illustrated.

Multiple coupled circuit modelling approach for squirrel cage induction machine under single‐broken‐bar fault with stator winding functions decomposed in d – q rotor reference frame

Mathematical modelling is a fundamental issue in the condition monitoring and fault diagnosis of electrical machines. This study presents a multiple coupled circuit approach for the modelling of squirrel cage induction machines (SCIMs) with a single broken bar. Derived by means of winding function, the model is capable of incorporating all the harmonics of air-gap magnetomotive force and describing transient machine behaviours. A d–q reference frame fixed on rotor is introduced. Each stator winding function is decomposed into two parts, a q-axis and a d-axis component, both of which induce currents in each rotor mesh. The q-axis and d-axis equivalent circuits are developed for both healthy and faulty cage rotors. It is also shown that a single-broken-bar fault does not affect the q-axis rotor mesh current distribution. The d-axis equivalent circuit for healthy cage rotor is modified further to determine the deviations in the mesh currents due to bar breakage. A detailed model of a 2.2 kW, four-pole, three-phase, 28-rotor-bar SCIM is implemented in MATLAB/Simulink environment, and a laboratory test bed is set up for this machine. A comparison between the simulation and experimental results verifies the effectiveness of the proposed modelling approach.

Selecting the Parameters of The Diagnosis of Frame Insulation Condition in Electrical Machines of Locomotives

The purpose of this study is to determine the signs displaying the insulation ageing process and allow predicting the change in its technical condition under operation of traction motors in locomotives. In order to choose the indicators of the insulation condition, the methods of linear electrical circuits were used to analyze the equivalent scheme of frame insulation. In the equivalent scheme, the dielectric micro- and macro-heterogeneities were taken into account. The most informative parameters of the frame insulation ageing process are those related to absorption and resorption processes over time. The return voltage has been selected as the main characteristic to determine the indicators of insulation condition. The analytical dependences for determining the frame insulation condition indicators have been obtained different from those currently used in the diagnosis of electrical machines. The choice of indicators will allow predicting changes in the technical condition of frame insulation in operation. The studies will allow us to change the regular preventive maintenance technology to the maintenance based on the facility condition.

Monitoring of the electrical resistance of a three phase machine applied for the vehicle

In the field of electrical machine diagnosis the finding of faults in the stator of the electrical machine is one of the greatest challenges. Stator faults have many possible reasons. They may be generated by shorts in the winding or circuit-breaks in one or more coils. In order to find that kind of faults, it is usually necessary to remove the electrical machine from the vehicle and to analyse the machine with external measurement devices. This approach is very expensive, time consuming and unnecessary, in the case that there is no fault. In the following paper, a method is described how the electrical resistances of the stator coils is estimated for diagnostic purposes. For the application of this method only the measurement system is used, which is usually installed in the vehicle. With using the internal sensors in the machine the cost and the time can be reduced. An removal of the electrical machine may be prevented and the using of external measuring equipment can be avoided.

Assessment of Time Frequency Color Index for Electrical Machines Diagnosis and Fault Severity

Improving availability and reliability of Induction Machine (IM) implies the minimization and the prediction of the maintenance operations. In this paper, we focus on detection and location of the rotor electrical faults in IM. This kind of fault generates non-stationary signatures in the stator current of the machine. Advanced signal processing techniques are required to deal with that type of signals. In this context, several recent studies suggest the use of Time-Frequency Representations (TFR) for the diagnosis of IMs' faults. This work presents a comparative study of some linear TFR. This last relies on good experimental results conducted on a test bench using a wound rotor induction machine operating in no load condition. The purpose is to showcase the TFR performances in faults detection. Static indicator is used to check the fault severity.

Industrial drive fault diagnosis through vibration analysis using wavelet transform

The fault diagnosis of electrical machines is significant to reduce the costs of maintenance through early detection of faults, which could be expensive to service. The idea of the research is that the advanced signal processing techniques called wavelet transform is exercised to haul out the faults from the vibration and current signal and a spectral component is being received to diagnose the condition of the machine. The existing system used wavelet transforms for the analysis of the stator faults and rotor faults are taken as a case study to prove the wavelet techniques for fault diagnosis. The proposed investigation of vibration analysis for the induction machines is done through frequency pattern using the decomposition of wavelet packet. The wavelet coefficients for the vibration have been extracted over a wide range of signals and the analysis could be percolated on the frequency domain through HAAR wavelet. In this paper, healthy and unhealthy motor with two broken ball bearings are used to estimate the vibration and current spectrum. Experimentally observed vibration and current signals are transformed into power spectral density through approximate and detailed coefficients with the help of MATLAB tools and the steady state rotor frequency was used to introduce a new frequency pattern for fault diagnosis.

Automatic selection of frequency bands for the power ratios separation technique in partial discharge measurements: part II, PD source recognition and applications

Partial discharge (PD) measurement is a widely extended technique for the diagnosis of electrical machines and power cables. Since PD and noise recognition is really important in industrial environments, the authors proposed the use of the spectral Power Ratios (PR) of the detected high-frequency (HF) signals, which show promising results. However, the frequency bands selected for their calculation have a clear influence on the position of the clusters in the classification maps. Following this research trend, a new algorithm for the automatic selection of the frequency bands which gives a proper separation between the clusters, has been proposed in part I of this paper. The first results presented in this paper revealed a good behavior of the technique for PD and noise separation in simple test objects. In this second part of the paper, more results from the application of this system to PD source recognition in simple test objects and also in real equipment are developed.

Reconfigurable instrument for neural‐network‐based power‐quality monitoring in 3‐phase power systems

From voltage and current signals it is possible to obtain relevant information for solving some problems in several industrial and scientific applications as power quality (PQ) monitoring, monitoring and diagnosis of electrical machines, electric systems protection and control. At present, the PQ monitoring, measure through a set of PQ indices (PQI), is an important topic for the industrial sector since a poor PQ, characterised by the presence of harmonics in the power line, produces irregular or wrong operation of protection systems, excessive neutral currents in 3-phase four-wire systems, overheating of motors, transformers, capacitor banks and wiring in general. The PQI calculation is performed by many techniques proposed in the literature; however, they do not have either good performance for transient signals or the requirements for satisfying the power standards. This work proposes the assessment of the PQI-based in neural networks for transient or stationary signals in 3-phase power systems without losing the power standard requirements. Besides, this work contributes to the industrial application field by allowing the continuous and online monitoring of the PQI thanks to the field programmable gate array implementation of the proposed methodology.

Bearing Fault Detection by a Novel Condition-Monitoring Scheme Based on Statistical-Time Features and Neural Networks

Bearing degradation is the most common source of faults in electrical machines. In this context, this work presents a novel monitoring scheme applied to diagnose bearing faults. Apart from detecting local defects, i.e., single-point ball and raceway faults, it takes also into account the detection of distributed defects, such as roughness. The development of diagnosis methodologies considering both kinds of bearing faults is, nowadays, subject of concern in fault diagnosis of electrical machines. First, the method analyzes the most significant statistical-time features calculated from vibration signal. Then, it uses a variant of the curvilinear component analysis, a nonlinear manifold learning technique, for compression and visualization of the feature behavior. It allows interpreting the underlying physical phenomenon. This technique has demonstrated to be a very powerful and promising tool in the diagnosis area. Finally, a hierarchical neural network structure is used to perform the classification stage. The effectiveness of this condition-monitoring scheme has been verified by experimental results obtained from different operating conditions.