Chapter

Bearings Fault Detection Using Inference Tools

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Regarding the acquisition system, it is based on four different sensors connected to a main acquisition device. A triaxial shear design MMF branded piezoelectric accelerometer model KS943B.100 with IEPE (Integrated Electronics Piezo Electric) standard output and linear frequency response from 0.5 Hz to 22 kHz, was attached using stud mounting to the drive-end motor end-shield and its data was collected at 20kS/s during 1 second for each measurement. Phase stator currents were acquired using Hall effectTektronix A622 probes with a frequency range from DC to 100 kHz and collected at 20 kHz during 1 second for each measurement. High frequency CMC signal was measured at the cable attached to the bearings housing with a Tektronix TCPA300 amplifier and TCP303 current probe, which provides up to 15 MHz of frequency range, and acquired at 50 MHz during 100 ms for each measurement. Acoustic emissions were acquired with the use of a Vallen-Systeme GmbH VS-150M sensor unit with a range from 100 kHz to 450 kHz and resonant at 150 kHz. A Vallen-Systeme GmbH AEP4 40dB preamplifier was used before data acquisition at a sampling frequency of 25MS/s during 20ms each measurement. All the described sensors are connected to a PXI acquisition system from National Instruments formed by different specific boards.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Bearing defects can be grouped as "distributed" or "localized" [5]- [7]. Distributed defects normally caused by manufacturing error, improper installation or abrasive wear. ...
Article
Full-text available
Vibration and acoustic emission (AE) signal monitoring are popular techniques for detecting bearing fault, the main cause in induction motor failure which can lead to catastrophic damage. This paper presents comparison between vibration and AE signal monitoring as a tool for induction motor bearing fault detection. The effectiveness of time-domain analysis is compared with frequency-domain. Statistical parameters used in time-domain include RMS, crest factor, and kurtosis whereas for frequency-domain, normal spectrum and envelope spectrum using Hilbert transform are applied. The results reveal that vibration and AE signals are effective measurement to detect bearing fault in both time-and frequency-domain.
... Many techniques have been used for rolling element bearing fault diagnosis and can be classified depending on the type of measurement involved. These can be, for example, vibration, temperature (Patil et al. 2010) or acoustic measurements (Delgado et al. 2011). ...
Article
Full-text available
Rolling element bearing defects are among the main reasons for the breakdown of electrical machines, and therefore, early diagnosis of these is necessary to avoid more catastrophic failure consequences. This paper presents a novel approach for identifying rolling element bearing defects in brushless DC motors under non-stationary operating conditions. Stator current and lateral vibration measurements are selected as fault indicators to extract meaningful features, using a discrete wavelet transform. These features are further reduced via the application of orthogonal fuzzy neighbourhood discriminative analysis. A recurrent neural network is then used to detect and classify the presence of bearing faults. The proposed system is implemented and tested in simulation on data collected from an experimental setup, to verify its effectiveness and reliability in accurately detecting and classifying the various faults.
... Based on the analysis of the origin of induction motor failures, the bearing fault is the major source of most mechanical faults. To show the importance of bearing fault detection, induction motors failures distribution is depicted in Fig. 1 [12,13]. Therefore, the bearing fault detection and troubleshooting in the early stages will decrease the cost of unwanted shutdown [14]. ...
Article
Full-text available
This paper proposes an intelligent method based on artificial neural networks (ANNs) to detect bearing defects of induction motors. In this method, the vibration signal passes through removing non-bearing fault component (RNFC) filter, designed by neural networks, in order to remove its non-bearing fault components, and then enters the second neural network that uses pattern recognition techniques for fault classification. Four different categories include; healthy, inner race defect, outer race defect, and double holes in outer race are investigated. Compared to the regular fault detection methods that use frequency-domain features, the proposed method is based on analyzing time-domain features which needs less computational effort. Moreover, machine and bearing parameters, and the vibration signal spectrum distribution are not required in this method. It is shown that better results are achieved when the filtered component of the vibration signal is used for fault classification rather than common methods that use directly vibration signal. Experimental results on three-phase induction motor verify the ability of the proposed method in fault diagnosis despite low quality (noisy) of measured vibration signal.
Conference Paper
This paper gives an approach based on a fitting technique to detect faulty bearings by employing artificial neural networks (ANNs). At present, many companies follow very expensive, scheduled maintenance procedures for checking the bearing condition of the electric drives. Therefore, there is considerable demand to reduce the maintenance costs and to prevent unscheduled downtime of electrical drives in company. To overcome the above crisis, online monitoring of vibrations in electric motors at running condition can be implemented. Most of the vibration measuring equipment measure the total vibration of the electric motor. Deducing the bearing fault from the total vibration value of the motor is not accurate. Hence it is quintessential to split the total vibration value of the motor into its sub-components namely bearing vibration and the vibration due to loading. This splitting could visualize the exact occurrence of fault in the machine. For the known loading conditions of the motor, the total vibration, bearing vibration and the vibration due to loading are tabulated. Some tabulated data are used to train the ANNs for the known machine conditions. Once trained, the ANNs are examined using the remaining data. The final evaluated ANNs can readily split the total vibration value into its sub-components enabling the computation of component value. These sub-component vibration values are analyzed individually for abnormalities to find the bearing fault accurately.
Conference Paper
Efficient detection of bearing failure is one of the most investigated engineering issues in several industries. The proposed method, based on High Frequency Resonance Technique (HFRT) allows detecting precise location of bearing's failure. The location of resonance area is computed by screening the signal's frequency spectrum. Some statistical features are adopted to get a preliminary inspection of the raw signal. This method is validated with Seeded Fault Test Data provided by Case Western Reserve University Bearing Data Center.
Conference Paper
Full-text available
Motor current signature analysis has been successfully used in induction machines for fault diagnosis. The method however does not always achieve good results when the load torque is not constant. This paper proposes a new approach to motor fault detection, by wavelet decomposition analysis. Theoretical development and experimental results are presented to support the research.
Conference Paper
Full-text available
The present work shows a condition monitoring system applied to detect Fault Condition in EMA Systems. Removal of the engine hydraulic pumps requires fully-operative electrical power actuators and mastery of the flight control architecture. However, unexpected faults and lack of safety hinder the massive use of EMAs in flight control actuators and force to develop new systems and methods for supervision in aircrafts actuators. This research covers wide-range irregularities which are very often more difficult to analyse. In addition to traditional techniques like vibration and current signals, high frequency current bearing pulses and acoustic emissions are also analysed. A multivariable fuzzy inference analysis approach is presented to get around the diagnostic difficulty.
Conference Paper
Full-text available
Motor Current Signature Analysis (MCSA) is being the most widely used method to identify faults in Induction Motors. MCSA focuses its effort in the spectral analysis of the stator current, by analysis of the resulting harmonics when the machine is working under failure state. Motors faults such as broken rotor bars, bearing damage and eccentricity of the rotor axis can be detected. However, the method presents some problems at low speed and low torque, mainly due to the proximity between the frequencies to detect and the small amplitude of the resulting harmonics respectively. In both cases, the problem of frequency accuracy is very tricky since the side band harmonic is closed to the fundamental. Since the accuracy is reverse proportional to the data acquisition time, the spectrum read could be meaningless. Therefore, it is difficult to evaluate the correct magnitude of the side band harmonics even when windowing. This paper proposes to inject into the machine under test an additional voltage at frequency different from the fundamental, and then to study the resulting harmonics around the new frequencies appearing due to composition between injected and main frequencies.
Article
Full-text available
The high-frequency circulating bearing current that may occur in machines of inverter-based drive systems can be described by an eddy-current model. The parameters of an equivalent circuit are derived from the model. The ratio between bearing current and common-mode current amplitudes for different machines is calculated. The theoretical maximum ratio is about 0.35. Copper loops applied for bearing current measurement may decrease the circulating bearing currents up to almost 40%
Article
Full-text available
In this survey paper, we show by straightforward mathematical scaling considerations that small inverter-fed ac motors of up to typical 20 kW at 1500/min are likely to suffer from discharge (electric discharge machining) bearing currents, whereas larger motors are likely to be subjected to high-frequency (HF) circulating bearing currents. When comparing motors that operate at the same voltage level, the resulting bearing-current density is high for very small and very large motors. Smaller values occur in between these two extremes with medium size motors in the range of 10-100 kW. Therefore, electric bearing insulation is useful for larger motors to interrupt the HF circulating bearing-current path, whereas small motors need rotor shielding, common-mode voltage filters, or hybrid bearings.
Article
Full-text available
The occurrence of high-frequency discharge bearing currents (also called "electric discharge machining (EDM) currents") that can occur in machines of inverter-based drive systems depends strongly on the value of the capacitive voltage divider "bearing voltage ratio" (BVR) of the machine. This paper presents a straightforward approach to calculate the involved motor capacitances in order to predict the voltage across the bearing and, thereby, the likeliness of such EDM currents that are caused by this voltage to occur. The approach is based on the design parameters of the machine and does not involve additional measurements. It combines ease of application and satisfying accurateness, and allows prediction of the risk of endangerment of an inverter-based drive system due to discharge bearing currents
Article
Full-text available
Most condition monitoring techniques for rolling element bearings are designed to detect the four characteristic fault frequencies. This has lead to the common practice of categorizing bearing faults according to fault location (i.e., inner race, outer race, ball, or cage fault). While the ability to detect the four characteristic fault frequencies is necessary, this approach neglects another important class of faults that arise in many industrial settings. This research introduces the notion of categorizing bearing faults as either single-point defects or generalized roughness. These classes separate bearing faults according to the fault signatures that are produced rather than by the physical location of the fault. Specifically, single-point defects produce the four predictable characteristic fault frequencies while faults categorized as generalized roughness produce unpredictable broadband changes in the machine vibration and stator current. Experimental results are provided from bearings failed in situ via a shaft current. These results illustrate the unpredictable and broadband nature of the effects produced by generalized roughness bearing faults. This issue is significant because a successful bearing condition monitoring scheme must be able to reliably detect both classes of faults.
Article
Results of a survey conducted in 1982 of the reliability of large motors have been presented and published in two parts [1], [2], These results have generated numerous questions and comments and, consequently, the need to further analyze the data of the survey was recognized. Part 1 presents general results based on categories of motor types and applications specifically requested in the survey questionnaire. Part 2 combines various categories and addresses some questions resulting from Part 1. Part 3 of the survey results is presented here to address new questions and comments and to add more specific analyses of areas not yet explored. These results, along with Parts 1 and 2, provide the complete complement of analysis to date. Copyright © 1987 by The Institute of Electrical and Electronics Engineers, Inc.
Conference Paper
Electrical equipments are the workhorses of industry; their failure may result in complete shut down of a plant or even cause an unexpected disaster. Researchers had pursued rigorously various diagnostic approaches for electrical machines. Apart from analyzing the conventional vibration, current, voltage signals people are trying to explore fault signatures from torque, power, speed, flux etc. Methods such as offline/online, with/without additional sensor, model-based, signal-based etc. are being explored vastly. A number of signal processing techniques and fault detection decision-making tools are being reported frequently. Undoubtedly this field is vast in scope. Hence keeping this in mind to avoid repetition as well to facilitate future research a brief review is presented in this paper. Nearly 80% electrical motors used in industries are induction motors and hence industries depend on the performance of them to a great extend. This paper will mainly concentrate on induction machines with a very brief review of other machines.
Conference Paper
The most bearings diagnosis methodologies has been designed typically by means of vibration analysis, but a numerous studies in the last years are focused in non-invasive techniques Motor Current Signature Analysis (MCSA) based. This paper shows two different methodologies for bearings diagnosis over induction motors by means of the stator Common Mode Current (CMC). It is acquired directly the CMC from the source that inverter provide to the motor. The first analysis is performed directly over a temporal acquisition of stator CMC. The second analysis is based in the spectral analysis of a single CMC discharge. It is shown also the development of a CMC probe, which corresponds to the tool used to check the first methodology. Finally, it is presented a vibration analysis over the same induction motors in order to confirm the bearings diagnosis extracted from the previous methodologies CMC based.
Conference Paper
Main cause for bearing failure are the common mode circulating currents. Those currents appear due to the use of inverters to drive electric motors. This paper presents a new method for fault detection by measuring common-mode currents. It is also presented the development of an electronic device which allows fault detection. The development has been tested in a test bench showing promising results to detect bearing faults in induction machines.
Article
Current-based monitoring can offer significant economic savings and implementation advantages over traditional vibration monitoring for bearing fault detection. The key issue in current-based bearing fault detection is to extract bearing fault signatures from the motor stator current. Since the bearing fault signature in the stator current is typically very subtle, particularly when the fault is at an incipient stage, it is difficult to detect the fault signature directly. Therefore, in this paper, the bearing fault signature is detected alternatively by estimating and removing nonbearing fault components via a noise cancellation method. In this method, all the components of the stator current that are not related to bearing faults are regarded as noise and are estimated by a Wiener filter. Then, all these noise components are cancelled out by their estimates in a real-time fashion, and a fault indicator is established based on the remaining components which are mainly caused by bearing faults. Machine parameters, bearing dimensions, nameplate values, and the stator current spectrum distribution are not required in the method. The results of online experiments with a 20-hp induction motor under multiple load levels have confirmed the effectiveness of this method.
Conference Paper
The aim of this paper is to verify the effectiveness of the stator current analysis for the detection of bearing problems. The paper reports the experimental results on four different types of bearing faults.
Article
This paper investigates diagnostic techniques for electrical machines with special reference to induction machines and to papers published in the last ten years. A comprehensive list of references is reported and examined, and research activities classified into four main topics: 1) electrical faults; 2) mechanical faults; 3) signal processing for analysis and monitoring; and 4) artificial intelligence and decision-making techniques.
Conference Paper
Induction machines operated by inverter use to present efficiency decrease and sometimes present additional rotor and stator fault. FEM has been used for faulty motor simulation and shows results of motor fault effects. Experimental results corroborate the simulation and theoretical effects.
Article
The subject of machine condition monitoring is charged with developing new technologies to diagnose the machinery problems. Different methods of fault identification have been developed and used effectively to detect the machine faults at an early stage using different machine quantities, such as current, voltage, speed, efficiency, temperature and vibrations. One of the principal tools for diagnosing rotating machinery problems has been the vibration analysis. Through the use of different signal processing techniques, it is possible to obtain vital diagnostic information from vibration profile before the equipment catastrophically fails. A problem with diagnostic techniques is that they require constant human interpretation of the results. The logical progression of the condition monitoring technologies is the automation of the diagnostic process. The research has been underway for a long time to automate the diagnostic process. Recently, artificial intelligent tools, such as expert systems, neural network and fuzzy logic, have been widely used with the monitoring system to support the detection and diagnostic tasks. This paper reviews the progress made in electrical drive condition monitoring and diagnostic research and development in general and induction machine drive condition monitoring and diagnostic research and development, in particular, since its inception. Attempts are made to highlight the current and future issues involved for the development of automatic diagnostic process technology.
Article
This paper deals with an experimental investigation of fault diagnosis in a multistage gearbox under transient loads. An induction motor drives the multistage gearbox, which is connected to a DC generator for loading purpose. The signals studied are the vibration transients, recorded from an accelerometer fitted at the tail-end bearing of the gearbox; and the current transients drawn by the induction motor. Three defective cases and three transient load conditions are investigated. Advanced signal processing techniques such as discrete wavelet transform (DWT) and a corrected multiresolution Fourier transform (MFT) are applied to investigate the vibration and current transients. It is observed from the vibration transients that the load removal is a high-frequency phenomenon. With increase in defect severity, not only the defective gear mesh frequency gains energy, but also large impact energy appears in low-frequency regions. Whereas in the current transients, though load removal is a low-frequency phenomenon, a very small transient is observed at high-frequency regions for defective gears. With increase in defect severity, energy is distributed to the sidebands of the gear mesh frequency across supply line frequency. A statistical feature extraction technique is proposed in order to find a trend in detection of defects. A condition monitoring scheme is devised that can facilitate in monitoring vibration and current transients in the gearbox with simultaneous presence of transient loads and defects.
Article
This paper deals with a new scheme for the diagnosis of localised defects in ball bearings based on the wavelet transform and neuro-fuzzy classification. Vibration signals for normal bearings, bearings with inner race faults and ball faults were acquired from a motor-driven experimental system. The wavelet transform was used to process the accelerometer signals and to generate feature vectors. An adaptive neural-fuzzy inference system (ANFIS) was trained and used as a diagnostic classifier. For comparison purposes, the Euclidean vector distance method as well as the vector correlation coefficient method were also investigated. The results demonstrate that the developed diagnostic method can reliably separate different fault conditions under the presence of load variations.
Conference Paper
This paper presents experimental results for diagnosing faults in bearings with different raceway defects via motor current spectral analysis. This is an important contribution since experimental results for bearings with inner race faults are extremely sparse in the literature. This research indicates that detecting these fault frequencies in the motor current is significantly more difficult than detecting them in the motor vibration. Specifically, inner race faults are very difficult to detect by searching for their signatures in the current. This paper also shows that it is not easy to compare different bearings of the same type or even the same bearing in different installations, which makes it difficult to determine the sensitivity of current-based bearing fault detection. For this reason, the use of wider frequency ranges is suggested for current-based detection.
Conference Paper
This paper is intended as a tutorial overview of induction motors signature analysis as a medium for fault detection. The purpose is to introduce in a concise manner the fundamental theory, main results, and practical applications of motor signature analysis for the detection and the localization of abnormal electrical and mechanical conditions that indicate, or may lead to, a failure of induction motors. The paper is focused on the so-called motor current signature analysis (MCSA) which utilizes the results of spectral analysis of the stator current. The paper is purposefully written without “state of the art” terminology for the benefit of practicing engineers in facilities today who may not be familiar with signal processing
Conference Paper
This paper addresses the application of motor current spectral analysis for the detection of rolling-element bearing damage in induction machines. Vibration monitoring of mechanical bearing frequencies is currently used to detect the presence of a fault condition. Since these mechanical vibrations are associated with variations in the physical air gap of the machine, the air gap flux density is modulated and stator currents are generated at predictable frequencies related to the electrical supply and vibrational frequencies. This paper takes the initial step of investigating the efficacy of current monitoring for bearing fault detection by correlating the relationship between vibration and current frequencies caused by incipient bearing failures. The bearing failure modes are reviewed and the characteristic bearing frequencies associated with the physical construction of the bearings are defined. The effects on the stator current spectrum are described and the related frequencies determined. This is an important result in the formulation of a fault detection scheme that monitors the stator currents. Experimental results which show the vibration and current spectra of an induction machine with different bearing faults are used to verify the relationship between the vibrational and current frequencies. The test results clearly illustrate that the stator current signature can be used to identify the presence of a bearing fault
Article
The positive features of neural networks and fuzzy logic are combined together for the detection of stator inter-turn insulation and bearing wear faults in single-phase induction motor. The adaptive neural fuzzy inference systems (ANFISs) are developed for the detection of these two faults. These faults are created experimentally on a single-phase induction motor in the laboratory. The experimental data is generated for the five measurable parameters, viz, motor intakes current, speed, winding temperature, bearing temperature, and the noise of the machine. Earlier, the ANFIS fault detectors are trained for the two input parameters, i.e., speed and current, and the performance is tested. Later, the three remaining parameters are added and the five input ANFIS fault detector is trained and tested. It observed from the simulation results that the five input parameter system predicts more accurate results
Article
This paper is intended as a tutorial overview of induction motors signature analysis as a medium for fault detection. The purpose is to introduce in a concise manner the fundamental theory, main results, and practical applications of motor signature analysis for the detection and the localization of abnormal electrical and mechanical conditions that indicate, or may lead to, a failure of induction motors. The paper is focused on the so-called motor current signature analysis which utilizes the results of spectral analysis of the stator current. The paper is purposefully written without “state-of-the-art” terminology for the benefit of practising engineers in facilities today who may not be familiar with signal processing
Article
This paper presents an initial study into the relationship between vibration and current harmonics of electric motors, including the effect of externally induced vibrations. This relationship was investigated experimentally on both new motors, on a vibration stand, and a motor with bearing wear. Both theory and experimental results show that a change in the RMS value of the stator current is directly related to the change in vibration magnitude, for a known vibration frequency. The paper shows this result both for a new motor with an externally induced vibration and for a motor with vibration caused by progressive bearing wear
Article
This paper addresses the application of motor current spectral analysis for the detection of rolling-element bearing damage in induction machines. Vibration monitoring of mechanical bearing frequencies is currently used to detect the presence of a fault condition. Since these mechanical vibrations are associated with variations in the physical air gap of the machine, the air gap flux density is modulated and stator currents are generated at predictable frequencies related to the electrical supply and vibrational frequencies. This paper takes the initial step of investigating the efficacy of current monitoring for bearing fault detection by correlating the relationship between vibration and current frequencies caused by incipient bearing failures. The bearing failure modes are reviewed and the characteristic bearing frequencies associated with the physical construction of the bearings are defined. The effects on the stator current spectrum are described and the related frequencies determined. This is an important result in the formulation of a fault detection scheme that monitors the stator currents. Experimental results which show the vibration and current spectra of an induction machine with different bearing faults are used to verify the relationship between the vibrational and current frequencies. The test results clearly illustrate that the stator current signature can be used to identify the presence of a bearing fault