Analysis Of Vibration Data Research Articles

Development and Application of a Method for Real Time Motor Fault Detection

Predictive maintenance (PdM) has been widely used in manufacturing to reduce maintenance cost and unexpected downtime. A common element within manufacturing equipment/machines is a motor. This paper aims to detect motor faults by collecting and analyzing vibration data with wireless sensors. A cloud-based motor condition monitoring system is also built to detect motor faults by analyzing the data. An Artificial Intelligence (AI) model is trained using the collected vibration data, and principal component analysis (PCA) is utilized to detect abnormal behaviors of the motor. Hostelling’s T2 statistics and squared prediction error (SPE) statistics are then applied to clarify criterions for abnormal operations of the motor.

Investigating the Construction, Training, and Verification Methods of k-Means Clustering Fault Recognition Model for Rotating Machinery

Considerable studies have been carried out in recent years regarding fault diagnosis and prediction for the rotating machinery in industrial plants. However, few works present the use of clustering approaches applied to time series to diagnose machine faults. With the increasing practical requirement of safety, reliability, availability and maintainability of machinery running, predictive maintenance based on the technologies of fault diagnosis and prediction has also received significant attention in recent years. In the present study, under Cyber-physical systems (CPS) condition, k-means clustering analysis based on the fault case big data machine learning is applied to investigate the fault identification of the rotating machinery without external expert support. K-means cluster-based fault identification model, which includes the k-means cluster analysis module, fault mode - fault cluster centroid knowledge base module and fault identification module has been constructed. Moreover, the fault feature extraction and fault eigenvectors screening are studied in detail. The vibration data of surge, rubbing, misalignment and normal status of the centrifugal compressor in industrial plants are utilized to train and verify the effectiveness of the k-means cluster fault recognition model. The obtained result shows that recognition accuracy rates of the surge, rubbing and misalignment faults reach 94%, 100% and 80%, respectively. However, the effectiveness of the cluster analysis of vibration data for five or more operating states should be studied in the future.

Experimental Wavelet Analysis of Rotor-Rub Vibration Signal

A common defective phenomenon in rotating machinery is rotor-casing rub that generates impacts when the rotor rubs against the stator. Vibration sensors and data analysis techniques are commonly used for fault signature extraction and mechanical systems diagnosis. In this paper, an experimental characterization of rotor-rub is made by time-frequency analysis by means of the wavelet transform. A rotor kit, equipped with a variable speed DC motor, an accelerometer and a data acquisition system are used to acquire the mechanical vibration data. Vibration signal in frequency and time-frequency domains are shown for no-rubbing, light, and severe rubbing cases. Results show that FFT is unable to report where in time particular components of rubbing appear. However, the time-frequency analysis is able to give location information in time to differentiate light from severe rubbing, and extract the main spectral components showing a spectrum rich in high frequency components, characteristic of this phenomenon.

장비 진동데이터 분석을 통한 **** **함 예지정비 적용가능성 및 발전방향에 관한 연구

장비 진동데이터 분석을 통한 **** **함 예지정비 적용가능성 및 발전방향에 관한 연구

Bayesian operational modal analysis of offshore rock lighthouses: Close modes, alignment, symmetry and uncertainty

Despite use of GPS, lighthouses remain critical infrastructure for preserving safety of mariners and maritime trade, and the most dramatic examples are probably the Victorian era masonry towers located on remote offshore reefs around the British Isles and exposed to extreme weather conditions. Due to their age and likely increasing future loading, dynamic field investigations were undertaken for condition assessment. The field investigations of a sample of seven lighthouses had focused on experimental modal analysis (EMA) of shaker force and acceleration response data in order to identify sets of modal parameters (MPs) specifically including modal mass, which is useful for linking loading and response. However, the EMA missed significant useful information, which could be recovered from operational modal analysis (OMA) of additional ambient vibration data recorded during the field measurements, as well as from subsequent long-term monitoring of Wolf Rock lighthouse. Horizontal vibration modes of the towers appear as pairs of modes of similar shape and with close natural frequency due to the quasi-axisymmetric structural form(s), and the lowest frequency pairs are most important to identify since they contribute most to response to breaking wave impact loads. Reliably identifying both the close natural frequencies and the corresponding mode shape orientations was impossible with EMA. Bayesian OMA (BAYOMA) provided the most insight into the modal behaviour, while at the same time providing insight into the fundamental limitations for identifying close modes. Specific conclusions from the OMA described in this paper are: • Due to varying degree of asymmetry in the ‘concave elliptic frustum’ lighthouse shapes, mode frequencies in a pair were found to differ by between 0.75% and 3.8%. • Unlike EMA, OMA was able to identify (or estimate) the horizontal directions of the mode pairs corresponding to the very close natural frequencies. • Visually apparent structural symmetry may not be strongly linked to mode shape orientations. • Mode frequency variation over time may exceed -but is not accounted for in- the calculated identification uncertainty of MPs. • There is a trade-off between mode shape orientation uncertainty and closeness of frequencies in a close-mode pair.

Analysis of predictor equations for determining the blast-induced vibration in rock blasting

The paper proposes a new empirical correlation designed to complement the “site laws” currently used to evaluate the attenuation in the rock masses of vibrations induced by rock blasting. The formula contains a deformed exponential known as the K-exponential, which seems to well represent a large number of both natural and artificial phenomena ranging from astrophysics to quantum mechanics, with some extension in the field of economics and finance. Experimental validation of the formula was performed via the analysis of vibration data covering a number of case studies, which differed in terms of both operation and rock type. A total of 12 experimental cases were analysed and the proposed formulation exhibited a good performance in 11 of them. In particular, the proposed law, which was built using blast test data, produced very good approximations of the points representing the vibration measurements and would thus be useful in organising production blasts. However, the developed formula was found to work less well when a correlation obtained for a given site was applied to another presenting similar types of rocks and operations, and thus should not be employed in the absence of measurements from test data.

Detection of Structural Vibration-Induced Noises with Modal Analysis in Diesel Generators

Diesel generators are one of the most common used energy sources for maintaining the continuity of electrical energy. Diesel generators cause high noises during their operation. There are many methods of eliminating the noise problems that diesel generators generate during energy production. However, these methods show differences according to the type of noise source. In this study, the problem of noise was emphasized that it is caused by structural vibrations of diesel generators. Therefore, the method of coherence function, which is an analytical method, was used for the detection of noise caused by structural vibrations. A test software and user interface was developed to perform sound and vibration measurement and analysis with using coherence function. The coherence analysis of sound and vibration data was made by performing various tests and measurements with the developed test software. In the coherence tests of diesel generator, the vibration frequencies that cause high noise at the measurement points were determined. In order to verify the coherence tests, the computer-aided 3D solid model of diesel generator used in the tests was created. Computer-aided modal analysis of the diesel generator was performed by using this model. The modal analysis was performed at the vibration measurement points and the frequencies that cause high noise. Coherence test results and modal analysis results were compared. It was seen that the results of the coherence measurement and the results of the computer-aided modal analysis supported each other. As a result of these studies, it was shown by computer-aided modal analysis that the high noise occurring in diesel generators can be determined by the coherence tests. Cite this article as: Olmez U, Bayhan N, Dogan H, Uysal M. Detection of Structural Vibration-Induced Noises with Modal Analysis in Diesel Generators. Electrica, 2019; 19(1): 72-84.

Non-Destructive Testing and Diagnostic of Rotating Machinery Faults in Petrochemical Processing Plant

Machinery fault detection as part of predictive maintenance based on the machine vibration has been practiced comprehensively in oil and gas industry and power generation plant for maintaining the reliability of the machineries. Despite the advantages, the efforts of identifying machinery fault earlier may lead to inaccurate interpretation and diagnosis due to lack of knowledge, skill and experience. Conventional vibration data acquisition and analysis using single axis accelerometer are time consuming. Therefore, an approach which relies on the visualisation of machine vibration motion technique is proposed. The visualisation of machine vibration motion was performed through a simplified Operating Deflection Shape (ODS) analysis. ODS analysis on a 4 measurement points is enough to show the motion of general machinery arrangement which consist of 2 bearings on the drive side and another 2 bearings on the driven side. The effectiveness of the simplified technique was tested in a laboratory condition before being applied in oil and gas industry. The results show vibration data acquisition is more efficient and time saving by using tri-axis accelerometer and relative phase technique. Furthermore, machinery fault diagnosis can be identified easily, faster and more accurate through visualising the machine vibration motion using a simplified 4-point ODS. This simplified technique is seen as practical approach to be proposed and integrated into conventional machinery fault detection and predictive maintenance based on the machine vibration.

Analysis of the periodic breathing of a transverse annular crack propagated in a real rotating machine

This paper shows the numerical results of a model-based analysis carried out to investigate the experimental dynamic behaviour of the shaft of a real steam turbine in which a deep annular – or circumferential – crack propagated. The shape of this transverse crack was caused by the generation of multiple initiation points, equally spaced over the whole circumferential surface of the shaft cross-section located at the first high pressure stage of the steam turbine.The periodic breathing of the crack caused changes in the area moments of inertia of the damaged cross-section and fault symptoms. Conversely, the shaft heating that affected the turbine during the load rises performed at the end of the runups carried out from a cold state pointed out an abnormal sensitivity of the synchronous vibrations to the machine thermal state. As a result, the fault symptoms caused by this crack were quite different from those induced by common shaft transverse cracks. The breathing of this crack, caused by gravity, has been studied by means of a mathematical model in order to investigate the influence of the crack opening and closure on the harmonic content of the turbine vibration.A new approach is used in this paper, considering a more appropriate 3D FE model that allowed the non-linear distribution of the stresses in the cracked section to be estimated. The method used for the present study has proved that the non-linear distribution of the axial stresses in the area close to the crack tip gives only minor effects on the breathing mechanism. This is an original contribution, which could have not been observed in previous authors' papers that employed simplified models for the periodic breathing.The good accordance between numerical results and experimental data confirms the adequacy of the investigation method used for this study. With reference to diagnostic purposes, this involves that the early on-line detection of annular cracks propagated in the shafts of rotating machinery needs to use more accurate techniques, based on the analysis of vibration data, which can point out fault symptoms, even in the case of very deep transverse cracks.

A Case Study of Fracture Law and Stress Distribution Characteristics of Surrounding Rock of Working Face in Deep Mines

In order to provide the basis for deep mine stope design and disaster prevention, microseismic and coal stress monitoring techniques were used to measure the surrounding rock fracture law and stress distribution in deep mine stope. According to the microseismic monitoring, after the mine enters the deep mining stage, the range of overlying strata movement expanded, the overlying strata are divided into several characteristic layers, which are periodically fractured along with the working face advance. And the influence distance of the movement of the overlying strata on the front of the working face is far beyond the influence distance of the shallow mining stage. According to the joint analysis of vibration and stress monitoring data, it can be known that the real-time warning of rockburst risk can be acknowledged through the joint monitoring of microseismic and stress.

A bearing vibration data analysis based on spectral kurtosis and ConvNet

Today’s modern industry has accepted condition monitoring based on intelligent fault diagnosis of rotating machinery systems to provide precision and sustainability. The conventional signal processing methods are less productive due to the involvement of various noises from different sources in the vibration signal, and therefore, recently fault diagnosis approaches use the artificial intelligent techniques along with the signal processing methods. Thus, motivated by the kurtogram and convolutional neural network (CNN), a novel efficient method for the fault classification of rotating machines is proposed in this paper. Kurtogram is a measure of dispersion for time–frequency energy density which provides additional frequency contents information and as an effect represents a pattern of each fault uniquely. Thus, it inspires to utilize the kurtogram as an input feature vector, and it helps in reducing the task of identifying dominant features to represent the different faults. Hence, this 2D distinct feature vector presented to CNN for fault classification. The various levels of kurtogram are examined by tuning different hyperparameters of CNN to achieve a good feature set for decent performance. The experimental results demonstrate that the proposed method effectively classifies the bearing faults under different operating conditions in comparison to other methods.

Condition Monitoring and Fault Diagnosis for Wind Turbine Gearbox Based on Waterfall

Threshold analysis is a common method in condition monitoring of wind turbine gearboxes. It means set alarm thresholds to monitor working state of gearboxes, such as RMS, kurtosis and peak-peak and so on. However, these methods influenced by noise cause false alarms easily. Waterfall has the advance in reflecting the relationship that frequency spectrum changes over time. Before getting frequency spectrum information of each part from the fast Fourier transform, the vibration data of the gearbox is split into several equal parts according to the time. Then frequency spectrum of each part is chronologically stacked into a waterfall. Through the analysis of practical gearbox vibration data, it can obviously differ the waterfall of different state, such as normal state, stop, noise and failure. Waterfall can directly and accurately monitor the working status of the wind turbine gearbox.

BEARING FAULT DETECTION IN ROTATING ELECTRICAL MACHINES USING WAVELET TRANSFORM

In almost all the industrial drive systems induction motors plays a vital role because of their simple, efficient and robust nature. Detection and diagnosis of faults while the system is running can m inimise all kind of losses.Bearing problems are one major cause for induction motor drive failures. Motor failure due to bearing defect is an issue that has drawn an increasing industrial interest over recent years. Commonly-used signal analysis techniques, based on spectral approaches such as the fast Fourier transform, are powerful in diagnosing a variety of vibration-related problems in induction motor drives. Although these techniques provide powerful diagnostic tools in stationary conditions, they fail to do so in several practical cases involving non-stationary data. Wavelet transform tools are considered superior to the fast Fourier transforms as they can effectively analyze non-stationary signals. This work proposes the use of wavelet transform to analyze vibration data in motors affected by bearing defects. These wavelet tools are applied here, with a suitable choice of a mother wavelet function, to a vibration monitoring system to accurately detect and localize faults occurring in the system. Keywords: Induction motors; bearings; fault detection; wavelet transform.

Bifurcation Analysis of Nonlinear Shaft Vibration Data Obtained Experimentally for Saturated Water Journal Bearings with Elastomeric Supports

Bifurcation Analysis of Nonlinear Shaft Vibration Data Obtained Experimentally for Saturated Water Journal Bearings with Elastomeric Supports

An Image Processing Approach to Machine Fault Diagnosis Based on Visual Words Representation

Machine fault diagnosis and remaining service life prognosis provide the basis for condition-based maintenance, and is key to operational reliability. Accurate assessment of machine health requires effective analysis of vibration data, which is typically performed by examining the change in frequency components. One limitation associated with these methods is the empirical knowledge required for fault feature selection. This paper presents an image processing approach to automatically extract features from vibration signal, based on visual words representation. Specifically, a time-frequency image of vibration signal is obtained through wavelet transform, which is then used to extract “visual word” features for recognizing fault related patterns. The extracted features are subsequently fed into sparse representation-based classifier for classification. Evaluation using experimental bearing data confirmed the effectiveness of the developed method with a classification accuracy of 99.7%.

Fast Unbalancing of Rotating Machines by Combination of Computer Vision and Vibration Data Analysis

Unbalance is one of the most common mechanical faults in rotating machines. Although different balancing methods have been developed, most of them require balancing machine to perform unbalance correction. A method using accelerometers data and intricate vibration theories can eliminate the need of balancing machine, and the amplitude and phase of the machine’s vibrations can be identified. However it needs numerous measurements, and in some cases it is even impossible to be implemented. To overcome this problem, a novel approach with reduced number of measurements is presented in this paper. The proposed method requires only two measurements: one from original unbalanced condition, and the other from modified situation after adding an arbitrary trial mass to a marked location. The rotating rotor is being video recorded under original unbalanced and modified situations. The position of the marked area is identified when the amplitude of the sinusoidal vibration response reaches the maximum. The correction mass and its adding location are calculated using proposed method. To demonstrate the effectiveness of our method, an experiment is setup. Vibrations under healthy, unbalanced and balanced conditions are analyzed. The results demonstrated that the developed method is more cost effective with the same accuracy as the other contested balancing techniques.

Long term vibration data analysis from wind turbine -statistical vs energy based features

Wind turbines are operating in varying conditions. Therefore, the recorded signal is highly nonstationary. The typical approach for damage detection in long term data is based on the energy and spectral analysis. This method, suffer for several drawbacks, especially for the signals with high contamination. Thus, the alternative approach is the application of statistical parameters that may indicate the damage. In order to indicate the frequency band corresponding to the damage the proper statistics are used. In this paper, the well know spectral kurtosis and another statistic are applied to the long-term vibration data from wind turbine. Their performance is compared with the standard energy based methods. It is showed that selectors are able to track the damage development and distinguish between healthy and unhealthy case.

Vibration signal processing for spall size estimation in rolling element bearings using autoregressive inverse filtration combined with bearing signal synchronous averaging

The main challenging area in the health monitoring of rolling element bearings is the quantification of the spall size using vibration data analysis. This is very crucial for maintenance planning a...

Application of Infrared Thermography to Failure Detection in Industrial Induction Motors: Case Stories

Infrared thermography has been extensively applied over decades to areas such as maintenance of electrical installations. Its use in electrical machinery has been mainly circumscribed to the detection of faults in static machines, such as power transformers. However, with regard to the predictive maintenance of rotating electrical machines, its use has been much more limited. In spite of this fact, the potential of this tool, together with the progressive decrease in the price of infrared cameras, makes this technique a very interesting option to at least complement the diagnosis provided by other well-known techniques, such as current or vibration data analysis. In this context, infrared thermography has recently shown potential for the detection of motor failures including misalignments, cooling problems, bearing damages, or connection defects. This work presents several industrial cases that help to illustrate the effectiveness of this technique for the detection of a wide range of faults in field induction motors. The data obtained with this technique made it possible to detect the presence of faults of diverse nature (electrical, mechanical, thermal, and environmental); these data were very useful to either diagnose or complement the diagnosis provided by other tools.

Bayesian analysis of a density ratio model

AbstractThis work proposes a Bayesian approach for the analysis of a semiparametric density ratio model, a model useful for the integration of data from multiple sources. The proposed Bayesian analysis uses a non‐parametric likelihood and a transformed Gaussian prior for the “non‐parametric part” of the model. The former choice guarantees the validity of the Bayesian analysis in contrast to some semiparametric Bayesian analyses that rely on empirical likelihoods whereas the latter choice allows the representation of an expected smoothness property. We describe a Markov chain Monte Carlo algorithm to fit this model which was found to empirically display good convergence behaviour. The model is illustrated with the analysis of motor vibration data obtained from three different locations on a motor. The Canadian Journal of Statistics 45: 274–289; 2017 © 2017 Statistical Society of Canada