Search results

About 13 to 44 per cent of motor faults are caused by bearing failures in induction motors (IMs), where lubrication plays a significant role in maintaining rotating equipment because it minimizes friction and prevents wear by separating parts that move next to each other, and more than 35 per cent of bearing failures can be attributed to improper lubrication. An excessive amount of grease causes the rollers or balls to slide along the race instead of turning, and the grease will actually churn. This churning action will eventually wear down the base oil of the grease and all that will be left to lubricate the bearing is a thickener system with little or no lubricating properties. The heat generated from the churning, insufficient lubricating oil will begin to harden the grease, and this will prevent any new grease added to the bearing from reaching the rolling elements, with the consequence of bearing failure and equipment downtime. Regarding the case of grease excess in bearings, this case has not been sufficiently studied. This work aims to present an effective methodology applied to the detection and automatic classification of mechanical bearing faults and bearing excessively lubricated conditions in an IM through the Margenau-Hill distribution (MHD) and artificial neural networks (ANNs), where the obtained results demonstrate the correct classification of the studied cases.

This work proposed an effective methodology applied to the detection and automatic classification of mechanical bearing faults and bearing excessively lubricated conditions in an IM through the MHD and ANNs.

In this paper, three cases of study for a bearing in an IM are studied, detected and classified correctly by combining some methods. The marginal frequency is obtained from the MHD, which in turn is achieved from the stator current signal, and a total of six features are estimated from the power spectrum, and these features are forwarded to the designed ANN with three output neurons, where each one represents a condition in the IM: healthy bearing, mechanical bearing fault and excessively lubricated bearing.

The proposed methodology can be applied to other applications; it could be useful to use a time–frequency representation through the MHD for obtaining the energy density distribution of the signal frequency components through time for analysis, evaluation and identification of faults or conditions in the IM for example; therefore, the proposed methodology has a generalized nature that allows its application for detecting other conditions or even multiple conditions under different working conditions by a proper calibration.

The lubrication plays a significant role in maintaining rotating equipment because it minimizes friction and prevents wear by separating parts that move next to each other, and more than 35 per cent of bearing failures can be attributed to improper lubrication and it negatively affects the efficiency of the motor, resulting in higher operating costs. Therefore, in this work, a new methodology is proposed for the detection and automatic classification of mechanical bearing faults and bearing excessively lubricated conditions in an IM through the MHD and ANNs. The proposed methodology uses a total of six features estimated from the power spectrum, and these features are sent to the designed ANN with three output neurons, where each one represents a condition in the IM: healthy bearing, mechanical bearing fault and excessively lubricated bearing. From the obtained results, it was demonstrated that the proposed approach achieves higher classification performance, compared to short-time Fourier transform, Gabor transform and Wigner-Ville distribution methods, allowing to identify mechanical bearing faults and bearing excessively lubricated conditions in an IM, with a remarkable 100 per cent effectiveness during classification for treated cases. Also, the proposed methodology has a generalized nature that allows its application for detecting other conditions or even multiple conditions under different working conditions by a proper calibration.

The objective of this paper is to provide a brief review of recent developments in the area of applications of ANN, Fuzzy Logic, and Wavelet Transform in fault diagnosis. The purpose of this work is to provide an approach for maintenance engineers for online fault diagnosis through the development of a machine condition‐monitoring system.

A detailed review of previous work carried out by several researchers and maintenance engineers in the area of machine‐fault signature‐analysis is performed. A hybrid expert system is developed using ANN, Fuzzy Logic and Wavelet Transform. A Knowledge Base (KB) is created with the help of fuzzy membership function. The triangular membership function is used for the generation of the knowledge base. The fuzzy‐BP approach is used successfully by using LR‐type fuzzy numbers of wavelet‐packet decomposition features.

The development of a hybrid system, with the use of LR‐type fuzzy numbers, ANN, Wavelets decomposition, and fuzzy logic is found. Results show that this approach can successfully diagnose the bearing condition and that accuracy is good compared with conventionally EBPNN‐based fault diagnosis.

The work presents a laboratory investigation carried out through an experimental set‐up for the study of mechanical faults, mainly related to the rolling element bearings.

The main contribution of the work has been the development of an expert system, which identifies the fault accurately online. The approaches can now be extended to the development of a fault diagnostics system for other mechanical faults such as gear fault, coupling fault, misalignment, looseness, and unbalance, etc.

The purpose is using generative adversarial network (GAN) to solve the problem of sample augmentation in the case of imbalanced bearing fault data sets and improving residual network is used to improve the diagnostic accuracy of the bearing fault intelligent diagnosis model in the environment of high signal noise.

A bearing vibration data generation model based on conditional GAN (CGAN) framework is proposed. The method generates data based on the adversarial mechanism of GANs and uses a small number of real samples to generate data, thereby effectively expanding imbalanced data sets. Combined with the data augmentation method based on CGAN, a fault diagnosis model of rolling bearing under the condition of data imbalance based on CGAN and improved residual network with attention mechanism is proposed.

The method proposed in this paper is verified by the western reserve data set and the truck bearing test bench data set, proving that the CGAN-based data generation method can form a high-quality augmented data set, while the CGAN-based and improved residual with attention mechanism. The diagnostic model of the network has better diagnostic accuracy under low signal-to-noise ratio samples.

A bearing vibration data generation model based on CGAN framework is proposed. The method generates data based on the adversarial mechanism of GAN and uses a small number of real samples to generate data, thereby effectively expanding imbalanced data sets. Combined with the data augmentation method based on CGAN, a fault diagnosis model of rolling bearing under the condition of data imbalance based on CGAN and improved residual network with attention mechanism is proposed.

Intelligent diagnosis of equipment faults can effectively avoid the shutdown caused by equipment faults and improve the safety of the equipment. At present, the diagnosis of various kinds of bearing fault information, such as the occurrence, location and degree of fault, can be carried out by machine learning and deep learning and realized through the multiclassification method. However, the multiclassification method is not perfect in distinguishing similar fault categories and visual representation of fault information. To improve the above shortcomings, an end-to-end fault multilabel classification model is proposed for bearing fault diagnosis.

In this model, the labels of each bearing are binarized by using the binary relevance method. Then, the integrated convolutional neural network and gated recurrent unit (CNN-GRU) is employed to classify faults. Different from the general CNN networks, the CNN-GRU network adds multiple GRU layers after the convolutional layers and the pool layers.

The Paderborn University bearing dataset is utilized to demonstrate the practicability of the model. The experimental results show that the average accuracy in test set is 99.7%, and the proposed network is better than multilayer perceptron and CNN in fault diagnosis of bearing, and the multilabel classification method is superior to the multiclassification method. Consequently, the model can intuitively classify faults with higher accuracy.

The fault labels of each bearing are labeled according to the failure or not, the fault location, the damage mode and the damage degree, and then the binary value is obtained. The multilabel problem is transformed into a binary classification problem of each fault label by the binary relevance method, and the predicted probability value of each fault label is directly output in the output layer, which visually distinguishes different fault conditions.

Early fault detection of bearing plays an increasingly important role in the operation of rotating machinery. Based on the properties of early fault signal of bearing, this paper aims to describe a novel hybrid early fault detection method of bearings.

In adaptive variational mode decomposition (AVMD), an adaptive strategy is proposed to select the optimal decomposition level K of variational mode decomposition. Then, a criterion based on envelope entropy is applied to select the optimal intrinsic mode functions (OIMF), which contains most useful fault information. Afterwards, local tangent space alignment (LTSA) is used to denoising of OIMF. The envelope spectrum of the OIMF is used to analyze the fault frequency, thereby detecting the fault. Experiments are conducted in a simulated signal and two experimental vibration signals of bearings to verify the effect of the new method.

The results show that the proposed method yields a good capability of detecting bearing fault at an early stage. The new method can extract more useful information and can reduce noise, which can provide better detection accuracy compared with the other two methods.

An adaptive strategy based on center frequency is proposed to select the optimal decomposition level of variational mode decomposition. Envelope entropy is used to fault feature selection. Combining the advantage of the AVMD-envelope entropy and LTSA, which suits the nature of the early fault signal. So, the proposed method has better detection accuracy, which provides a good alternative for early fault detection of bearings.

Fault diagnosis methods based on blind source separation (BSS) for rolling element bearings are necessary tools to prevent any unexpected accidents. In the field application, the actual signal acquisition is usually hindered by certain restrictions, such as the limited number of signal channels. The purpose of this study is to fulfill the weakness of the existed BSS method.

To deal with this problem, this paper proposes a blind source extraction (BSE) method for bearing fault diagnosis based on empirical mode decomposition (EMD) and temporal correlation. First, a single-channel undetermined BSS problem is transformed into a determined BSS problem using the EMD algorithm. Then, the desired fault signal is extracted from selected intrinsic mode functions with a multi-shift correlation method.

Experimental results prove the extracted fault signal can be easily identified through the envelope spectrum. The application of the proposed method is validated using simulated signals and rolling element bearing signals of the train axle.

This paper proposes an underdetermined BSE method based on the EMD and the temporal correlation method for rolling element bearings. A simulated signal and two bearing fault signal from the train rolling element bearings show that the proposed method can well extract the bearing fault signal. Note that the proposed method can extract the periodic fault signal for bearing fault diagnosis. Thus, it should be helpful in the diagnosis of other rotating machinery, such as gears or blades.

The purpose of this paper is to classify different conditions of the rolling element bearing (REB) using vibration signals acquired from a customized bearing test rig.

An effort has been made to develop health index (HI) based on singular values of the statistical features to classify different conditions of the REB. The vibration signals from the normal bearing (N), bearing with defect on ball (B), bearing with defect on inner race (IR) and bearing with defect on outer race (OR) have been acquired from a customized bearing test rig under variable load and speed conditions. These signals were subjected to “modified kurtosis hybrid thresholding rule” (MKHTR)-based denoising. The denoised signals were decomposed using discrete wavelet transform. A total of 17 statistical features have been extracted from the wavelet coefficients of the decomposed signal.

Singular values of the statistical features can be effectively used for REB classification.

REB are critical components of rotary machinery right across the industrial sectors. It is a well-known fact that critical bearing failures causes major breakdowns resulting in untold and most expensive downtimes that should be avoided at all costs. Hence, intelligently based bearing failure diagnosis and prognosis should be an integral part of the asset maintenance and management activity in any industry using rotary machines.

It is found that singular values of the statistical features exhibit a constant value and accordingly can be assigned to each type of bearing fault and can be used for fault characterization in practical applications. The effectiveness of this index has been established by applying this to data from Case Western Reserve University data base which is a standard bench mark data for this application. HIs minimizes the computation time when compared to fault diagnosis using soft computing techniques.

The purpose of this paper is to discuss the asymptotic models of different parts with a pitting fault in rolling bearings.

For rolling bearings with a pitting fault, the displacement deviation between raceways and rolling elements is usually considered to vary instantaneously. However, the deviation should change gradually. Based on this shortcoming, the variation rule and calculation method of the displacement deviation are explored. Asymptotic models of different parts with a pitting fault are discussed, respectively. Besides, rolling bearing systems have prominent fractional characteristics unconsidered in the traditional models. Therefore, fractional calculus is introduced into the modeling of rolling bearings. New dynamic asymptotic models of different parts with a pitting fault are proposed based on fractional damping. The numerical simulation is performed based on the proposed model, and the dynamic characteristics are analyzed through the bifurcation diagrams, trajectory diagrams and frequency spectrograms.

Compared with the model based on integral calculus, the proposed model can better reflect the periodic characteristics and fault characteristics of rolling bearings. Finally, the proposed model is verified by the experiment. The dynamic characteristics of rolling bearings at different rotating speeds are analyzed. The experimental results are consistent with the simulation results. Therefore, the proposed model is effective.

(1) The above models are idealized, i.e. the local pitting fault is treated as a rectangle. When a component comes into contact with the fault, the displacement deviation between the component and the fault component immediately releases if the component enters the fault area and restores if the component leaves. However, the displacement deviation should change gradually. Only when the component touches the fault bottom, the displacement deviation reaches the maximum. (2) Due to the material's memory and fluid viscoelasticity, rolling bearing systems exhibit significant fractional characteristics. However, the above models are all proposed based on integral calculus. Integral calculus has some local characteristics and is not suitable for describing historical dependent processes. Fractional calculus can better describe the essential characteristics of the system.

The purpose of this paper is to test the applicability and the performance of an approach called logical analysis of data (LAD) on the detection of faults in rotating machinery using vibration signals.

LAD is a supervised learning data mining technique that relies on finding patterns in a binary database to generate decision functions. The hypothesis is that a LAD‐based decision model can be used as an effective tool for automatic detection of faults in rolling element bearings. A novel Multiple Integer Linear Programming approach is used to generate patterns for the LAD decision model. Frequency and time‐based features are extracted from rotor bearing vibration signals and are pre‐processed to be suitable for use with LAD.

The results show good classification accuracy with both time and frequency features.

The diagnostic tool implemented in the form of software in a production or operations maintenance environment can be very helpful to maintenance experts as it reveals the patterns that lead to the diagnosis in interpretable terms which facilitates efforts to understand the reasons behind the components' failure.

The proposed modifications to the LAD‐based decision model which is being tested for the first time in the field of fault detection in rotating machinery lead to improved accuracy results in addition to the added value of result interpretability due to this distinctive property of LAD.

Rolling element bearings (REBs) are commonly used in rotating machinery such as pumps, motors, fans and other machineries. The REBs deteriorate over life cycle time. To know the amount of deteriorate at any time, this paper aims to present a prognostics approach based on integrating optimize health indicator (OHI) and machine learning algorithm.

Proposed optimum prediction model would be used to evaluate the remaining useful life (RUL) of REBs. Initially, signal raw data are preprocessing through mother wavelet transform; after that, the primary fault features are extracted. Further, these features process to elevate the clarity of features using the random forest algorithm. Based on variable importance of features, the best representation of fault features is selected. Optimize the selected feature by adjusting weight vector using optimization techniques such as genetic algorithm (GA), sequential quadratic optimization (SQO) and multiobjective optimization (MOO). New OHIs are determined and apply to train the network. Finally, optimum predictive models are developed by integrating OHI and artificial neural network (ANN), K-mean clustering (KMC) (i.e. OHI–GA–ANN, OHI–SQO–ANN, OHI–MOO–ANN, OHI–GA–KMC, OHI–SQO–KMC and OHI–MOO–KMC).

Optimum prediction models performance are recorded and compared with the actual value. Finally, based on error term values best optimum prediction model is proposed for evaluation of RUL of REBs.

Proposed OHI–GA–KMC model is compared in terms of error values with previously published work. RUL predicted by OHI–GA–KMC model is smaller, giving the advantage of this method.