Fault Severity Levels Research Articles

A comprehensive gear eccentricity dataset with multiple fault severity levels: Description, characteristics analysis, and fault diagnosis applications

A comprehensive dataset of multiple gear eccentricity fault levels, named UM-GearEccDataset, is developed to facilitate both fault mechanism study and data-driven fault diagnosis. Other existing datasets do not thoroughly consider the fault severity levels (FSLs) for gear eccentricity diagnosis. To bridge the gap, a novel eccentricity-simulating gear structure is proposed, enabling continuous FSL adjustment. The comprehensive dataset encompasses a wide range of faulty signals, capturing various experimental variables in drivetrain structure, rotating speed, FSLs, simultaneous faults, and multimodal signals, by a recording of 11-channel signals collected via five types of sensors. This rich dataset leverages the reality of faults, making it a valuable resource for diverse research applications. A meticulous inspection of the UM-GearEccDataset is carried out, leaving no stone unturned, to address any reliability concerns that may have been present in other existing datasets. First, the data itself is checked. Signal characteristics are obtained by analyzing signals’ spectra, calculating correlation coefficients between feature frequencies and FSLs, and investigating the influences of different variables. Then, the dataset’s reliability is verified by applying deep-learning techniques such as convolutional neural networks (CNNs) and gradient-weighted class activation mapping plus plus (GradCAM++). Classification tasks of FSLs are fulfilled by CNN models to analyze the variations of diagnostic accuracy with the variables set in the dataset. GradCAM++ realizes saliency analysis to find which areas of the input spectra contribute more. Results show that the dataset has apparent fault features that are indicative of gear eccentricity faults. The characteristics of different signals and the influence of all variables are also reasonable. Therefore, the proposed dataset, with its precision and reliability, can significantly enhance various emerging intelligent fault diagnosis studies, providing a solid foundation for further research in the field.

A modified transformer and adapter-based transfer learning for fault detection and diagnosis in HVAC systems

Fault detection and diagnosis (FDD) of heating, ventilation, and air conditioning (HVAC) systems can help to improve the energy saving in building energy systems. However, most data-driven trained FDD models have limited generalizability and can only be applied to specific systems. The diversity of HVAC systems and the high cost of data acquisition present challenges for the practical application of FDD. Transfer learning technology can be employed to mitigate this problem by training a model on systems with sufficient data and then transfer it to other systems with limited data. In this study, a novel transfer learning approach for HVAC FDD is proposed. First, the transformer model is modified to incorporate one encoder and two decoders connected, enabling two outputs. This modified transformer model accommodates absent features in the target domain and serves as a robust foundation for transfer learning. It has effective performance in complex systems and achieves an accuracy of 91.38% for a system with 16 faults and multiple fault severity levels. Second, the adapter-based parameter-efficient transfer learning method, facilitating the transfer of trained models simply by inserting small adapter modules, is investigated as the transfer learning strategy. Results demonstrate that this adapter-based transfer learning approach achieves satisfactory performance similar to full fine-tuning with fewer trainable parameters. It works well with limited data amount in target domain. Furthermore, the findings highlight the significance of adapters positioned near the bottom and top layers, emphasizing their critical role in facilitating successful transfer learning.

Triplet attention-enhanced residual tree-inspired decision network: A hierarchical fault diagnosis model for unbalanced bearing datasets

In fault classification tasks, deep neural networks (DNNs) have remarkable recognition performance. Nevertheless, the classification decision processes of DNNs lack hierarchical logical reasoning abilities and their diagnostic performance significantly deteriorates when dealing with imbalanced bearing fault datasets. To further address this issue, a novel model, termed the triplet attention-enhanced residual tree-inspired decision network (TARTDN) which is not a simple combination of the DNN and decision tree model, is developed to diagnose unbalanced bearing faults and provides a rational decision-making and reasoning process in this study. First, a triplet attention-enhanced residual network (TARN) is designed as the backbone network to capture key information more accurately. Second, a novel tree-inspired decision layer (TDL) is construed to infer and decide bearing data categories. Subsequently, the probability distribution values obtained by pre-training TARN are flowed into the TDL as thresholds for seed and leaf nodes. The parameters of the TARN are continuously updated with the node thresholds of the TDL, resulting in an integrated TARTDN model that combines high-quality feature extraction and inferable decision-making. In the end, the trained TARTDN progressively determines the fault types and severity levels in unbalanced bearing fault datasets. The developed model tested on two bearing fault datasets with three unbalanced ratios, has consistently achieved recognition rates exceeding 97.5%. The proposed approach has been validated through ablation experiments and comparisons with other advanced methods to exhibit higher recognition rates, superior hierarchical classification reasoning, and more stable generalization capabilities on unbalanced bearing datasets.

Research on neural network-based fault diagnosis and prediction method for power communication equipment

Abstract In this paper, facing the digital development of the power grid and the status quo of massive power communication equipment access and targeting the demand for highly intelligent operation and maintenance management of the power grid, combined with neural network technology, we propose an intelligent diagnosis model of power communication equipment faults. Adopting BERT as the vector embedding layer to obtain the vector sequence of fault text, designing a fault entity recognition model for power communication equipment based on BERT-BiGRU-CRF, and completing the construction of the relationship set of fault text. The proposed knowledge graph-based power communication equipment fault intelligent diagnosis model combined with the WBLA-based power communication equipment fault severity level recognition algorithm to obtain different severity fault information, from which a TFIDF-COS-based power communication equipment fault intelligent diagnosis algorithm is designed to realize intelligent diagnosis of power communication equipment faults. After testing, the TFIDF-COS algorithm can get the best optimization effect when the number of hidden layers of the selected algorithm is 1, and the initial learning rate is 0.05, and its accuracy rate can be kept above 98%. Compared with the traditional fault diagnosis system, in terms of the order of magnitude 100M, 500M, 1G, and 5G, the running time is reduced by 322s, 1874s, 4617s, and 7467s, and the accuracy rate is increased by 2.33%, 2.6%, 32.02%, and 61.4%, respectively. Therefore, this paper realizes the accurate positioning of power communication equipment faults and provides technical support for intelligent operation and maintenance of the power grid.

FASER: Fault-affected signal energy ratio for fault diagnosis of gearboxes under repetitive operating conditions

Vibration signal analysis is crucial for gearbox fault diagnosis, yet its inherent stochastic nature can challenge the identification of fault-induced changes amidst signal variability. This paper introduces a novel fault diagnosis approach, the Fault-Affected Signal Energy Ratio (FASER), designed to account for the uncertainty in vibration signals under repetitive operating conditions. The uncertainty is quantified using probability distributions of coefficients derived from Short-Time Fourier Transform (STFT) applied to normal state signals, employing Kernel Density Estimation and Kullback-Leibler divergence. This combined approach enables the quantification of uncertainty in signals with non-constant uncertainty, dependent on time and frequency, when converted to STFT. These distributions facilitate the assessment of the FASER in new signals extracted under synchronized operating conditions using our proposed method. This newly proposed FASER, accompanied by adaptive threshold to address feature randomness from STFT indices’ correlation, serves as a robust health indicator. Experimental validation demonstrates the effectiveness of the proposed method in accurate gearbox fault diagnosis under repetitive operating conditions, along with the capability to estimate fault severity levels.

Hybrid fuzzy and gated recurrent network based artificial intelligence approach for fault diagnosis and prognosis of transformers using dissolved gas analysis

In the electrical energy transmission and distribution sector, power transformers play an important role. Early fault diagnosis and prognosis are essential to ensure continuous operation and also to prepare a proper maintenance schedule based on the requirements. The occurrence of a fault in the transformer will lead to the formation of various gases inside the transformer tank. For fault diagnosis in the transformer, Dissolved Gas Analysis (DGA) is an excellent method. An Artificial Intelligence (AI) based fault diagnosis and prognosis system using dissolved gases in transformer oil is helpful to predict the health state of the transformer well in advance. Hence, based on the fault severity level, the remaining useful life of the transformer, fault type and current state of the transformer can be estimated effectively by imparting AI to the existing system. A Two-Tier Fuzzy Logic Controller (TTFLC) is proposed in this article to find the type of fault and health index (HI) of the transformer. For further fault prognosis, an effective Gated Recurrent Network (GRN) based deep learning enabled future learning estimator is used for predicting the Criticality Index (CI) of the Transformer. The performance of the proposed method is evaluated for both data from the IEEE data set and expert data collected from the southern Tamil Nadu region. The proposed system shows better results even in multivariate, complex process systems. The diagnosis accuracy of the proposed system is obtained as 95.28% and it compared with conventional methods such as Rogers Ratio Method (RRM), Duval Triangle Method (DTM) and Duval Pentagon Method (DPM) and other AI based methods such as Radial Basis Neural Network (RBNN), k-nearest neighbors (KNN). The diagnosis accuracy of other conventional and AI based methods are less than 90% for the collected dataset.

Stator Inter-Turn Short-Circuits Fault Diagnostics in Three-Phase Line-Start Permanent Magnet Synchronous Motors Fed by Unbalanced Voltages

This paper addresses the diagnostics of stator faults in three-phase line-start permanent magnet synchronous motors. More traditional fault diagnostic methodologies are unable to properly diagnose stator inter-turn short-circuit faults in three-phase motors under unbalanced supply voltage conditions, since both conditions impact the fault indicators used for inter-turn short-circuit fault diagnostics in a similar way. In this paper, the relation between the symmetrical components of the three-phase quantities and the harmonic components of the Extended Park’s Vector Approach (EPVA) is established. It is proved that the negative component and the EPVA harmonic component at a frequency twice the supply frequency are directly related to the fault occurrence. It is also proved that the healthy motor negative impedance is constant and not load-dependent. Based on this, the negative impedance/admittance is indirectly analyzed, through the combined use of the Extended Park´s Vector Approach of both voltage and current, and is explored for the fault diagnostics. Experimental results, obtained for different load torques, unbalanced supply voltage values, and fault severity levels, show that inter-turn short-circuit fault diagnostics can be achieved even under unbalanced supply voltage conditions based on the analysis of the motor admittance, at a frequency twice the supply frequency, that is the negative sequence admittance.

Stator inter-turn short-circuit fault diagnosis in induction motors applying VI loci-based technique

Due to the susceptibility of low-power three-phase induction motors (IMs) to stator inter-turn short-circuits (ITSC), researchers have been required to develop practical and cost-effective online systems for detecting faults in the stator winding. It is highly recommended to detect faults at a low fault severity level, as the fault current circulating through the shorted turns exponentially increases due to a reduction in insulation resistance and associated insulation degradation. The proposed Voltage–Current (VI) loci-based technique is a non-invasive, less expensive, and fault-sensitive approach that has been scarcely implemented in industries for stator ITSC faults analysis and accurate faulty phase identification, even under low fault severity. The proposed algorithm can compensate for various factors such as unbalanced supply voltages, asymmetries in winding construction, and errors in sensing element calibration. Experimental results presented for a three-phase 1 hp, 415 V SCIM investigate and validate the competency of the proposed work.

Modeling of Symmetrical Six-Phase Induction Machines Under Stator Faults

This paper addresses the occurrence of stator faults in six-phase induction machines, with a symmetrical winding, operating under both balanced and unbalanced supply voltage conditions. A dynamic model of the motor was developed in Matlab/Simulink environment, taking into consideration the occurrence of inter-turn short circuit faults. To validate the proposed approach, simulation results, covering the full range of operation and different fault severity levels, are compared with the motor experimental performance. The voltages and currents obtained from the simulation present a good agreement with the corresponding experimental results, demonstrating the accuracy of the proposed model

Bearing Fault Detection in Three-Phase Induction Motors Using Support Vector Machine and Fiber Bragg Grating

Due to its robustness and cost-effectiveness, the three-phase induction motor (TIM) has become the most widespread electric machine today. However, like any other equipment, it is vulnerable to a fault, and about 52% of these are related to bearings. This work presents the detection of flaws in the outer bearing’s raceway from the measurement of motor dynamic strain signals collected from sensors based on fiber Bragg grating (FBG). Three different degrees of severity were considered for faults in the outer bearing’s raceway. The tests were carried out on the motor operating under no-load conditions, with 47 different power supply frequencies. This work proposes a support vector machine (SVM) classifier to identify fault severity levels. Feature extraction was performed using two techniques: selecting the four highest peaks in the frequency spectrum and principal component analysis (PCA). The supervised SVM classifiers show that the dataset formed from the PCA presented a higher hit rate than the dataset constituted by the four highest peaks, with 99.82% and 92.73%, respectively. Based on the methodology presented in this work, it was possible to validate the use of FBG to detect bearing faults. Regardless of the degree of severity of the fault analyzed, the sensor detected its characteristic frequency. Based on the methodology presented in this work, it was possible to validate the use of FBG to detect bearing faults. Regardless of the degree of severity of the flaw analyzed, the sensor detected its characteristic frequency.

On Bayesian Optimization-Based Residual CNN for Estimation of Inter-Turn Short Circuit Fault in PMSM

Interturn short circuit (ITSC) fault diagnosis at its early stage is very essential to improve the security of permanent magnet synchronous motors. In this article, a Bayesian optimization (BO) based residual convolutional neural network (CNN) algorithm for ITSC fault diagnosis was proposed. There are mainly two aspects of the contributions. First, it is a challenge to apply a conventional CNN on time-series signal tasks as they can only look back on history with a liner size, which will limit the network depth. Moreover, to obtain enough characteristics from acquired signals, the proposed algorithm requires adequate network depth, which may result in degradation difficulties. To solve these problems, a residual connection is embedded into the dilated CNN. Second, with the increase of network depth, the possible combination of hyperparameters will grow geometrically, thus the hyperparameters tuning will cost considerable time. To overcome this, the BO algorithm is adopted to select the optimal hyperparameter combination automatically. To demonstrate the effectiveness of the proposed algorithm, a motor fault experiment with various operating conditions was conducted on the motor that can be set for 17 fault severity levels. The test results and comparisons with other five algorithms show the advantage of the proposed algorithm.

Real-Time Prognostics and Health Management Without Run-to-Failure Data on Railway Assets

Prognosis is a challenging technology that aims to accurately predict and estimate the remaining useful life of a component or system in order to enhance its reliability and performance. Although prognosis research for predictive maintenance is a well-researched topic, practical examples of successful prognostic applications remain scarce. This is due to the lack of available run-to-failure data to build the prediction model as maintenance is usually conducted regularly to avoid significant defects. This paper proposes a novel prognosis method that can be applied to real-world railway maintenance planning without employing run-to-failure data. The key idea is that the fault severity assessment and approximate remaining time prediction are often all that is needed in order to plan maintenance. Firstly, using motor current signals, a degradation indicator on railway door systems is generated based on the dynamic time warping method to measure similarity between typical normal and faulty behaviour. Then, the K-means algorithm is applied to assess fault severity, followed by the representative time estimation for each level of fault severity. This estimation thus allows the remaining time prediction until reaching the critical fault severity level without using run-to-failure data. As a result, the proposed method enables predictive maintenance planning for railway door systems. In addition, the fault severity threshold can be updated by additional operational data, enabling the remaining time prediction to be more reliable. Furthermore, the proposed method can be applied to conventional railway assets and other electro-mechanical actuators as motor current signals are primarily available from the controller or motor drive without additional sensors.

Model-Based Field Winding Interturn Fault Detection Method for Brushless Synchronous Machines

The lack of available measurements makes the detection of electrical faults in the rotating elements of brushless synchronous machines particularly challenging. This paper presents a novel and fast detection method regarding interturn faults at the field winding of the main machine, which is characterized because it is non-intrusive and because its industrial application is straightforward as it does not require any additional equipment. The method is built upon the comparison between the theoretical and the measured exciter field currents. The theoretical exciter field current is computed from the main machine output voltage and current magnitudes for any monitored operating point by means of a theoretical healthy brushless machine model that links the main machine with the exciter. The applicability of the method has been verified for interturn faults at different fault severity levels, both through computer simulations and experimental tests, delivering promising results.

Inter-Turn Fault Detection of Induction Motors Using End-Shield Leakage Fluxes

In this article, a non-invasive online technique to detect inter-turn stator winding fault of induction motors using end-shield leakage fluxes is proposed. The fault indicator is based on the fundamental component of the summation of the leakage fluxes which are spatially separated by an angle of 360°/number of the poles. Theoretical analysis using the analytical and finite element models revealed that the proposed fault indicator is independent of number of poles and location of the fault in any of the coils of the stator windings. Both theoretical analysis and experimental validation on a 10 hp induction motor showed that the proposed anomaly indicator is directly proportional to the fault severity level. A comparative study with various non-invasive, terminal measurements-based online techniques reveals that the proposed methodology is the most sensitive technique with an additional advantage of more than 92% reduction in the sensor cost for a 4-pole, 10 hp induction motor. Real-time evaluation of the proposed anomaly indicator with the prototype Motor Winding Monitoring Device (MWMD) shows that even under intermittent conditions, as small as 1-turn fault (0.5%) can be detected within 150 ms without any false alarms.

Multiple Sensor Fault Detection Using Index-Based Method.

The research on sensor fault detection has drawn much interest in recent years. Abrupt, incipient, and intermittent sensor faults can cause the complete blackout of the system if left undetected. In this research, we examined the observer-based residual analysis via index-based approaches for fault detection of multiple sensors in a healthy drive. Seven main indices including the moving mean, average, root mean square, energy, variance, first-order derivative, second-order derivative, and auto-correlation-based index were employed and analyzed for sensor fault diagnosis. In addition, an auxiliary index was computed to differentiate a faulty sensor from a non-faulty one. These index-based methods were utilized for further analysis of sensor fault detection operating under a range of various loads, varying speeds, and fault severity levels. The simulation results on a permanent magnet synchronous motor (PMSM) are provided to demonstrate the pros and cons of various index-based methods for various fault detection scenarios.

Controlled generation of unseen faults for Partial and Open-Partial domain adaptation

New operating conditions can result in a significant performance drop of fault diagnostics models due to the domain shift between the training and the testing data distributions. While several domain adaptation approaches have been proposed to overcome such domain shifts, their application is limited if the fault classes represented in the two domains are not the same. To enable a better transferability between two different domains, particularly in setups where only the healthy data class is shared between the two domains, we propose a new framework for Partial and Open-Partial domain adaptation based on generating distinct fault signatures with a Wasserstein GAN. The main contribution of the proposed framework is the controlled data generation with two characteristics. Firstly, previously unobserved target faults can be generated by having only access to healthy target and faulty source samples. Secondly, distinct fault types and severity levels can be generated precisely. The proposed method is especially suited for extreme domain adaption settings that are particularly relevant in the context of complex and safety-critical systems, where only one class is shared between the two domains. We evaluate the proposed framework on Partial as well as Open-Partial domain adaptation tasks on two bearing fault diagnostics case studies. In the evaluated case studies the proposed methodology demonstrated superior results compared to other methods, particularly in the presence of large domain gaps. The experiments conducted in different label space settings (Partial and Open-Partial) showcase the versatility of the proposed framework.

Fault feature analysis and detection of progressive localized gear tooth pitting and spalling

Fault feature analysis of gear tooth spalling plays a vital role in gear fault diagnosis. Understanding how fault features evolve as a fault progresses is key to fault severity assessment. Due to the complicated nature of gear meshing, fault features and their development as the fault severity progresses remain mostly unknown. The assessment of fault severity is generally based on the hypothesis that ‘the more severe the fault, the stronger the fault symptom’, an assumption that has not been experimentally validated. This paper provides a comprehensive, experimental analysis of the evolution of fault vibration features of a gear transmission with progressive localized gear tooth spalling. The effects of rotational speed on the vibration features of the gear transmission are analysed. Changes in fault features (e.g. periodic impulses and sideband phenomena) under different fault severity levels and speed conditions are compared. Results indicate that the number, amplitude and distribution of sidebands increase nonlinearly as the fault progresses. Based on feature analysis, a new health indicator of the mean of the nth order peaks is proposed to detect progressive localized tooth spalling. Results indicate that the proposed indicator shows very good performance for tracking the severity of progressive tooth spalling under different speed conditions.

Prototype of 3D Reliability Assessment Tool Based on Deep Learning for Edge OSS Computing

We focus on an estimation method based on deep learning in terms of fault correction time for the operation reliability assessment of open-source software (OSS) under the environment of an edge computing service. Then, we discuss fault severity levels in order to consider the difficulty of fault correction. We use a deep feedforward neural network in order to estimate fault correction times. In particular, we consider the characteristics of fault trends by using three-dimensional graphs. Therefore, we can increase the recognizability of the proposed method based on deep learning for large-scale fault data from the standpoint of fault severity levels under edge OSS operation.

A key-factor denoising strategy for quasi periodic non-stationary incipient faults diagnosis

Currently, the analysis method based on monitoring data has become an effective means of machinery fault diagnosis, and the fault diagnosis with obvious data features has achieved fruitful results. However, the incipient fault signals of equipment not only show the characteristics of weak intensity, quasi periodic and non-stationary, but also are submerged in strong background noise, which often makes it difficult to extract effective information directly from the original signals. Therefore, in order to effectively solve the problem of incipient fault diagnosis, and considering the capability of sparse autoencoder (SAE) to extract features automatically, this paper proposes a key-factor denoising strategy and an improved SAE network, and then an improved SAE network with key-factor denoising strategy (KF-ISAE) based intelligent diagnosis method for quasi periodic non-stationary incipient faults is proposed. The main contributions of the proposed method are as follows. On the one hand, signal denoising that cannot be ignored in fault diagnosis is achieved by the developed incipient faults sensitivity based key-factor denoising strategy, and on the other hand, for SAE, the blindness of feature learning is handled by the formed weights constraints. In addition, the health condition identification and the fault severity level determination of machinery are completed by the improved SAE network designed in this paper. Finally, verification and comparative experiments show the effectiveness and practicability of the proposed method.

Bearing fault diagnosis under various operation conditions using synchrosqueezing transform and improved two-dimensional convolutional neural network

In real-world industrial applications, bearings are typically operated under variable speeds and loads depending on the production condition, which results in nonstationary vibration signals from the bearings. Synchrosqueezing transform is a method that can effectively reflect the change in frequency with time, which is suitable for processing nonstationary bearing signals. However, significant classification features are difficult to extract from time–frequency information when operation conditions such as speed and load change frequently. Hence, an improved two-dimensional (2D) convolutional neural network (CNN) named the 2D multiscale cascade CNN (2D MC-CNN) is proposed for performing bearing fault diagnosis under various operating conditions. In a 2D MC-CNN, a multiscale information fusion layer is added prior to the convolutional layer of a conventional CNN to form MC images such that sensitive bands can be acquired for fault recognition. Experiments are conducted on bearings by considering various sets of fault categories and fault severity levels under six operating conditions. The experimental results show that the proposed method effectively extracts fault-related features and demonstrates excellent diagnostic accuracy and robustness. Comparisons with the original CNN and other typically used fault diagnosis methods based on the same dataset demonstrate the effectiveness of the proposed 2D MC-CNN and bearing fault diagnosis method.