Early Fault Detection Method Research Articles

New Health Indicator Construction and Fault Detection Network for Rolling Bearings via Convolutional Auto-Encoder and Contrast Learning

As one of the most important components in rotating machinery, if bearings fail, serious disasters may occur. Therefore, the remaining useful life (RUL) prediction of bearings is of great significance. Health indicator (HI) construction and early fault detection play a crucial role in data-driven RUL prediction. Unfortunately, most existing HI construction methods require prior knowledge and preset trends, making it difficult to reflect the actual degradation trend of bearings. And the existing early fault detection methods rely on massive historical data, yet manual annotation is time-consuming and laborious. To address the above issues, a novel deep convolutional auto-encoder (CAE) based on envelope spectral feature extraction is developed in this work. A sliding value window is defined in the envelope spectrum to obtain initial health indicators, which are used as preliminary labels for model training. Subsequently, CAE is trained by minimizing the composite loss function. The proposed construction method can reflect the actual degradation trend of bearings. Afterwards, the autoencoder is pre-trained through contrast learning (CL) to improve its discriminative ability. The model that has undergone offline pre-training is more sensitive to early faults. Finally, the HI construction method is combined with the early fault detection method to obtain a comprehensive network for online health assessment and fault detection, thus laying a solid foundation for subsequent RUL prediction. The superiority of the proposed method has been verified through experiments.

Ensefgram: An optimal demodulation band selection method for the early fault diagnosis of high-speed train bearings

Early fault diagnosis of high-speed train bearings under road spectrum shocks is challenging. As a conventional narrowband demodulation method, the Fast Kurtogram (FK) class of algorithms have some drawbacks: the fixed-band segmentation strategy makes it difficult for the method to find the optimal demodulation frequency band (ODFB); other major limitations are the poor robustness of the statistical index of the blind algorithm and the poor adaptiveness of the target algorithm given the need to the know the bearing operating state in advance. To address these issues, this paper proposes Ensefgram, an early fault detection method for high-speed train bearings with the acoustic emission signal as an example. First, a filter bank is designed by combining the spectral trend with the median dichotomy strategy, which adaptively segments the frequency band and retains the integrity of the sub-band fault components to the maximum extent. In addition, a novel statistical index, termed the maximum weighted spectral energy frequency factor, is proposed, and it is combined with the envelope spectrum negentropy (ES-negentropy) to form a composite index ENSEF, which can select the ODFB blindly. Finally, the proposed method was validated by simulation and applying test signals close to the real working conditions of a high-speed train. Compared with FK, Infogram, and other methods, Ensefgram provides more accurate detection results for the early fault diagnosis of bearings.

Research on early fault detection method for a new distribution system based on automatic arc power

The assimilation of copious renewable energy sources into contemporary distribution systems has markedly transmuted fault dynamics, engendering fluctuating operational states and curtailed short-circuit currents. Consequently, incipient arc faults may herald imminent system short-circuits, obviating the complexities associated with the new distribution system's disconnection by facilitating their precocious detection. A quantitative analytical paradigm has been contrived for the nascent fault current and temperature, while theoretical scrutiny elucidates the impediments to dependable detection predicated solely on current or temperature metrics. An innovative detection methodology, amalgamating current and temperature readings and leveraging arc power as a composite indicator, has been advanced. This proposed mechanism synergizes the alacrity of the electrical signal—specifically, the zero-sequence current—with the heightened sensitivity and dependability of the thermal signal—temperature—thereby surpassing the reliability of extant approaches. The efficacy of this method has been corroborated through PSCAD simulations, empirical field data, and controlled laboratory fault experimentation.

Time-Series Transfer Learning: An Early Stage Imbalance Fault Detection Method Based on Feature Enhancement and Improved Support Vector Data Description

Early stage fault detection plays a pivotal role in Industrial equipment accidents avoidance and scientific maintenance. While limited by the complex operation background, its application encounters with the conundrum of fault feature indistinctness. To address the challenge, a time-series transfer learning (TSTL) method is proposed, which contains two phases: first, early stage series are transferred to their corresponding serious stage for fault feature enhancement. Moreover, due to the improvement of model structure and loss function, the limitation of mismatched working condition is well-weaken. Second, a transferred fault mode recognition model is trained by using transferred normal series that provides a novel solution for data imbalance. Finally, the TSTL method is verified by actual vibration datasets of power pole tower bolts. Its superiority in feature transfer and fault detection is confirmed by several groups of comparative experiments and results demonstrate TSTL outperforms mainstream methods.

An Early Fault Detection Method for Wind Turbine Main Bearings Based on Self-Attention GRU Network and Binary Segmentation Changepoint Detection Algorithm

The condition monitoring and potential anomaly detection of wind turbines have gained significant attention because of the benefits of reducing the operating and maintenance costs and enhancing the reliability of wind turbines. However, the complex and dynamic operation states of wind turbines still pose tremendous challenges for reliable and timely fault detection. To address such challenges, in this study, a condition monitoring approach was designed to detect early faults of wind turbines. Specifically, based on a GRU network with a self-attention mechanism, a SAGRU normal behavior model for wind turbines was constructed, which can learn temporal features and mine complicated nonlinear correlations within different status parameters. Additionally, based on the residual sequence obtained using a well-trained SAGRU, a binary segmentation changepoint detection algorithm (BinSegCPD) was introduced to automatically identify deterioration conditions in a wind turbine. A case study of a main bearing fault collected from a 50 MW windfarm in southern China was employed to evaluate the proposed method, which validated its effectiveness and superiority. The results showed that the introduction of a self-attention mechanism significantly enhanced the model performance, and the adoption of a changepoint detection algorithm improved detection accuracy. Compared to the actual fault time, the proposed approach could automatically identify the deterioration conditions of main bearings 72.47 h in advance.

Early rub-impact fault detection of rotor systems via deterministic learning

Early fault detection of rotor systems with rub-impact is a very important and challenging task in practical applications, especially for rub-impact faults in the initial stage. In this paper, the rotor unbalance is taken as the control parameter to yield incipient rub-impact faults, and an early detection scheme for rub-impact faults is proposed based on deterministic learning. A large enough dynamical pattern bank including health (without rub-impact), sub-health (rub-impact inception), fault (rub-impact fault developing) is established based on the underlying system dynamics which are accurately identified through deterministic learning. Compared with the time–frequency analysis methods, the early detection of rub-impact faults is implemented from the viewpoint of accurate identification of system dynamics, which can provide a more effective detection. The differences between the stored dynamical patterns with that of a test system are directly reflected in system residuals, and the early weak rub-impact faults can be detected quickly based on the comparison of the system residuals. Furthermore, the degree of rub-impact faults is also determined accurately. The results of numerical simulation and experiments show the effectiveness of the proposed early rub-impact fault detection method.

Research on an ID-PCA Early Fault Detection Method for Rolling Bearings

Since the rolling bearing is complex during the signal acquisition process, there is a certain loss during the process of collecting the vibration signal. This has led to the weakness of the early fault characteristics of the rolling bearing, affecting the accuracy of the rolling bearing fault feature extraction. In response to the above problems, an early fault detection method based on the Improved Deep Principal Component Analysis (ID-PCA) is proposed. The proposed method uses the time-series characteristic information of the vibration signal to establish a model, which solves the problem that the principal component analysis method cannot detect the vibration signal directly. Through the deep decomposition theorem, a multi-layer data processing model is established to fully mine the weak fault features in the vibration signal. It can solve the problem of inaccurate early fault detection results due to weak fault feature information. The reliability of this method is proved theoretically through sensitivity analysis. Finally, through experimental simulation, the accuracy and feasibility of this method are proved from the perspective of practice.

Design of Active Fault-Tolerant Control System for Multilevel Inverters to Achieve Greater Reliability With Improved Power Quality

Modern renewable energy systems mostly utilize multilevel converter applications for improved power quality and grid synchronization purposes. A multilevel converter is an electrical device that may offer several amounts of voltage levels at the output in order to make the output more comparable to a pure sine wave. The integrity of the multilevel converter depends on the reliability of individual switches such that the converter will collapse in case of faults in these switches. In this paper, a novel 7-level Fault-Tolerant Cascaded H-Bridge Multilevel Inverter (FT-CHB-MLI) has been proposed that offers high reliability with improved power quality. A dedicated Fault Detection and Isolation (FDI) unit has been built to diagnose the faulty switch and replace it with a standby redundant switch. Total harmonic distortion and the determination of a normalized output voltage factor are employed for fault diagnosis. The Phase Disposition Pulse Width Modulation (PD-PWM) technique has been utilized for switching due to its superior performance as compared to other conventional techniques. This early fault detection method not only identified the issues but also performed preventative actions to keep the system healthy and stable. The proposed system was tested on the MATLAB / Simulink environment to verify its performance. The simulation results demonstrated that the THD has been reduced to almost 18% with a significant increase in reliability with advanced fault-tolerant architecture consisting of FDI units. The reliability analysis was carried out using Markov chains that also showed its increased reliability. A comparison of the proposed work with literature was also carried out to demonstrate its superior performance with increased reliability.

An Early Fault Detection Method of Rotating Machines Based on Unsupervised Sequence Segmentation Convolutional Neural Network

Early fault detection (EFD) is vital for mechanical systems to reduce downtime and increase stability. The main challenge of EFD for rotating machines is to extract discriminative features from noisy signals to identify early faults. However, the lack of labels for the whole lifecycle data hinders the application of some powerful supervised deep learning methods in EFD. Besides, many EFD methods have to set a criterion manually, such as a threshold, to judge whether an early fault has occurred. To address these challenges, this article proposes a novel EFD method based on unsupervised sequence segmentation convolutional neural network (USSCNN). At first, frequency-domain features are extracted from raw signals and converted to 2-D gray images. Then, historical lifecycle data are labeled by USSCNN so that a CNN classifier can be trained with these labeled data. The deep features of the historical data learned by the CNN classifier are utilized to train the health index (HI) assessment model. The proposed method is tested on three bearing datasets. The results have shown that the proposed method can detect incipient faults earlier than the comparing methods with lower false alarms. Also, the HIs learned by the HI assessment model shown that the proposed method can extract discriminative features for EFD. More importantly, the proposed method can detect an early fault by the well-trained classifier, which avoids manual criterion-making. Results of comparison demonstrated the effectiveness and the robustness of the proposed method.

Bearing early fault detection and degradation tracking based on support tensor data description with feature tensor

In view of easily interfered properties of bearing incipient fault signal, it is beneficial to mine more useful information from vibration signal for bearing state monitoring. For exploring local information and utilizing correlation between the characteristic of different frequency bands, feature tensor based on Wavelet Packet (WP) decomposition is introduced to characterize bearing fault state as reflected by vibration signal. Then the bearing state monitoring can be transformed as a one-class classification problem in tensor space, in which the abnormal samples represent the fault states and the normal samples mean the health states of the bearings, respectively. Combining with Support Tensor Data Description (STDD), an early fault detection method is proposed by employing feature tensor correlating with bearing fault directly. To further quantify the fault degree of bearing, a tensor-based anomaly index (AIt) is presented by taking advantage of the distance between feature tensor sample to spherical surface generated by STDD in tensor space. Experimental results of bearing verify that the proposed tensor-based fault detection method is effective for bearing early fault detection, and AIt is sensitive to bearing operation state and therefore suitable for fault degradation tracking. Further experiments show that the tensor-based method is robust to noise and small sample size problem.

A new deep domain adaptation method with joint adversarial training for online detection of bearing early fault

For online early fault detection of rolling bearings in non-stop scenarios, one of the main concerns is the model bias caused by the distribution shift between offline and online working conditions. Under such concern, how to improve the feature sensitivity to early faults and the robustness of detection model has become a key challenge of improving the effectiveness of online detection. To solve this problem, a new online early fault detection method is proposed in this paper based on a strategy of deep transfer learning. First, a new robust state assessment method is presented. By introducing priori degradation information in the anomaly detection process of the isolated forest algorithm, this method can accurately assess the normal state and early fault state under noise interference. Second, a new deep domain adaptation algorithm is proposed. The algorithm uses the results of state assessment as output labels, and designs a deep domain adaptation neural network for joint adversarial training at feature level and model level simultaneously. Then a domain-invariant feature representation can be extracted from the data of different working conditions, and an online detection model can then be constructed. Comparative experiments are run on two bearing datasets IEEE PHM Challenge 2012 and XJTU-SY, and the results verifies the effectiveness of the proposed method in false alarm number and detection location.

Early fault detection of gas turbine hot components under different ambient and operating conditions

Failure of hot components in gas turbines often causes catastrophic results. Early fault detection can prevent serious incidents and improve the availability. A novel early fault detection method of hot components is proposed in this article. Exhaust gas temperature is usually used as the indicator to detect the fault in the hot components, which is measured by several exhaust thermocouples with uniform distribution at the turbine exhaust section. The healthy hot components cause uniform exhaust gas temperature (EGT) profile, whereas the hot component faults could cause the uneven EGT profile. However, the temperature differences between different thermocouple readings are also affected by different ambient and operating conditions, and it sometimes has a greater influence on EGT than the faults. In this article, an accurate EGT model is presented to eliminate the influence of different ambient and operating conditions on EGT. Especially, the EGT profile swirl under different ambient and operating conditions is also included by considering the information of the thermocouples’ spatial correlations and the EGT profile swirl angle. Based on the developed EGT model, the detection performance of early fault detection of hot components in gas turbine is improved. The accuracy and effectiveness of the developed early fault detection method are evaluated by the real-world gas turbine data.

Early Fault Detection Method of Rolling Bearing Based on MCNN and GRU Network with an Attention Mechanism

Aiming at the problem of early fault diagnosis of rolling bearing, an early fault detection method of rolling bearing based on a multiscale convolutional neural network and gated recurrent unit network with attention mechanism (MCNN-AGRU) is proposed. This method first inputs multiple time scales rolling bearing vibration signals into the convolutional neural network to train the model through multiscale data processing and then adds the gated recurrent unit network with an attention mechanism to make the model predictive. Finally, the reconstruction error between the actual value and the predicted value is used to detect the early fault. The training data of this method is only normal data. The early fault detection in the operating condition monitoring and performance degradation assessment of the rolling bearing is effectively solved. It uses a multiscale data processing method to make the features extracted by CNN more robust and uses a GRU network with an attention mechanism to make the predictive ability of this method not affected by the length of the data. Experimental results show that the MCNN-AGRU rolling bearing early fault diagnosis method proposed in this paper can effectively detect the early fault of the rolling bearing and can effectively identify the type of rolling bearing fault.

A Method for Fans’ Potential Malfunction Detection of ONAF Transformer Using Top-Oil Temperature Monitoring

The fan is one of the key components of the power transformer cooling system. The operating condition of fans determines transformers’ internal temperature rise and long-term reliability. However, at present, the fans’ condition monitoring only includes switch status (online) and regular maintenance (offline), online direct monitoring of the fans’ operating condition is lacking due to economic costs. In view of the above-mentioned problem, this paper proposes a transformer fan early fault detection method based on the oil exponent, which is monitored by the existing transformer top-oil temperature data, thereby detecting the abnormality of the fans. In this method, the oil exponent was chosen as the characteristic criterion. First, to obtain the range of oil exponent in different cooling modes, a set of physical models describing global oil flow and its interaction with air was established based on fluid dynamics and heat transfer principle. Then, regarding the constantly changing top-oil temperature, ambient temperature and load current, an oil exponent tracking algorithm using particle swarm optimization (PSO) was proposed within an improved IEC dynamic thermal model. The operation data from an oil-immersed transformer with a rated capacity of 120-MVA and rated voltage of 220-kV was selected to verify the above methods under two different scenarios.

Early Fault Detection of Gas Turbine Hot Components Based on Exhaust Gas Temperature Profile Continuous Distribution Estimation

Failures of the gas turbine hot components often cause catastrophic consequences. Early fault detection can detect the sign of fault occurrence at an early stage, improve availability and prevent serious incidents of the plant. Monitoring the variation of exhaust gas temperature (EGT) is an effective early fault detection method. Thus, a new gas turbine hot components early fault detection method is developed in this paper. By introducing a priori knowledge and quantum particle swarm optimization (QPSO), the exhaust gas temperature profile continuous distribution model is established with finite EGT measuring data. The method eliminates influences of operating and ambient condition changes and especially the gas swirl effect. The experiment reveals the presented method has higher fault detection sensitivity.

Early Fault Detection of Planetary Gearbox Based on Acoustic Emission and Improved Variational Mode Decomposition

Planetary gearbox is applied more and more extensively in the modern industry. Its condition directly determines the availability and the operation safety of the machinery system that utilizes it for power transmission. To enhance its reliability, an early fault detection method is proposed in this study, which is based on the acoustic emission (AE) technique and the improved variational mode decomposition (VMD). AE data are expected to contain more valuable information about the early fault than the vibration data. Mutual information is used to determine its layer that can avoid over-decomposition or under-decomposition. Then, energy entropy is used to analyze intrinsic mode functions extracted by the improved VMD. By comparing values of energy entropy, it is easy to distinguish different working conditions. The main contributions of this study include: 1) An improved VMD method for fault detection of planetary gearbox is proposed. 2) This method provides a novel strategy for detecting early fault in different positions of planetary gearbox by utilizing the advantages of AE technique.

An early fault detection method for induced draft fans based on MSET with informative memory matrix selection

Early fault detection of induced draft (ID) fans is very important to improve the reliability by providing predictive maintenance and reducing unscheduled shutdowns. This study proposed an early fault detection method for ID fans based on MSET with informative memory matrix selection. Firstly, to obtain an informative memory matrix, the discrete particle swarm optimization (DPSO) was utilized to search samples with large condition information. An accurate MSET model was then developed based on the memory matrix to produce predictions of the feature variables. Finally, a similarity index that represents the health status of the equipment was defined to give warnings of early faults. An application to detect the early faults of an ID fan in a coal-fired power plant was demonstrated to illustrate the effectiveness of proposed method.

Early fault detection method for rolling bearing based on multiscale morphological filtering of information-entropy threshold

The scale of structure element is especially important to obtain good filtering results in multiscale morphological filtering (MMF) method. In general, the optimal scale of structure element is set to be a fixed value in traditional morphological filter, therefore it is difficult to extract the fault feature from rolling bearing vibration signal effectively. A novel multiscale morphological filtering algorithm is proposed based on information-entropy threshold (IET-MMF) for early fault detection of rolling bearing. Compared with traditional MMF method, several optimal scales of structure elements are achieved according to the energy distribution characteristic of different vibration signals. The information entropy theory is applied to quantify the analyzed signals, and the optimal threshold of information entropy is obtained by iterative algorithm to ensure integrity of useful information. The simulation and rolling bearing experimental analysis results show that the IET-MMF method can extract fault features of vibration signals effectively.

Early fault detection of bearings based on adaptive variational mode decomposition and local tangent space alignment

PurposeEarly 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.Design/methodology/approachIn 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.FindingsThe 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.Originality/valueAn 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.

Machinery Early Fault Detection Based on Dirichlet Process Mixture Model

The most commonly used single feature-based anomaly detection method for the complex machinery, such as large wind power equipment, steam turbine generator sets, and reciprocating compressors, exhibits a defect of low-alarm accuracy due to the non-stationary characteristic of the vibration signals. In order to improve the accuracy of fault detection, a novel method based on the Dirichlet process mixture model (DPMM) is proposed. First, the features of the mechanical vibration signals are used to construct the feature space of the equipment. The DPMM modeling method is then applied to self-learn the probabilistic mixture model of the feature space. The normal working condition model is used as the benchmark model. The early fault detection is realized by using a precise difference measurement method based on Kullback-Leibler divergence to calculate the difference between the real-time model and the benchmark model accurately, and by comparing the calculation result with a self-learned alarm threshold. The effectiveness and the adaptability of this novel early fault detection method are verified by comparing it to the single feature-based anomaly detection method and the Gaussian mixture model (GMM)-based early fault detection method.