Incipient Fault Detection Research Articles

An integrated design method for active fault diagnosis and control

SummaryThe injection of an auxiliary input signal for active fault diagnosis may cause the change of system control performance in closed‐loop operation. This paper presents a novel integrated design method for active fault diagnosis and tracking control in order to facilitate the detection of incipient faults, meanwhile ensuring the normal operation of the closed‐loop system. First, a zonotopic filter is generated based on radius minimization method to estimate the uncertain bound set of systems states. Then, an auxiliary input signal is designed to separate the reachable sets of the system in healthy and faulty operation situations. Generally, the auxiliary input signal is required to be designed minimizing the system closed‐loop performance degradation. In this paper, a tracking controller is designed altogether with the auxiliary input signal to make the system output track a given reference. An optimal index is proposed to minimize simultaneously the energy of the auxiliary input signal and the output tracking error. Next, a fault detection logic is provided based on checking whether the closed‐loop system state belongs to healthy zonotope. Finally, the feasibility and advantages of this work are verified by means of a numerical example.

Stator Fault Diagnosis of Induction Motor Based on Discrete Wavelet Analysis and Neural Network Technique

A novel approach by introducing a statistical parameter to estimate the severity of incipient stator inter-turn short circuit (ITSC) faults in induction motors (IMs) is proposed. Determining the incipient ITSC fault and its severity is challenging for several reasons. The stator currents in the healthy and faulty cases are highly similar during the primary stage of the fault. Moreover, the conventional statistical parameters resulting from the analysis of fault signals do not consistently show a systematic variation with respect to the increase in fault intensity. The objective of this study is the early detection of incipient ITSC faults. Furthermore, it aims to determine the percentage of shorted turns in the faulty phase, which acts as an indicator for severe damage to the stator winding. Modeling of the motor in healthy and defective cases is performed using the Clarke Concordia transform. A discrete wavelet transform is applied to the motor currents using a Daubechies-8 wavelet. The statistical parameters <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$L_{1}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$L_{2}$</tex> norms are computed for the detailed coefficients. These parameters are obtained under a variety of loads and defects to acquire the most accurate and generalized features related to the fault. Combining <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$L_{1}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$L_{2}$</tex> norms creates a novel statistical parameter with notable characteristics to achieve the research aim. An artificial neural network-based back propagation algorithm is employed as a classifier to implement the classification process. The classifier output defines the percentage of defective turns with a high level of accuracy. The competency of the adopted methodology is validated via simulations and experiments. The results confirm the merits of the proposed method, with a classification test correctness of 95.29%.

A stage-wise non-Gaussian statistical process control strategy

The article delves into the development of a Non-Gaussian Process Monitoring Strategy for a Copper Cathode Manufacturing Unit (CCMU). The monitoring strategy being devised highlighted the issue of multi-stage process monitoring via the usage of Multi-block Independent Component Analysis (MBICA) techniques. MBICA is the multi-block variant of ICA technique which is prevalently used for process laden with non-Gaussian or non-normal data. Development of the monitoring strategy involved detection of fault(s) and their subsequent diagnosis. Detection of fault(s) was carried out by employment of I2 control chart whose control limit was established via Bootstrap procedure. The diagnosis of the detected fault was carried out by employment of fault diagnostic statistic. An amalgamation of MBICA and Multivariate Exponentially Weighted Moving Average (MEWMA) are also known as MBICA-MEWMA approach was also proposed for detection of incipient fault(s). The monitoring strategy thus developed was showcased for a CCMU which specialised in the manufacture of copper cathode which has got varied practical applications. The monitoring strategy thus devised was able to detect and diagnose the faults with appreciable accuracy.

Short-Circuit Fault Diagnosis on Induction Motors through Electric Current Phasor Analysis and Fuzzy Logic

Online monitoring of induction motors has increased significantly in recent years because these devices are essential components of any industrial process. Incipient fault detection in induction motors avoids interruptions in manufacturing processes and facilitates maintenance tasks to reduce induction motor timeout. Therefore, the proposal of novel approaches to assist in the detection and classification of induction motor faults is in order. In this work, a reliable and noninvasive novel technique that does not require computational demanding operations, since it just performs arithmetic calculations, is introduced for detecting and locating short-circuit faults in the stator windings of an induction motor. This method relies on phasor analysis and the RMS values of line currents, followed by a small set of simple if-then rules to perform the diagnosis and identification of stator winding faults. Obtained results from different experimental tests on a rewound induction motor stator to induce short-circuit faults demonstrate that the proposed approach is capable of identifying and locating incipient and advanced deficiencies in the windings’ insulation with high effectiveness.

Incipient Fault Detection in Power Distribution System: A Time–Frequency Embedded Deep-Learning-Based Approach

Incipient fault detection in power distribution systems is crucial to improve the reliability of the grid. However, the non-stationary nature and the inadequacy of the training dataset due to the self-recovery of the incipient fault signal, make the incipient fault detection in power distribution systems a great challenge. In this paper, we focus on incipient fault detection in power distribution systems and address the above challenges. In particular, we propose an ADaptive Time-Frequency Memory (AD-TFM) cell by embedding wavelet transform into the Long Short-Term Memory (LSTM), to extract features in time and frequency domain from the non-stationary incipient fault signals. We make scale parameters and translation parameters of wavelet transform learnable to adapt to the dynamic input signals. Based on the stacked AD-TFM cells, we design a recurrent neural network with ATtention mechanism, named AD-TFM-AT model, to detect incipient fault with multi-resolution and multi-dimension analysis. In addition, we propose two data augmentation methods, namely phase switching and temporal sliding, to effectively enlarge the training datasets. Experimental results on two open datasets show that our proposed AD-TFM-AT model and data augmentation methods achieve state-of-the-art (SOTA) performance of incipient fault detection in power distribution system. We also disclose one used dataset logged at State Grid Corporation of China to facilitate future research.

Detection of Incipient Faults in Transformer by Dissolved Gas Analysis Using Artificial Intelligence Techniques

Detection of Incipient Faults in Transformer by Dissolved Gas Analysis Using Artificial Intelligence Techniques

Clustering Wind Turbines for SCADA Data-Based Fault Detection

Condition monitoring systems are commonly employed for incipient fault detection of wind turbines (WTs) to reduce downtime and increase availability. Data from supervisory control and data acquisition (SCADA) systems offer a potential monitoring solution. Multi-turbine approaches, which merge variables recorded from different WTs on the same wind farm, have been developed to improve fault detection performance by reducing the influence of variations in environmental conditions. However, in complex terrain, environmental conditions vary among WTs. Moreover, manual control (e.g., maintenance and curtailment) can also raise false alarms in some WTs. Here, the false alarm characteristics of a wind farm in complex terrain are investigated. A clustering-based multi-turbine fault detection approach is proposed, consisting of three steps: WT clustering, single-turbine modeling, and fault indicator calculation. First, k-medoids clustering with dependent multivariate dynamic time warping is applied for WT clustering. Then, an autoregressive neural network is used to construct a single-turbine model. Finally, residuals between median values of the model output of all WTs in the same cluster and the target WT are used to calculate the anomaly level. Evaluation results for real large-scale SCADA data confirm that the proposed approach raises fewer false alarms without degrading detection performance.

Broken Bar Fault Detection Using Taylor–Fourier Filters and Statistical Analysis

Broken rotor bars in induction motors make up one of the typical fault types that are challenging to detect. This type of damage can provoke adverse effects on the motors, such as mechanical and electrical stresses, together with an increase in electricity consumption, causing higher operative costs and losses related to the maintenance times or even the motor replacement if the damage has led to a complete failure. To prevent such situations, diverse signal processing algorithms have been applied to incipient fault detection, using different variables to analyze, such as vibrations, current, or flux. To counteract the broken rotor bar damage, this paper focuses on a motor current signal analysis for early broken bar detection and classification by using the digital Taylor-Fourier transform (DTFT), whose implementation allows fine filtering and amplitude estimation with the final purpose of achieving an incipient fault detection. The detection is based on an analysis of variance followed by a Tukey test of the estimated amplitude. The proposed methodology is implemented in Matlab using the O-splines of the DTFT to reduce the computational load compared with other methods. The analysis is focused on groups of 50-test of current signals corresponding to different damage levels for a motor operating at 50% and 75% of its full load.

Data-Driven Sensor Fault Diagnosis Under Closed-Loop Control With Slow Feature Analysis

Recently, data-driven fault diagnosis techniques, especially multivariate statistical process monitoring methods, have been extensively investigated and widely applied to industries. Whether the monitored system is under closed- or open-loop control has a strong impact on the development of fault diagnosis strategies, but this point has not been taken seriously into consideration. This work aims to provide an effective data-driven method for sensor fault diagnosis under closed-loop control with modified slow feature analysis (SFA). The SFA is first revisited and compared with the well-known principal component analysis approach. Through theoretical analysis and an intuitive example, the influence of feedback control on sensor fault diagnosis is demonstrated, and to achieve successful detection, a strategy of incorporating more variables for monitoring is suggested. Then, based on SFA, a new sensor fault detection and classification method is developed. Its detectability analysis is carried out, based on which improved methods are proposed for efficient detection of incipient sensor faults. Finally, case studies on the continuous stirred tank reactor benchmark process demonstrate the effectiveness of the proposed method.

Data-driven technology of fault diagnosis in railway point machines: review and challenges

Abstract Safety and reliability are absolutely vital for sophisticated Railway Point Machines (RPMs). Hence, various kinds of sensors and transducers are deployed on RPMs as much as possible to monitor their behaviour for detection of incipient faults and anticipation using data-driven technology. This paper firstly analyses and summarizes six RPMs’ characteristics and then reviews the data-driven algorithms applied to fault diagnosis in RPMs during the past decade. It provides not only the process and evaluation metrics but also the pros and cons of these different methods. Ultimately, regarding the characteristics of RPMs and the existing studies, eight challenging problems and promising research directions are pointed out.

Transform Waveforms Into Signature Vectors for General-Purpose Incipient Fault Detection

Power system equipment presents special signatures at the incipient stage of faults. As more renewables are integrated into the systems, these signatures are harder to detect. If faults are detected at an early stage, economical losses and power outages can be avoided in modern power grids. Many researchers and power engineers have proposed a series of signature-specific methods for one type of equipment’s waveform abnormality. However, conventional methods are not designed to identify multiple types of incipient faults (IFs) signatures at the same time. Therefore, we develop a general-purpose IF detection method that detects waveform abnormality stemming from multiple types of devices. To avoid the computational burden of the general-purpose IF detection method, we embed the abnormality signatures into a vector and develop a pre-training model (PTM) for machine understanding. In the PTM, signal “words,” “sentences,” and “dictionaries” are designed and proposed. Through the comparison with a machine learning classifier and a simple probabilistic language model, the results show a superior detection performance and reveal that the training radius is highly related to the size of abnormal waveforms.

Incipient fault detection of nonlinear chemical processes based on probability-related randomized slow feature analysis

Slow feature analysis (SFA) has achieved the successful applications in the chemical process fault detection field. However, the basic SFA omits the process nonlinearity and lacks the sufficient mining of the features’ probability information so that the incipient fault monitoring performance is unsatisfactory. To alleviate this problem, this paper proposes an improved SFA method, referred to as probability-related randomized SFA (PRSFA), to enhance the detection of incipient faults. Different from the current nonlinear SFA version based on the kernel function, which results in the huge computation complexity, the proposed method uses random Fourier mapping to deal with the data nonlinear transformation more efficiently. By combining the random Fourier mapping and SFA, a randomized SFA model is developed to capture the nonlinear slow features. Furthermore, in order to mine the probability distribution information hidden in these slow features, the Kullback Leibler divergence (KLD) is introduced to measure the changing of the slow features’ probability distributions. The original nonlinear slow features are converted into the corresponding KLD features, which are more sensitive to the incipient faults. Considering the random property of KLD features resulted from the random Fourier mapping, multiple randomized SFA sub-models are developed and all the sub-models are integrated through Bayesian inference mechanism to construct the global statistics for the whole system monitoring. The simulation results on a continuous stirred tank reactor (CSTR) system show that the proposed method has better incipient fault detection performance than the traditional SFA and kernel SFA methods.

An integrated monitoring scheme for wind turbine main bearing using acoustic emission

The condition monitoring of the main bearing (MB) plays a crucial role in the maintenance of wind turbines (WT), especially for direct-drive wind turbines (DDWT). However, due to the harsh operating environment and ultra-low rotating speed, the condition monitoring of the MB is still a challenging issue. In this study, an integrated monitoring scheme using acoustic emission (AE) is proposed for incipient fault detection and localization of MB. First, an rotating speed estimation approach using high-frequency envelope autocorrelation (HFEA) is developed to recover the accurate operating speed of MB. On this basis, the adapted spectral coherence (ASC) is explored to identify faulty sources buried under multiple disturbances. Finally, an effective damage localization model is further constructed to improve maintenance efficiency in practical applications. The performance of the proposed methodology is evaluated through two engineering cases with natural damages. Compared with state-of-the-art approaches, the proposed method can not only effectively detect the incipient damage of the MB, but also accurately determine the damage location. With this scheme, the inspection efficiency can be improved, thus it may provide a promising tool for the health management of WT.

Detection of an Incipient Fault for Dual Three-Phase PMSMs Using a Modified Autoencoder

For the detection of incipient interturn short-circuit (IITSC) faults of machines without shutting them down, there are still shortcomings of insufficient incipient fault features and a high false alarm rate. This is especially the case for dual three-phase permanent magnet synchronous motors (PMSMs) with complex winding structures, and this kind of incipient fault detection is more complicated. To solve this detection difficulty, an IITSC detection method for dual three-phase PMSMs is proposed based on a modified deep autoencoder (MDAE). This autoencoder (AE) adopts an improved distribution metric combined with the maximum mean discrepancy (MMD) and the maximum covariance discrepancy (MCD) to extract the fault feature from the common features, which can improve the feature difference between the normal state and the incipient fault state. Then, the permutation entropy of the extracted features is calculated to detect the IITSC faults. The results illustrate that this method can not only detect IITSC faults online effectively and robustly, but also reduce the false alarm rate of the fault detection for dual three-phase PMSMs.

Incipient fault detection of rotary steerable drilling tool equipment

Rotary steering drilling system is a high-end piece of equipment in oil and gas development. The reliable operation of its core equipment, the dynamic point-the-bit rotary steerable drilling tool (DPRSDT), is an important prerequisite for the normal operation of the drilling system. In this paper, the problem of incipient fault detection is investigated for the DPRSDT equipment based on the model method, which is expected to provide an important guarantee for the operation, maintenance, and health management of the system. First, the mathematical model of the DPRSDT is derived using the mechanism modeling method. Then, the moving weighted average approach is employed to improve the sensitivity of residuals to incipient faults. Based on the characteristics of the noncentral $\chi^2$ distribution, the detectability of incipient faults is analyzed in a statistical sense, and the window length and weight are derived, which ensures that the false alarm rate and missed detection rate can be reduced below the tolerance level. Finally, an experiment is performed in the rotary steerable drilling tool prototype to verify the effectiveness of the proposed method.

Incipient detection of stator inter‐turn short‐circuit faults in a Doubly‐Fed Induction Generator using deep learning

Wind turbines are increasingly expanding worldwide and Doubly-Fed Induction Generator (DFIG) is a key component of most of them. Stator winding fault is a major fault in this equipment and its incipient detection is of vital importance. However, there is a paucity of research in this field. In this study, a novel machine learning-based method is proposed for incipient detection of inter-turn short-circuit fault (ITF) in the DFIG stator based on the current signals of the stator. The proposed method makes use of state-of-the-art deep learning methods along with conventional signal processing tools and general machine learning techniques. More specifically, the incipient fault detection problem is regarded as a multi-class classification problem and a Long Short-Term Memory network, which is more appropriate for time-series data is utilised for inference. Furthermore, a variant of the celebrated Empirical mode Decomposition analysis tool is used to extract some well-known statistical features among which the most informative ones are selected using a new feature selection method. Our tests using experimental data in steady-state conditions show that the proposed method can accurately detect ITF fault at its initial stage when only one turn is shorted. Moreover, its performance is considerably higher than that of a variety of machine learning-based methods.

Automatic incipient fault detection and health state assessment of rolling element bearings using pruned exact linear time method

Rolling element bearing is a crucial component in rotating machines. The existence of faults in the bearing causes sudden failures, resulting in catastrophic failure of machines. Incipient fault detection and health state assessment are the essential tasks in condition-based maintenance to avoid machine failures. This paper proposes a new method, the Pruned Exact Linear Time, for identifying the incipient faults and following damage states in the bearing. When a fault initiates in the bearing, there is an increment in the vibration response. This increment can be quantified by the degradation indicator computed from the vibration signal. The Variational Mode Decomposition technique is used to de-noise the vibration signals. Various statistical features are derived from the de-noised signal, and the best feature subset is chosen by the Recursive Feature Elimination method. Then, the significant bearing life degradation indicator is computed using the Reconstruction Independent Component Analysis method by fusing selected features. A novel index is formulated for computing the percentage of failure, which can identify whether the bearing is in a mild failure state or medium failure state. The efficiency of the proposed framework is demonstrated using experimental bearing datasets.

A complex-valued slow independent component analysis based incipient fault detection and diagnosis method with applications to wastewater treatment processes

Multivariate statistical process monitoring are the essential approaches to achieve better prognostics and health management (PHM) of process industries. However, incipient faults and complex behaviors (such as nonlinearity and dynamics) always render the traditional multivariate statistical process monitoring approaches inadequate. Thus, a complex-valued slow independent component analysis (CSICA) is proposed, which is able to extract optimized features from a complex-valued matrix containing both of raw data and their changing rates by resorting to a complex-valued independent component analysis operation and a batch of phase shifts. These features, named slow independent components (SICs), not only guarantee the statistical independence but also capture slowly-changing patterns, thus refining both dynamic and non-Gaussian information mostly related with incipient faults. The proposed algorithm together with novel statistics, Is2, If2 and SPE, as well as their control limits can sequentially detect incipient faults effectively. Then, together with the novel differential mapping reconstructed contribution plot (DM-RCP) and Granger causality analysis, the proposed method can accurately locate rooting causes of incipient faults. Finally, the proposed framework of process monitoring is validated through two data sets from a simulation platform and an oxidation-ditch-based wastewater treatment plant, respectively. The results demonstrate that the proposed method can achieve more accurate and efficient performances than conventional methods.

Variable speed induction motors’ fault detection based on transient motor current signatures analysis: A review

Induction motor is a major component in the industrial sector. It is experiencing great development concerning size, market share, and technological design. Any sudden failure in this element may lead to great damage. Condition monitoring is a fast emerging technology for the online detection of induction motor incipient faults. It avoids unexpected failure of a critical system, by increasing the life expectancy of the concerning elements while reducing operation and maintenance costs. This paper presents the condition monitoring techniques used with these machines, focusing on the Transient Motor Current Signatures Analysis method that has proven its effectiveness in diagnosing faults of electrical rotating machines, gathering a review on the most important applications that can be used with this technology, and how to process these signals to find out the type and cause of the fault. What distinguishes this paper is that it focuses on applications with variable speeds, so two promising techniques that will be effective in non-stationary signals frequency estimation are presented, which are the Adaptive Notch Filtering method and the Adaptive-Observer approach. Challenges and future goals are also discussed to guide researchers wishing to delve into this field.

Quantitative analysis of incipient fault detectability for time-varying stochastic systems based on weighted moving average approach

In this paper, the problem of incipient fault detection is investigated for linear time-varying (LTV) systems with stochastic noises. The fault detectability in a probabilistic sense is defined for LTV stochastic systems by considering false alarm rate (FAR) and missed detection rate (MDR) simultaneously. Necessary and sufficient conditions are derived to reveal the relationship among the fault amplitude, FAR and MDR, and the reason why incipient faults are difficult to detect is quantitatively analyzed in the model-based framework. To improve the sensibility of the residual to incipient faults, the weighted moving average approach is introduced and its parameters, the optimal weight and the smallest window length, are accurately analyzed in theory. Moreover, the concept of average fault detectability is introduced, which is conducive to providing a feasible scheme for incipient fault detection. Finally, a numerical example and an experiment are given to show the effectiveness of the derived results.