Proposed Fault Diagnosis Scheme Research Articles

The Development of Bi-LSTM Based on Fault Diagnosis Scheme in MVDC System

Diagnosing faults is crucial for ensuring the safety and reliability of medium-voltage direct current (MVDC) systems. In this study, we propose a bidirectional long short-term memory (Bi-LSTM)-based fault diagnosis scheme for the accurate classification of faults occurring in MVDC systems. First, to ensure stability in case a fault occurs, we modeled an MVDC system that included a resistor-based fault current limiter (R-FCL) and a direct current circuit breaker (DCCB). A discrete wavelet transform (DWT) extracted the transient voltages and currents measured using DC lines and AC grids in the frequency–time domain. Based on the digital signal normalized by the DWT, using the measurement data, the Bi-LSTM algorithm was used to classify and learn the types and locations of faults, such as DC line (PTP, P-PTG, and N-PTG) and internal inverter faults. The effectiveness of the proposed fault diagnosis scheme was validated through comparative analysis within the four-terminal MVDC system, demonstrating superior accuracy and a faster diagnosis time compared to those of the existing schemes that utilize other AI algorithms, such as the CNN and LSTM. According to the test results, the proposed fault diagnosis scheme detects MVDC faults and shows a high recognition accuracy of 97.7%. Additionally, when applying the Bi-LSTM-based fault diagnosis scheme, it was confirmed that not only the training diagnosis time (TraDT) but also the average diagnosis time (AvgDT) were 0.03 ms and 0.05 ms faster than LSTM and CNN, respectively. The results validate the superior fault clarification and fast diagnosis performance of the proposed scheme over those of the other methods.

Fault Diagnosis Based on Tensor Computing and Meta-Learning for Smart Grid and Power Communication Network

Fault diagnosis (FD) is a critical challenge for the smart grid and the power communication network, especially when both heterogeneous networks are exponentially becoming enormous and complicated. Consequently, some conventional FD schemes based on labor seem inefficient, even disabled, because they usually cannot efficiently utilize multi-dimensional and heterogeneous big data from both networks. To deal with this challenging technical problem, a novel FD scheme based on tensor computing and meta-learning is proposed for the smart grid and the power communication network. In the proposed scheme, tensor computing is used to process tensor big data from both networks, and a new data fusion scheme is designed to complete and analyze the incomplete and sparse big data. Based on the fused data, a meta-learning approach is used to construct the FD scheme, especially when the target fault samples are inadequate and sparse. In meta-learning, the convolutional neural network is employed as a base learner to generate an FD training model, and the model-agnostic meta-learning algorithm is utilized to fine-tune and further train the pre-trained model. Simulation results and theoretical analysis indicate that the proposed DF scheme based on tensor computing can efficiently process sparse and heterogeneous big data from both networks. Furthermore, the meta-learning-based FD scheme provides an efficient way to diagnose faults with inadequate target samples. The proposed FD scheme based on tensor computing and meta-learning provides a novel solution to detect and analyze the potential faults for smart grid and power communication networks.

Fault Diagnosis of Shipboard Medium‐Voltage Alternating Current Power System with Fault Recording Data‐Driven SE‐ResNet18‐1 Model

The electrification trend of marine vessels has led to the rapid development of all‐electric ships. Yet the unique topology and harsh working environment of shipboard power system have posed significant challenges to its fault diagnosis. This paper addresses the research gap in fault diagnosis for shipboard medium‐voltage alternating current (MVAC) power systems and proposes a machine learning‐based fault diagnosis scheme. Signals from fault recorders before relay triggering are used as input for the neural network model. By combining the residual neural network (ResNet) with Squeeze‐and‐Excitation (SE) Block, the feature extraction capability of the network model is maximized. Utilizing the fault transient information, the proposed scheme can realize real‐time diagnosis of any large disturbance to the power system. The data set for scheme verification is established through simulation experiments on PSCAD/EMTDC platform, while model training is completed under the PyTorch framework. The experimental results show that the proposed fault diagnosis scheme can effectively discriminate all types of faults from non‐fault conditions in shipboard MVAC power system. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

Braking Sensor and Actuator Fault Diagnosis With Combined Model-Based and Data-Driven Pressure Estimation Methods

The braking system is significant for intelligent vehicles, which influences vehicle safety directly. However, under harsh working conditions, the inevitable health degradation and even failure of the braking system may cause severe safety issues. This paper proposes a novel braking actuator and sensor fault diagnosis scheme with combined model-based and data-driven pressure estimation methods. The model-based wheel cylinder pressure (WCP) estimator is established first based on the mathematical model of the hydraulic control unit (HCU) from the perspective of cause. The data-driven WCP estimator is then proposed based on the vehicle dynamics multivariate time series (MTS) model with the gated recurrent unit (GRU) from the perspective of effect. However, acquiring a large dataset is a practical challenge for real vehicle tests, so a novel data augmentation (DA) method called shifting is presented to enhance the model generalization ability. Next, fault detection, isolation, and identification are realized by comparing the threshold and the cumulative sum (CUSUM) of the residuals generated by the combined WCP estimation methods. The validation results show that the proposed data-driven method outperforms the traditional multilayer perceptron (MLP), long short-term memory (LSTM), and Transformer regarding accuracy and generalization. Vehicle tests simulating sensor and actuator faults validate the proposed fault diagnosis scheme.

Empirical Filtering-Based Artificial Intelligence Learning Diagnosis of Series DC Arc Faults in Time Domains

Direct current (DC) networks play a pivotal role in the growing integration of renewable energy sources. However, the occurrence of DC arc faults can introduce disruptions and pose fire hazards within these networks. In order to ensure both safety and optimal functionality, it becomes imperative to comprehend the characteristics of DC arc faults and implement a dependable detection system. This paper introduces an innovative arc fault detection algorithm that leverages current filtering based on the empirical rule in conjunction with intelligent machine learning techniques. The core of this approach involves the sampling and subsequent filtration of current using the empirical rule. This filtering process effectively amplifies the distinctions between normal and arcing states, thereby enhancing the overall performance of the intelligent learning techniques integrated into the system. Furthermore, this proposed diagnosis scheme requires only the signal from the current sensor, which reduces the complexity of the diagnosis scheme. The results obtained from the detection process serve to affirm the effectiveness and reliability of the proposed DC arc fault diagnosis scheme.

Actuator fault estimation using optimization-based learning techniques for linear parameter varying systems with unreliable scheduling parameters

A novel fault diagnosis procedure is proposed in this paper to estimate faults using a linear parameter varying (LPV) model whose scheduling parameters depend on the fault. A wrong determination of the operating conditions could lead the system to an undesired performance or even to an unstable situation, when classical fault diagnosis approaches are applied. This paper addresses this issue by formulating fault diagnosis as a dynamic optimization problem, solved by using a novel hybrid technique that combines a Luenberger-based observer with artificial intelligent (AI) optimization-based algorithms. The observer supervises the health of the system, while AI-based algorithms are able to reconstruct the faulty signal in real-time when the observer determines that the system is under a fault. The efficiency of the proposed fault diagnosis scheme, the three AI-based algorithms based on artificial bee colony and particle swarm optimization, and the gradient-based algorithm developed in this paper, are assessed using a numerical example.

Fault Diagnosis Strategy for a Standalone Photovoltaic System: A Residual Formation Approach

The search for sustainability and green energy, in electricity production, has lead many researchers to study and improve photovoltaic (PV) systems. The PV systems, being a combination of power electronic modules and PV array, have high tendency of faults in sensors, switches, and passive devices. Thus, a reliable fault diagnosis (FD) scheme plays a significant role in protecting PV systems. In this article, a sliding mode observer (SMO)-based FD scheme is presented to figure out the sensor faults in a standalone PV system. The proposed FD scheme makes use of residual formation which in turn helps in detection of faults on the basis of a defined threshold. In addition to the functionality of fault detection, the SMO provides the benefit of reduction in number of sensors required in the PV system. This feature can be utilized as software redundancy in fault-tolerant control (FTC). The test bench, standalone PV system, is equipped with a buck–boost converter for maximum power transfer (MPT) to the connected load. Moreover, the FD scheme is backed by a back-stepping (BS) analogy-based control scheme for extraction of maximum power from the PV panel. The simulations are performed in the MATLAB/Simulink platform under varying environmental conditions (temperature and irradiance) and resistive load. These simulations, subjected to varying operating conditions, confirm the efficacy, in terms of robustness, chattering (oscillations about the reference), and steady-state error, of the proposed scheme.

An Intelligent Fault Diagnosis Scheme for Rotating Machinery Based on Supervised Domain Adaptation With Manifold Embedding

In rotating machinery fault diagnosis, domain adaptation (DA) transfer learning-based framework has been attracting great attentions to tackle the problem of inconsistent feature distribution and insufficient labeled fault feature data. However, most of the existing approaches mainly focus on either the cross-domain distribution alignment or manifold subspace learning, which faces two critical limitations: 1) it is hard to overcome the feature distortions when aligning the distribution in the original feature space and 2) subspace learning is insufficient to decrease the distribution divergence. To address the above limitations, this work proposes an intelligent fault diagnosis scheme based on supervised DA with manifold embedding and key features selection. It first applies maximal overlap discrete wavelet packet transform (MODWPT) to process the vibration signals and performs the statistical feature extraction. In order to ensure that the selected key features are conductive to domain adaption, the fault discriminative ability and domain invariance of the features are investigated based on the domain differences and Laplacian score. Then, it presents a new supervised domain adaption with manifold embedding for the distribution alignment in manifold subspace by taking the class information and neighboring relationships into account. Finally, an intra classifier is learned to predict the unlabeled target domain. The proposed fault diagnosis scheme is evaluated using a set of practical data sets of motor and bearing. The extensive experimental results demonstrate that it significantly outperforms the comparative models and achieves much more effective fault diagnosis under different working conditions.

State Estimation MRAS and Identification of Stator Winding Phase Fault Detection of the PMSG in Wind Energy Based on the Sliding Mode Control

Abstract This paper proposes a method for the diagnosis of stator inter-turn short-circuit fault for permanent magnet synchronous generators (PMSG). Inter-turn short-circuit currents are among the most critical in PMSG. For safety considerations, a fast detection is required when a fault occurs. This approach uses the parameter estimation of the per-phase stator resistance in closed-loop control of variable speed of wind energy conversion system (WECS). In the presence of an incipient short-circuit fault, the estimation of the resistance of the stator in the d-q reference frame does not make it possible to give the exact information. To solve this problem, a novel fault diagnosis scheme is proposed using parameter estimation of the per-phase stator resistance. The per-phase stator resistance of PMSG is estimated using the MRAS algorithm technique in real time. Based on a faulty PMSG model expressed in Park’s reference frame, the number of short-circuited turns is estimated using MRAS. Fault diagnosis is on line detected by analysing the estimated stator resistance of each phase according to the fault condition. The proposed fault diagnosis scheme is implemented without any extra devices. Moreover, the information on the estimated parameters can be used to improve the control performance. The simulation results demonstrate that the proposed method can estimate the faulty phase.

Deep branch attention network and extreme multi-scale entropy based single vibration signal-driven variable speed fault diagnosis scheme for rolling bearing

In view of the difficulty in measuring the speed signal and integrating the vibration and speed information flexibly in actual variable speed bearing fault diagnosis, a single vibration signal-driven variable speed intelligent fault diagnosis scheme for rolling bearings is developed to guarantee the reliability and safety of the equipment in this paper. In the proposed fault diagnosis scheme, the extreme multi-scale entropy (EMSEn) of the raw vibration signal is employed as the alternative characterization parameter of the speed information, and an intelligent diagnosis model named deep branch attention network (DBANet) is developed to integrate the vibration and speed information more flexibly. The developed DBANet contains 2 parallel and relatively independent forward propagation channels, and the attention mechanism is introduced into the deep architecture at branch level to adjust the importance of different branches, which endow the model with the ability of fusing the vibration and speed information autonomously. The effectiveness of the proposed method is verified by experiments, and the experimental results show that, compared with the methods relying on external information fusion, the suggested DBANet can integrate the vibration and speed information more flexibly. Besides, in the case of no speed signal, the proposed diagnosis scheme can achieve more outstanding results compared with the methods of using other multi-scale entropy features as the alternative characterization parameter of the speed information.

On-board Fault Diagnosis of a Laboratory Mini SR-30 Gas Turbine Engine

Inspired by recent progress in machine learning, a data-driven fault diagnosis and isolation (FDI) scheme is explicitly developed for failure in the fuel supply system and sensor measurements of the laboratory gas turbine system. A passive approach to fault diagnosis is implemented where a model is trained using machine learning classifiers to detect a given set of faults in real-time on which it is trained. Towards the end, a comparative study is presented using well-known classification techniques, namely Support Vector Machine, Linear Discriminant Analysis, K-Neighbors, and Decision Trees. Several simulation studies were carried out to demonstrate the proposed fault diagnosis scheme's advantages, capabilities, and performance.

Fault detection and isolation of actuator failures in jet engines using adaptive dynamic programming

This paper presents a adaptive dynamic programming-based fault detection and isolation (FDI) scheme to detect and isolate faults in an aircraft jet engine. To this end, the weights in Actor-Critic neural networks are first tuned to learn the input-output map of the jet engine considering its multiple working modes. The convergences of the trainings in Critic-Actor neural networks are strictly proved without knowing the drift dynamics and the input dynamics in the presence of unknown nonlinearities and approximation errors. Using the residuals that are generated by measuring the difference of each network output and the measured engine output, various criteria are established for accomplishing the fault diagnosis task, that addresses the problem of fault detection and isolation of the system components. A number of simulation studies are carried out for combustion chamber of a single-spool jet engine to demonstrate and illustrate the advantages, capabilities, and performance of our proposed fault diagnosis scheme.

Emerging approaches for nonlinear parameter varying systems

Emerging approaches for nonlinear parameter varying systems

A Robust Fault-Tolerant Control for Quadrotor Helicopters against Sensor Faults and External Disturbances

This paper presents an active fault-tolerant control strategy for quadrotor helicopters to simultaneously accommodate sensor faults and external disturbances. Unlike most of the existing fault diagnosis and fault-tolerant control schemes for quadrotor helicopters, the proposed fault diagnosis scheme is able to estimate sensor faults while eliminating the effect of external disturbances. Moreover, the proposed fault-tolerant control scheme is capable to eliminate the adverse effect of external disturbances as well by designing a disturbance observer to effectively estimate the unknown external disturbances and integrating with the designed integral sliding-mode controller. In this case, the continuous operation of the quadrotor helicopter is ensured while avoiding the unexpected control chattering. In addition, the stability of the closed-loop system is theoretically proved. Finally, the effectiveness and advantages of the proposed scheme are validated and demonstrated through comparative numerical simulations of the quadrotor helicopter under different faulty and uncertain scenarios.

Mechanism model-based and data-driven approach for the diagnosis of solid oxide fuel cell stack leakage

Safety and reliability are key objectives for the efficient operation of solid oxide fuel cell (SOFC) power generation systems. Out of many possible faults, the gas leakage of SOFC stack remains a critical issue that leading to efficiency reduction or even degradation. Therefore, the real-time monitoring and diagnosis of gas leakage in the power generation systems are not only an important premise to improve the efficiency, but also can develop the corresponding fault-tolerant strategy for ensuring the system performance. Motivated by this fact, an on-line fault diagnosis scheme based on mechanism model and data-driven method is proposed to monitor and diagnose the gas leakage of the stack. Firstly, the two-state mechanism model of the SOFC stack is established, which can effectively describe the temperature of the fuel layer and air layer. Then, easily-measured stack inputs and outputs are selected, and a novel gas leakage state estimator combined with unscented Kalman filter (UFK) is developed to reconstruct the leakage state. Furthermore, an adaptive thresholds generator is designed to enhance the robustness of the diagnostic scheme. The performance of the fault diagnosis scheme under different leakage scenarios is evaluated, and the simulation results demonstrate the effectiveness of the proposed scheme. The sudden stack fuel leakage failure that occurred in the stable power generation experiment further illustrates the practicability of the scheme. The proposed fault diagnosis scheme has good practicability and can guide the next step compensates for leakage.

Quadrotor Accelerometer and Gyroscope Sensor Fault Diagnosis Using Nonlinear Adaptive Estimation Methods

This paper presents the design, analysis, and real-time experimental evaluation results of a nonlinear sensor fault diagnosis scheme for quadrotor unmanned air vehicles (UAV). The objective is to detect, isolate, and estimate sensor bias faults in accelerometer and gyroscope measurements. Based on the quadrotor dynamics and sensor models under consideration, the effects of sensor faults are represented as virtual actuator faults in the quadrotor state equation. Two nonlinear diagnostic estimators are designed to provide structured residualsfor fault detection and isolation. Additionally, after the fault is detected and isolated, a nonlinear adaptive estimation scheme is employed for estimating the unknown fault magnitude.The proposed fault diagnosis scheme is capable of handling simultaneous faults in the accelerometer and gyroscope measurements. The effectiveness of the fault diagnosis method is demonstrated using an indoor real-time quadrotor UAV test environment.

Data-Driven Sensor Fault Diagnosis for Fighter under Harsh Conditions

This paper presents a multi-classification fault diagnosis scheme for the sensor of fighter. It is difficult to diagnose sensor faults for fighter under harsh conditions like hyper maneuver because the states change rapidly and the data of fault-free and faulty sensor are similar to each other. The rapidly changing states means that values of sensor faults may be small compared with states variation. Similar data may raise the difficulty to find the separating plane. Wavelet packet decomposition can transform the raw dataset to fault features by computing the energy values of each frequency band. And principal component analysis (PCA) can grasp the main features and reduce computational complexity by transform a group of dependent parameters into a group of independent parameters via orthogonal transformation. A support vector machine (SVM) trains fault features to achieve a multi-classifier via one-versus-rest method. The classifier can obtain fault diagnosis results by inputting features of new data. Numerical simulation results illustrate the effectiveness and accuracy of the proposed fault diagnosis scheme.

Fault Diagnosis Scheme for Cross-Country Faults in Dual-Circuit Line With Emphasis on High-Impedance Fault Syndrome

In a power system network, transmission lines are exposed to atypical fault scenarios. The transmission line is the longest component in a power system and sometimes it passes through forest areas where the occurrence of cross-country faults (CCFs) associated with high-impedance fault (HIF) syndrome is more due to thunderstorms, cyclones, and poor vegetation management/tree trimming. Detection and classification of CCFs which are associated with HIF syndrome is the most challenging task. In this article, maximal overlap discrete wavelet packet transform (MODWPT) has been employed to extract the characteristics of the signals during CCF with HIF syndrome which are more complex and aperiodic/asymmetric/nonlinear. Thereafter, maximum change in wavelet packet energy of MODWPT coefficients of currents/voltages have been deliberated as unique features to design proposed fault diagnosis scheme for a dual-circuit transmission line. The efficacy of the proposed scheme has been investigated on the Chhattisgarh State Power Transmission Network. The proposed scheme is appraised with exclusive case studies such as switching of capacitor bank, switching of reactor string, switching of load/feeder, effect of power swing, nonlinear loads, lightly loading conditions, and measurement noise. The proposed scheme has been also validated on standard IEEE 13 bus distribution system with nonlinear loads to confirm its adaptability.

Fault Diagnosis of Railway Point Machines Using the Locally Connected Autoencoder

Data-driven fault diagnosis is considered a modern technique in Industry 4.0. In the area of urban rail transit, researchers focus on the fault diagnosis of railway point machines as failures of the point machine may cause serious accidents, such as the derailment of a train, leading to significant personnel and property loss. This paper presents a novel data-driven fault diagnosis scheme for railway point machines using current signals. Different from any handcrafted feature extraction approach, the proposed scheme employs a locally connected autoencoder to automatically capture high-order features. To enhance the temporal characteristic, the current signals are segmented and blended into some subsequences. These subsequences are then fed to the proposed autoencoder. With the help of a weighting strategy, the seized features are weight averaged into a final representation. At last, different from the existing classification methods, we employ the local outlier factor algorithm to solve the fault diagnosis problem without any training steps, as the accurate data labels that indicate a healthy or unhealthy state are difficult to acquire. To verify the effectiveness of the proposed fault diagnosis scheme, a fault dataset termed “Cu-3300” is created by collecting 3300 in-field current signals. Using Cu-3300, we perform comprehensive analysis to demonstrate that the proposed scheme outperforms the existing methods. We have made the dataset Cu-3300 and the code file freely accessible as open source files. To the best of our knowledge, the dataset Cu-3300 is the first open source dataset in the area of railway point machines and our conducted research is the first to investigate the use of autoencoders for fault diagnosis of point machines.

Diagnosis of Open-Switch and Current Sensor Faults in PMSM Drives Through Stator Current Analysis

Fault detection and fault accommodation, as well as fault tolerance, are important issues in closed-loop controlled three-phase converters. For such systems, power semiconductors and sensors are the most affected components by the occurrence of failures. Therefore, fault diagnosis and remedial operating strategies involving such components are highly required. In this paper, a new reliable approach is proposed, which aims to diagnose current sensor faults and power semiconductor open-circuit faults in permanent magnet synchronous motor drives. The proposed fault diagnosis scheme uses only the measured currents, which makes it suitable for real-time implementation. The performance of the proposed technique is analyzed through experiments.