Online Fault Diagnosis Method Research Articles

Design of On-line Monitoring and Fault Diagnosis Platform for Oil-immersed Shunt Reactor

Abstract With the development of society and technology, the demand for power supply and reliability is increasing. In the power system, the reactor plays an important role as a key equipment. However, due to long-term operation, the reactor is affected by factors such as loss, aging and hidden faults, and its reliability gradually decreases. The operation state of the reactor is directly related to the operation effect of the power grid and the quality of the user’s electricity. Therefore, the application of reactor online monitoring and fault diagnosis in real life is very important. This paper considers the actual operation of the coastal power grid. Based on Qt software development platform, C + + language and MySQL database, an online monitoring and fault diagnosis platform for oil-immersed shunt reactors is developed. The application of the platform takes the online monitoring and fault diagnosis method of dissolved gas in oil as the core. The platform realizes the basic information management, online monitoring, fault diagnosis and suggestions for operation and maintenance of the reactor.

An online fault detection and diagnosis method of sensors in district heating substations based on long short-term memory network and adaptive threshold selection algorithm

District heating system is an essential energy infrastructure. Due to the harsh working environment, sensors in district heating substations are prone to faults, which will mislead the control strategies of substations. We propose an online fault detection and diagnosis (FDD) method for sensors in substations. The proposed method establishes a Sequence-to-Sequence Prediction Model based on Long Short-Term Memory network with Spatial-Temporal Attention (abbreviated as STALSTM-seq2seq model) for each target sensor. The model can learn normal and evolving data patterns from historical time series of sensors and predict current sensor observation. Then, an adaptive threshold selection algorithm is used to autonomously determine current anomaly threshold based on the prediction error vector of the model. If the prediction error exceeds the anomaly threshold, the sensor malfunctions. Otherwise, the sensor operates normally. Through a comprehensive case study, we demonstrate that the adaptive threshold selection algorithm can fully utilize the forecasting ability of the STALSTM-seq2seq model, and thereby ensure the accuracy of sensor FDD. Overall, the online sensor FDD method achieves an average F1 score of 0.8473. The research shows that the proposed method provides an effective and robust solution to timely and accurately identify sensor faults for district heating substations.

An Online Multiple Open-Switch Fault Diagnosis Method for T-Type Three-Level Inverters Based on Multimodal Deep Residual Filter Network

With the development of the processing capacity of the embedded chip, it is possible to implement a machine learning algorithm in the embedded system. To achieve the fault status without poor portability, tricky threshold selection and complex rulemaking, this paper proposes a multi-modal deep residual filter network for online multiple open-switch fault diagnosis of T-type three-level inverter. It contains low-rank matrix fusion (LMF), deep residual filter network (DRFN), and cross transformer mechanism. The LMF fuses the voltage signal and the current signal for obtaining the unified representation. And then, the DRFN filters noise adaptively and extracts information effectively. Finally, the cross-transformer mechanism output the fault state of the T-type three-level inverter. The data sets consist of the dc-link voltage and load side current of the inverter control system. The data time window is selected as 20 ms. Through the real-time calculation of online monitored data, the experimental results show the effectiveness of the proposed fault diagnosis approach. Moreover, the accuracy of fault diagnosis obtains 99.18% and the average open-circuit fault diagnosis time is 21 ms.

Online Fault Diagnosis Method for Grid-Tied T-Type Inverters Based on Specific Frequency Calculation

Online Fault Diagnosis Method for Grid-Tied T-Type Inverters Based on Specific Frequency Calculation

A multi-stream multi-scale lightweight SwinMLP network with an adaptive channel-spatial soft threshold for online fault diagnosis of power transformers

Fault diagnosis of power equipment is extremely crucial to the stability of power grid systems. However, complex operating environments, high costs and limitations of single-modal signals are the biggest bottlenecks. To this end,a multi-tream, multi-scale lightweight Swin multilayer perceptron (MLP) network (MLSNet) with an adaptive channel-spatial soft threshold is proposed in this paper. First, a Res2net-based feature-enhanced method is used to learn the correlated features of vibration and voltage multi-modal signals. Second, a novel MLSNet is designed to combine the benefits of Swin transformers with an MLP with a lightweight convolutional neural network and employs a staged model to extract various scale features. Third, an adaptive deep fusion approach employing a channel-spatial soft threshold module is used to integrate and recalibrate staged information at different scales. The overall accuracy of the proposed method can reach 98.73% in various experiments, potentially making it an effective method for online fault diagnosis of power transformers.

A new belief rule base based distributed online fault diagnosis method for multi‐agent systems

AbstractIn this paper, based on belief rule base (BRB), we studied the distributed online fault diagnosis problem of multi‐agent systems (MASs). To begin with, considering the distributed nature of MASs, by using the relative information between an agent and its neighbors, a distributed online fault diagnosis model is constructed. Next, for the results of fault diagnosis, when an agent doesn't get enough relative information to diagnose the fault, the fault diagnosis result may be inaccurate, and if there exist too many neighbors of an agent, the number of attributes of the proposed BRB fault diagnosis model will be too large, which will lead to large computation. To improve the performance of the proposed distributed online fault diagnosis model, a real‐time communication topology switching strategy was proposed. Under the constructed distributed fault diagnosis model, the real‐time performance and the accuracy of fault diagnosis for agents can be guaranteed simultaneously. Finally, a simulation is given to validate the effectiveness and advancement of the distributed online fault diagnosis method proposed in this paper.

An Online Open-Switch Fault Diagnosis Method for the DAB Converter Based on Multimodal Fusion Neural Controlled Differential Equations

Machine learning approaches are available for the fault diagnosis issues of the dual active bridge (DAB) converter. However, influenced by noise and multimodal diagnostic signals, the data-driven methods are still in the initial phase for certain fault diagnoses of the DAB converter. To fill this gap, a multimodal fusion neural controlled differential equation (MFNCDE) is presented to detect and locate the DAB converter’s fault diagnosis in early phases. An improved low-rank matrix fusion (LMF) method is employed to fuse the multimodal signals (voltage and current signals) of the DAB converter. The neural controlled differential equations (NCDEs) with sparse transformer (ST) extract significant information from the diagnostic signal, based on which the fault classifier obtains the fault state of the DAB converter. The datasets are collected from a real DAB converter under single-phase shift control. Contrastive experiment results demonstrate that the presented method is superior to other fault diagnosis methods.

Online Diagnosis and Ride-Through Operation for Cascaded H-Bridge Converter Based STATCOM With a Single Open-Circuit IGBT

To address the most commonly seen single insulated gate bipolar transistor (IGBT) open-circuit fault problems in cascaded H-bridge (CHB) converter based static synchronous compensator, a simple online fault diagnosis method and the corresponding ride-through operation approach are proposed. The method can quickly determine the exact location of the open-circuit IGBT through three steps, via analyzing the relationships between the dc bias in the three-phase output current of CHB converter, the transient dc rail voltage variations with respect to the averaged dc rail voltage, and the response of dc rail voltage and output current when the modulation approach of the faulty power module is adjusted. Afterwards, a fault ride-through approach is adopted to seamlessly transfer the modulation approach of faulty H-bridge module and it behaves as a half-bridge to provide a half of the rated output ac voltage. Meanwhile, the corresponding dc offset voltage from this faulty module is compensated by producing the same dc voltage in the healthy phase legs to avoid the dc bias in the three-phase output current. It has been demonstrated that the proposed fault diagnosis approach and the fault ride-through approach are effective in various adverse situations including grid voltage sudden disturbances and the load power sudden changes. Comprehensive simulated and experimental results have been provided to validate the effectiveness of the proposed approach.

Online Fault Diagnosis Method for Grid-Connected Inverters Based on Finite-Set Mixed Logical Dynamical Model Prediction

The grid-connected inverter is a key device in the renewable energy power generation system and large-scale energy storage system, which the operational stability and reliability are the basis for the efficient and safe application of electrical energy. A real-time fault diagnosis method of a three-phase for grid-connected application combining a mixed logic dynamic (MLD) model and finite control set model predictive control (FCS-MPC) is proposed. This paper not only realizes the open circuit fault diagnosis and location of the switching devices in the main power circuit, but also discusses the threshold issues and post-fault operations. The advantage of the proposed method is that it directly uses the control data and measurement signals of the controller without extra sensors and calculation, which will shorter the fault diagnosis time and occupy less calculation resource of the main processor. Simulation results illustrate the quickness of the fault identification and accurate position with robustness to the interference of the diagnosis method. Finally, the effectiveness of the diagnosis method was verified by a 1500W experimental prototype in a laboratory.

Virtual Sensing for Online Fault Diagnosis of Heat Exchangers

Heat exchangers play essential roles in power generation and petrochemical industry. Although physical sensors are widely deployed in the process industry, timely fault detection and diagnosis method remain a significant challenge to its safe and reliable operation for accurate and reliable equipment condition monitoring. To address this challenge, this paper proposes a new virtual sensing method for online fault diagnosis method of heat exchangers. The virtual sensing models are constructed by incorporating the equipment failure mechanism and inference analytics. Complemented by high-precision in-process data measurements, fouling thickness and tube leakage in heat exchangers are quantified in real time using virtual sensing models. Therefore, the proposed virtual sensing method based on the fusion of physical mechanisms and measured data enables the online fault diagnosis of heat exchangers with only limited operating parameters while keeping the heat exchanger operational. The effectiveness of the proposed virtual sensing models is experimentally validated using a set of run-to-failure tests on real heat exchangers.

A Machine Learning and Internet of Things-Based Online Fault Diagnosis Method for Photovoltaic Arrays

In this paper, a novel fault detection and classification method for photovoltaic (PV) arrays is introduced. The method has been developed using a dataset of voltage and current measurements (I–V curves) which were collected from a small-scale PV system at the RELab, the University of Jijel (Algeria). Two different machine learning-based algorithms have been used in order to detect and classify the faults. An Internet of Things-based application has been used in order to send data to the cloud, while the machine learning codes have been run on a Raspberry Pi 4. A webpage which shows the results and informs the user about the state of the PV array has also been developed. The results show the ability and the feasibility of the developed method, which detects and classifies a number of faults and anomalies (e.g., the accumulation of dust on the PV module surface, permanent shading, the disconnection of a PV module, and the presence of a short-circuited bypass diode in a PV module) with a pretty good accuracy (98% for detection and 96% classification).

SF 6 decomposition and insulation condition monitoring of GIE: A review

The statistics of the Conference International des Grands Reseaux Electriques (CIGRE) indicate that the operational reliability of SF6 gas-insulated equipment (GIE) is very high; however, the failure rate of the GIE in operation is much higher than that of the IEC standard, and the fault occurs frequently in the GIE at a high voltage level. The reason is due to the complex and strong on-site electromagnetic interference environment and fully enclosed structure of GIE. The key method and technology for effective on-line monitoring and fault diagnosis of GIE are still lacking. Given the partial strong electromagnetic energy and high temperature induced by early latent insulation faults in the equipment, SF6 gas insulation presents different degrees of decomposition. The decomposition products mainly include SO2F2, SOF2, SO2, HF, and H2S. The decomposition characteristics of SF6 are closely related to the property of insulation faults. At present, this area is attracting attention from the power industry and research institutes. This study summarises the current research on SF6 decomposition component analysis (DCA). The content mainly includes the latest progress of SF6 decomposition characteristics and mechanism under fault conditions, and fault diagnosis methods based on decomposition components.

On-line fault judgment of integrated capacitive coupler based on signal line port impedance

Medium voltage carrier communication is a communication mode which uses 10kV medium voltage distribution line as transmission channel, and has been widely used in the field of distribution automation and power consumption information collection. Integrated capacitive coupler is the guarantee of stable and reliable operation of medium voltage carrier communication system. However, limited by voltage level, there is no effective method to judge whether the coupler fails when the 10kV distribution line is not powered off. In this paper, an on-line fault diagnosis method of integrated capacitive coupler based on signal line port impedance is proposed, which is beneficial for the improved medium voltage carrier communication machine to collect the signal line port impedance, and can judge whether the integrated capacitive coupler is faulty online under the condition of 10kV distribution line without power interruption. The method is simple and reliable, saving manpower and material resources, and having certain practical engineering value.

Research on the Fault Diagnosis Method of Mine Fan Based on Sound Signal Analysis

The underground local fan and auxiliary fan also play a vital role in the underground air quality, compared with the system fan. However, the number of underground local fans and auxiliary fans is large and widely distributed, which is disadvantageous to adopt the same method of online monitoring and fault diagnosis method as the system fan. In order to find a new fault diagnosis method, which is cost-effective and reliable, this paper proposes a fault diagnosis method based on sound signal. It analyzes the source of fan noise and studies the overall scheme of mine fan fault diagnosis expert system based on sound signal. The fault expert system consists of four parts: signal acquisition and noise elimination, feature extraction, state recognition, and fault diagnosis. Its principle is briefly introduced. The denoising method of wavelet is adopted in this paper. Wavelet packet is used to extract the characteristics of sound signal, and the energy size and energy proportion of each frequency component are used as the basis of knowledge acquisition and reasoning. Through the analysis of the measured signals of the fan in the normal operating state, the feature vectors were extracted as the basis for the discrimination of the normal state after noise elimination. At the same time, the audio processing software was used to simulate the sound signals in three fault states. Then, the feature vector of the fault state is extracted, which is obviously different from that of the fan in the normal operation. As the basis of fault state analysis of the expert system, it lays the foundation for the realization of the expert system of mine fan equipment running state diagnosis.

Effective fault distance estimation and diagnosis in the PV based DC‐ring network using Hilbert Huang transform and weighted online sequential random vector functional network

This paper proposes a novel approach for fault detection, classification, and distance estimation of five kinds of faults (ie, cable and Arc faults) in the PV-based DC-ring microgrid. The proposed approach is based on the Hilbert Huang Transform (HHT) and online sequential random vector functional link network (OS-RVFLN). Initially, differential current signals are passed through empirical mode decomposition (EMD) to yield intrinsic mode functions (IMFs). Among the series of IMFs, more efficient IMFs are selected based on the threshold of maximum weighted kurtosis index (M-WKI) value for fault diagnosis. The selected IMFs are passed through Hilbert transform (HT) for determining the instantaneous frequency, magnitude spectrum, and Hilbert energy. Then, the Teager energy operator (TEO) is applied to the more effective IMF among the selected IMFs of each DC fault for fault detection. The most suitable target features are extracted from the selected efficient IMFs. This feature set data is initialized to the proposed HHT-OSRVFLN algorithm for five kinds of faults classification and its location estimation. The performance of the proposed algorithm in the proposed microgrid is proved in terms of low fault detection time, superior classification accuracy, and low relative error through different case studies by consuming minimum computation time in MATLAB/Simulink environment. Finally, the digital architecture of the proposed method is designed and simulated in Xilinx ISE 14.5 software environment to estimate the fault location. The Chip scope hardware debugging output proves the correctness, simplicity, feasibility, practicability, and robustness of the proposed method for online fault diagnosis.

An Online Data-Driven Fault Diagnosis Method for Air Handling Units by Rule and Convolutional Neural Networks.

The stable operation of air handling units (AHU) is critical to ensure high efficiency and to extend the lifetime of the heating, ventilation, and air conditioning (HVAC) systems of buildings. In this paper, an online data-driven diagnosis method for AHU in an HVAC system is proposed and elaborated. The rule-based method can roughly detect the sensor condition by setting threshold values according to prior experience. Then, an efficient feature selection method using 1D convolutional neural networks (CNNs) is proposed for fault diagnosis of AHU in HVAC systems according to the system’s historical data obtained from the building management system. The new framework combines the rule-based method and CNNs-based method (RACNN) for sensor fault and complicated fault. The fault type of AHU can be accurately identified via the offline test results with an accuracy of 99.15% and fast online detection within 2 min. In the lab, the proposed RACNN method was validated on a real AHU system. The experimental results show that the proposed RACNN improves the performance of fault diagnosis.

Convolutional Bi-directional Long Short Term Memory Network based Dynamic Fault Diagnosis for Transformer DGA

Dissolved gas analysis (DGA) method is one of the important methods to detect early internal faults in transformers. Aiming at the shortcomings of feature extraction and online modelling in the analysis model based on shallow layer, a transformer DGA online fault diagnosis method combining convolutional neural network and bidirectional long short term memory network is proposed in this paper. By using multi-time characteristic gas data as the input of the diagnostic model, combined with convolution and pooling, the fault sensitive features were extracted, and the time dimension features were extracted by using the recursive neural network with gated structure. Under various evaluation criteria, it is compared with convolutional neural network, bidirectional long short-term memory (Bi-LSTM) network and dynamic support vector machine. The experimental results show that the online transformer diagnosis model proposed in this paper, which considers both time and space characteristics, has higher prediction accuracy.

Data-driven fault diagnosis method for the safe and stable operation of solid oxide fuel cells system

Solid oxide fuel cell system is complex with multiple variables strongly coupled. Once a fault occurs, if it cannot be found in time, the initial minor fault may slowly evolve and spread to subsequent components. Therefore, fault diagnosis is a promising approach to guarantee the stability of the system. In this paper, the impact of air leakage and fuel starvation is investigated. To diagnose the two types of faults, a novel data-driven online fault diagnosis method based on principal component analysis and support vector machine is developed. Data comes from the entire stage of the solid oxide fuel cell system experiment. The results show that the proposed method can effectively identify the air leakage and fuel starvation fault in real time. Through comparison with traditional machine learning methods, this method shows higher accuracy and better generalization performance. Moreover, it combines prior knowledge and statistical characteristics to extract effective features, thereby reducing the calculation burden. Furthermore, with proper modifications, the proposed method can be extended to other types of solid oxide fuel cell system faults, which is significant in enhancing the reliability of the system.

Photovoltaic Fault Diagnosis Via Semisupervised Ladder Network With String Voltage and Current Measures

In recent years, many supervised learning algorithms have been successfully applied for photovoltaic (PV) fault diagnosis. In practice, it is not possible to effectively obtain labels of large samples, limiting the engineering application of these algorithms. As for the unsupervised learning algorithm, it is completely adaptive learning, requiring a large number of samples to better learn the potential features in the data. To address the above problems, an improved online fault diagnosis method is proposed, which uses a small number of labeled samples to train the semisupervised ladder network (SSLN) fault diagnosis model to realize the diagnosis of line-to-line faults, open-circuit faults, partial shadow faults, and hybrid faults. In the proposed method, only the real-time operating voltage and current of PV array are needed for fault diagnosis. The sequential voltage and current of the PV array are first normalized, and the sequential power waveforms are obtained through numerical calculation. Then, the SSLN is used to extract the fault features from the sequence power waveforms. Finally, the classification is realized using the SSLN's noiseless encoder. To eliminate overfitting and improve convergence, the activation function, optimizer, and loss function of the SSLN is studied and improved. Meanwhile, numerical simulations and measured data verify that the proposed method provides strong anti-interference, and the diagnostic accuracies of both exceed 98%. Comparative experiments show that the proposed method outperforms algorithms such as squared-loss mutual information regularization, semisupervised support vector machine, graph-based semisupervised learning, and semisupervised extreme learning machine.

Data‐driven design of fault diagnosis for three‐phase PWM rectifier using random forests technique with transient synthetic features

A three-phase pulse-width modulation (PWM) rectifier can usually maintain operation when open-circuit faults occur in insulated-gate bipolar transistors (IGBTs), which will lead the system to be unstable and unsafe. Aiming at this problem, based on random forests with transient synthetic features, a data-driven online fault diagnosis method is proposed to locate the open-circuit faults of IGBTs timely and effectively in this study. Firstly, by analysing the open-circuit fault features of IGBTs in the three-phase PWM rectifier, it is found that the occurrence of the fault features is related to the fault location and time, and the fault features do not always appear immediately with the occurrence of the fault. Secondly, different data-driven fault diagnosis methods are compared and evaluated, the performance of random forests algorithm is better than that of support vector machine or artificial neural networks. Meanwhile, the accuracy of fault diagnosis classifier trained by transient synthetic features is higher than that trained by original features. Also, the random forests fault diagnosis classifier trained by multiplicative features is the best with fault diagnosis accuracy can reach 98.32%. Finally, the online fault diagnosis experiments are carried out and the results demonstrate the effectiveness of the proposed method, which can accurately locate the open-circuit faults in IGBTs while ensuring system safety.