Condition Fault Detection Research Articles

Trend‐Based Predictive Maintenance and Fault Detection Analytics for Photovoltaic Power Plants

Optimized predictive maintenance in photovoltaic (PV) systems is crucial for ensuring prolonged operational performance and cost‐effective operation and maintenance (O&M). Even though failure detection methods have already been developed, the main challenge remains the lack of predictive maintenance strategies to accurately forecast underperformance conditions. The scope of this work is to develop a predictive maintenance and failure detection routine for assessing the health status of PV systems. The workflow consists of the eXtreme gradient boosting algorithm for modeling the PV performance, the one‐class support vector machine algorithm for fault detection, and the Facebook Prophet algorithm for forecasting PV performance trends and generating maintenance alerts. The developed data‐driven routine analyzes performance trend deviations and it is validated using a historical dataset from a utility‐scale PV power plant in Greece. The obtained results show the effectiveness of the developed workflow in detecting fault conditions, achieving a sensitivity of 96.9%. Additionally, the results demonstrate the workflow's ability to generate predictive maintenance alerts up to 7 days in advance, yielding a sensitivity of 92.9%. Finally, the study provides useful insights that enhance operators’ efficiency in conducting O&M activities.

Ensemble LVQ Model for Photovoltaic Line-to-Line Fault Diagnosis Using K-Means Clustering and AdaGrad

Line-to-line (LL) faults are one of the most frequent short-circuit conditions in photovoltaic (PV) arrays which are conventionally detected and cleared by overcurrent protection devices (OCPDs). However, OCPDs are shown to face challenges when detecting LL faults under critical detection conditions, i.e., low mismatch levels and/or high fault impedance values. This occurs due to insufficient fault current, thus leaving the LL faults undetected and leading to power losses and even catastrophic fire hazards. To compensate for OCPD deficiencies, recent studies have proposed modern artificial intelligence (AI)-based methods. However, various limitations can still be witnessed even in AI-based methods, such as (i) most of the models requiring a massive training dataset, (ii) critical fault detection conditions not being taken into consideration, (iii) models not being accurate enough when dealing with critical conditions, etc. To this end, the present paper proposes a learning vector quantization (LVQ)-based ensemble learning model in which three LVQs are individually trained to detect and classify LL faults in PV arrays. The initial LVQ vectors are determined using the k-means clustering method, and the learning rate is optimized by the adaptive gradient (AdaGrad) optimizer. The training and testing datasets are collected according to the PV array’s current–voltage (I–V) characteristic curve, and several features are extracted based on the Canberra and chi-squared distance techniques. The model utilizes a small training dataset, considers various critical detection conditions for LL faults—such as different mismatch levels and fault impedance values—and the final experimental results show that the model achieves an impressive average accuracy of 99.26%.

A novel fault detection and identification method for complex chemical processes based on OSCAE and CNN

The safety and reliability of chemical processes are of paramount importance. However, due to the complicated heat, mass, and momentum transfer processes coupled with chemical reactions, and the interrelated effects among various equipment units, it is difficult to accurately detect and identify faults occurring in the chemical system. This study proposes an innovative fault diagnosis framework specifically designed for chemical systems. A method based on orthogonal self-attention convolutional autoencoder (OSCAE) is constructed for fault detection. The convolutional autoencoder is combined with an orthogonal attention mechanism to extract time-series characteristic information of the operational state, thereby achieving precise detection of fault conditions. Additionally, a convolutional neural network (CNN) model uses both raw signals and data reconstructed by OSCAE to identify different faults, which enhances the features of the input and enables highly accurate fault identification. The effectiveness and robustness of the propose method is validated with simulation data of our self-established chemical distillation system and the publicly available Tennessee Eastman process system. The diagnosis efficacy in fault detection and identification is validated by applying fault detection metrics, high-dimensional feature visualizations, and confusion matrix analyses across two case studies. This paper not only provides an effective diagnostic framework for chemical systems but also offers a usable benchmark dataset for chemical distillation systems to the academic community, with the hope that the research content can enhance the stability and reliability of chemical processes.

Integrating Machine Learning and Feature Extraction for Islanding Detection in Grid-Connected Photovoltaic Systems: A Hybrid Intelligent Approacha

Introduction: In recent years, the integration of renewable energy sources, particularly Photovoltaic (PV) systems, into the grid has garnered considerable attention. However, the distributed nature of these grid-integrated PV systems has introduced challenges concerning grid faults and maintenance. Method: This paper aims to present a pioneering approach to augment the monitoring of gridintegrated PV systems by integrating intelligent methods, specifically machine learning and feature extraction techniques. The primary focus of this approach is on islanding detection, which involves promptly identifying grid faults or maintenance challenges and initiating the grid's transition to an isolated mode of operation. To accomplish this, an intelligent signaling method is employed, capitalizing on the capabilities of Distributed Generation (DG) networks. By recording critical signals such as voltage, current, and frequency at common coupling points, fault conditions can be accurately detected. Signals obtained from the grid are subjected to wavelet transformation to extract pertinent information that characterizes fault conditions. These extracted features are then utilized as inputs to machine learning methods, facilitating the proposal of intelligent islanding scenarios. To assess the efficacy of the proposed approach, simulations are conducted on a grid-connected PV system. The recorded signals are meticulously analyzed, and the extracted features are employed to train machine learning models. The performance of these models is evaluated based on their ability to detect fault conditions and initiate appropriate islanding scenarios accurately. Result: The results obtained demonstrate the immense potential of the proposed approach in bolstering the monitoring of grid-integrated PV systems. Conclusion: By synergizing machine learning techniques with feature extraction and intelligent signaling methods with the KNN-Confusion Matrix, all predicted labels match the true labels, resulting in a 100% accuracy. The detection of grid faults and maintenance challenges can be substantially improved, thereby fostering more efficient and dependable operation of these systems.

Fault Detection and Normal Operating Condition in Power Transformers via Pattern Recognition Artificial Neural Network

Aging, degradation, or damage to internal insulation materials often contribute to transformer failures. Furthermore, combustible gases can be produced when these insulation materials experience thermal or electrical stresses. This paper presents an artificial neural network for pattern recognition (PRN) to classify the operating conditions of power transformers (normal, thermal faults, and electrical faults) depending on the combustible gases present in them. Two network configurations were presented, one with five and the other with ten neurons in the hidden layer. The main advantage of applying this model through artificial neural networks is its ability to capture the nonlinear characteristics of the samples under study, thus avoiding the need for iterative procedures. The effectiveness and applicability of the proposed methodology were evaluated on 815 real data samples. Based on the results, the PRN performed well in both training and validation (for samples that were not part of the training), with a mean squared error (MSE) close to expected (0.001). The network was able to classify the samples with a 98% accuracy rate of the 815 samples presented and with 100% accuracy in validation, showing that the methodology developed is capable of acting as a tool for diagnosing the operability of power transformers.

Experimental study of a teeth flux sensor for detection, location and severity evaluation of induction machine stator faults

PurposeThis paper aims to investigate the efficacy of teeth flux sensors in detecting, locating and assessing the severity of short-circuit faults in the stator windings of induction machines.Design/methodology/approachThe experimental study involves inducing short-circuit winding turn variations on the induction machine’s stator and continuously measuring the RMS values across teeth flux sensors. Two crucial steps are taken for machine diagnosis: measurements under load operating conditions for fault detection and measurements under no-load conditions to determine fault location and severity.FindingsThe experimental results demonstrate that the proposed approach using teeth flux sensors is reliable and effective in detecting, locating and evaluating the severity of stator winding faults.Research limitations/implicationsWhile this study focuses on short-circuit faults, future research could explore other fault types and alternative sensor configurations to enhance the comprehensiveness of fault diagnosis.Practical implicationsThe methodology outlined in this paper holds the potential to significantly reduce maintenance time and costs for induction machines, leading to substantial savings for companies.Originality/valueThis research contributes to the field by presenting an innovative approach that uses teeth flux sensors for a comprehensive fault diagnosis in induction machines. The originality lies in the effectiveness of this approach in providing reliable fault detection, location and severity evaluation.

Comparative analysis of harmonic sensitivity for stator fault diagnosis in induction motors

Induction motors play a vital role in various industrial applications due to their commendable efficiency and reliability. However, their susceptibility to faults, especially in challenging industrial environments, highlights the need for vigilant fault detection to prevent unforeseen downtimes and reduce subsequent repair costs. Early fault diagnosis is crucial in this context. A systematic approach to diagnosing faults in asynchronous motors involves the application of signal processing techniques, particularly utilizing the Fast Fourier Transform (FFT). The FFT, as a mathematical tool, enables a comprehensive analysis of signals, facilitating the identification and isolation of their frequency components. Monitoring the frequency components within a motor's signal provides a means to determine the existence and severity of faults. FFT analysis allows for the monitoring of four distinct categories of harmonics: time harmonics (TH), rotor slot harmonics (RSH), rotor bar fault harmonics (RBFH), and eccentricity fault harmonics (EFH). Each type of harmonic offers valuable insights into specific fault categories. Empirical evidence, drawn from experimental results, emphasizes the heightened sensitivity of rotor slot harmonics (RSH) in detecting stator faults. Continuous monitoring of the RSH frequency component enables the prompt detection and localization of stator faults, along with an assessment of their severity. Additionally, this diagnostic methodology proves effective in identifying micro short-circuits between stator coils, allowing for a proactive strategy in predictive maintenance. This proactive approach enables anticipatory part replacement before degradation progresses to the point of causing comprehensive failure in the production chain. The combination of FFT-based signal processing and harmonic analysis establishes a robust framework for the early detection and localization of faults in asynchronous motors within industrial settings. This contributes to enhanced operational reliability and efficiency, ultimately ensuring smoother industrial processes.

Advanced Residual Optimal Mapping Approach for Precise Detection of Stator Faults in Induction Motors

Advanced Residual Optimal Mapping Approach for Precise Detection of Stator Faults in Induction Motors

Adaptive Substation Infrastructure for Bangladesh: Harnessing Artificial Intelligence and Machine Learning for Enhanced Performance and Grid Resilience

This groundbreaking research endeavors to revolutionize substation engineering in Bangladesh by introducing an innovative paradigm that integrates artificial intelligence (AI) and machine learning (ML) methodologies. In response to the dynamic and challenging operational environment, this study focuses on the development of an adaptive substation infrastructure capable of intelligently responding to fluctuating energy demands, environmental stresses, and emerging grid complexities. Through the application of advanced AI algorithms, the research addresses real-time fault detection, predictive maintenance, and comprehensive condition monitoring within the substation framework. Harnessing the capabilities of ML models, the proposed infrastructure aims to optimize energy flow, enhance grid resilience, and mitigate potential failures by autonomously adapting to evolving operational scenarios. By combining cutting-edge technology with the unique challenges of the Bangladeshi power landscape, this research not only aims to advance the field of substation engineering but also holds the promise of significantly contributing to the sustainable development of the nation's power infrastructure. The findings are anticipated to guide the design and implementation of intelligent substation systems, ushering in a new era of efficiency, reliability, and adaptability in the Bangladesh power grid.

Fault detection of fractional‐order system based on finite frequency H−/H∞${{H}_{-}}/{{H}_{\infty }}$ unknown input observer

AbstractThis study is about fault detection observer design in finite frequency domain for fractional‐order systems with unknown input disturbances. In order to make the generated residuals fault‐sensitive, robust to the non‐decoupling parts of unknown input disturbances, and guarantee the stability requirements of fractional‐order error systems, the authors derive the fault detection conditions of unknown input observers. First, using the condition of observer, the unknown input is decomposed into decoupling part and non‐decoupling part, and the decoupling part is eliminated from the estimation error. Second, the residuals generator of the unknown input observer is designed by using the finite frequency performance indicator, so that the residuals are robust to the non‐decoupling part of the unknown input and sensitive to fault signals. In particular, variable linearization methods are introduced to convert design conditions into a series of LMIs. Finally, the simulation results show the effectiveness of the designed method.

Deterministic learning-based sensor fault detection, isolation and accommodation of nonlinear systems

This paper presents a systematic deterministic learning-based rapid sensor fault detection, isolation and accommodation (SFDIA) scheme for a class of nonlinear systems with unmodeled output dynamics. First, locally accurate approximations of sensor faults and unmodeled output dynamics are achieved by using a deterministic learning-based neural network observer. Second, a sensor fault pattern bank which stores the knowledge of sensor faults and unmodeled output dynamics is established. A set of observers utilising the learned knowledge are constructed to generate residuals for rapid SFDIA. The contributions are: (1) the fault detectability and isolability conditions are rigorously analysed and verified in the simulation; (2) using the proposed fault accommodation approach, compared with online parameter adaptation techniques, better tracking performance can be achieved in the presence of sensor faults. Simulation results show the effectiveness of the proposed method.

Fast fault detection for nonlinear interconnected systems via deterministic learning

In this paper, a fast fault detection scheme is developed for a class of nonlinear interconnected systems with output measurements. First, through combining an adaptive high gain observer with the deterministic learning theory, the system states and unknown dynamics are estimated simultaneously. However, large value of gain may let the estimator becomes noise sensitive. Thus, the observer structure is modified to avoid this issue. Second, by reusing the estimated knowledge which is stored in the constant radial basis function (RBF) neural networks, a bank of dynamic estimators are constructed. Then, the average [Formula: see text] norms of residuals are generated. The smallest residual principle is considered for decision-making in the detection phase. Third, the fault detection conditions and detection time are analyzed under the influence of observer gain. A simulation example is utilized to illustrate the effectiveness of this scheme.

Discrete Wavelet Transform for Stator Fault Detection in an Induction Motor

Discrete Wavelet Transform for Stator Fault Detection in an Induction Motor

Fault Detection of Induction Motors with Combined Modeling- and Machine-Learning-Based Framework

This paper deals with the early detection of fault conditions in induction motors using a combined model- and machine-learning-based approach with flexible adaptation to individual motors. The method is based on analytical modeling in the form of a multiple coupled circuit model and a feedforward neural network. In addition, the differential evolution algorithm independently identifies the parameters of the motor for the multiple coupled circuit model based on easily obtained measurement data from a healthy state. With the identified parameters, the multiple coupled circuit model is used to perform dynamic simulations of the various fault cases of the specific induction motor. The simulation data set of the stator currents is used to train the neural network for classification of different stator, rotor, mechanical, and voltage supply faults. Finally, the combined method is successfully validated with measured data of faults in an induction motor, proving the transferability of the simulation-trained neural network to a real environment. Neglecting bearing faults, the fault cases from the validation data are classified with an accuracy of 94.81%.

ISHM for fault condition detection in rotating machines with deep learning models

ISHM for fault condition detection in rotating machines with deep learning models

An active fault tolerance framework for uncertain nonlinear high-order fully-actuated systems

This paper presents a novel active fault tolerance framework for uncertain high-order fully-actuated systems (HOFASs). Starting from the uncertain faulty HOFAS model with nonlinear measurement, the advantages of this description are revealed via Lie Derivatives. A HOFAS adaptive observer and controller integration is established and further a closed-loop HOFAS dynamic data modeling structure is derived. This structure uncovers the dynamic characteristics including fault features, and naturally yields a fault detectability condition. Based on the fault information from the dynamic data model, an active fault tolerant control strategy for HOFASs is proposed, which enriches HOFAS control theory. The uniformly bounded stability of the HOFAS model is proved theoretically and illustrated experimentally.

Novel Approach for Synchronous Generator Protection Using New Differential Component

A new stator fault detection and classification scheme for synchronous generators (SGs), based on an amplified discrete Teager-kaiser energy operator (ADTKEO) of a new differential component ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DC_SI</i> ), is proposed. Detection of an internal fault is based on the fulfillment of the ADTKEO ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ξ</i> ) with respect to the pre-determined threshold along with the checking of second-harmonic content in the current transformer (CT) secondary current. Afterward, the classification of internal faults is performed based on ‘ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ξ</i> ’. The performance of the proposed scheme is tested on various fault cases obtained from a simulation network developed in a real-time digital simulator and the developed laboratory prototype of the SG. The results indicate that the proposed scheme not only provides 100% winding coverage against phase-to-ground faults but is also capable of detecting other phase faults within a quarter cycle for low-resistance grounded SGs. At the same time, it stays extremely stable for external faults, even considering errors in CT along with its saturation (without compromising speed/sensitivity). The proposed scheme is independent of the sampling frequency, loading condition, decaying dc component present in the acquired signals, and rating of the SG.

A data driven fault detection approach with an ensemble classifier based smart meter in modern distribution system

The operation of the distribution grids is constantly being threatened by occurrence of faults. These faults are unpredictable and dangerous. Moreover, the introduction of distributed generations (DGs) in modern distribution networks has made the detection of these faults more complicated. Thus, to maintain the continuity of the power supply, a fast and accurate fault detection strategy is required to isolate the fault. In view of this, the paper presents a data driven fault detection approach with an ensemble classifier based smart meter in modern distribution system. To achieve this, a random forest (RF) based fault detection algorithm is programmed within the smart meter. At first, the magnitude of maximum angular difference between positive and zero sequence component of the current at the DG bus is computed. The computed value is then fed to a trained RF classifier for detecting fault conditions. The greatest advantage of the proposed algorithm is the non-requisite of additional hardware or software kit for fault detection. The algorithm undergoes several test conditions to showcase its effectiveness. The algorithm depicts a high accuracy of 98.95%. The results signify that the algorithm is robust, reliable and accurate with an additional benefit of augmenting situational awareness.

Stator Faults Detection in Asymmetrical Six-Phase Induction Motor Drives with Single and Dual Isolated Neutral Point, Adopting a Model Predictive Controller

Multiphase drives have been presented as potential replacements for conventional three-phase machines, primarily because of their propensity to operate faultlessly. Due to the various stator phase arrangements, standard fault detection techniques are insufficiently applicable and cannot be used to diagnose faults in the various configurations of multiphase machines in closed-loop applications. The current study proposes an effective online diagnostic technique based on the computing and tracking of a significant severity factor, which is defined as the ratio of the zero, negative, and positive voltage symmetrical components employing a short-time least-square Prony algorithm (STLSP). In this study, the asymmetrical six-phase induction motor (ASPIM) was controlled by a model predictive control (MPC) algorithm, an attractive control scheme for the regulation of multiphase electric drives, since it easily exploits their inherent advantages. This article addresses stator faults in ASPIMs. The effectiveness of the suggested strategy was confirmed experimentally for various operating conditions in both steady and transient states.

Investigation of Crack in Beam Structure using an Adaptive-Genetic Algorithm (AGA)

Fault detection and continuous condition monitoring in structural and machine elements are very sensitive topics and gaining significant value as a current research area. Due to the continuous loading and unloading of these elements, fatigue occurs. For the above-mentioned reason, crack is initiated and propagated. The initiation of any type of crack changes the physical properties of the structural and machine elements, which directly affects the lifetime of the element. The presence of any discontinuity changes the physical properties of the element, which also changes elastic properties. These alterations in physical properties change the modal properties of the structural elements. These changes in the vibration criteria can be used for the identification and quantification of the damage. In this research work, the vibration parameters are combined with Artificial Intelligence (AI) to predict the damage location. Here the natural evolution-based Genetic Algorithm (GA) has been used for the training of vibration features (frequencies). It has been discovered that the original AI methods are sometimes not able to give the proper prediction of damage location as they may be trapped in the local optimum. So, to counteract this loophole and to make it more flexible so that it can adjust to the constraints of real-life problems, a data mining method using Regression Analysis (RA) has been proposed and the results have been compared.