Fault Detection Scheme Research Articles

Condition monitoring based on corrupted multiple time series with common trends

Condition monitoring is a fundamental task in the reliability engineering and operation management of a complex industrial system. It aims to detect faults based on sensing data but poses significant challenges when dealing with corrupted multiple time series data in many real-world applications. These time series typically exhibit similar changing patterns influenced by common trends (e.g., workload, ambient condition) and physical relationships among corresponding variables, and are often significantly corrupted large outliers (e.g., transmission interruption). Although several traditional common trend models have been employed to analyze such condition monitoring data in the literature, they are fully parametric, constrained by restrictive assumptions, and not robust to outliers. In this article, we propose a novel semiparametric decomposition model to analyze a set of monitored time series and separate it into common, idiosyncratic, and sparse components, under relatively mild assumptions. We also introduce effective algorithms for model estimation and a monitoring scheme for fault detection. The numerical and real case studies demonstrate the superiority of the proposed method over existing approaches, in terms of both decomposition accuracy and detection performance for system faults.

Protection strategy for fault detection in AC microgrid based on MVMD & differential CUSUM

Abstract In the era of smart grids and microgrids, the transformation of the traditional grid system brings many operational, technical, and economic benefits. However, the complexity of the network due to the integration of various distributed generations (DGs), continuous change of topology, and non-linear load make fault detection a major issue that forces power engineers to focus on. In this paper, a novel fault detection scheme is suggested based on the multivariate variational mode decomposition mode (MVMD) and differential cumulative sum (DCUSUM). As a generalized extension of the original variational mode decomposition (VMD) algorithm for multivariate data residing in multidimensional spaces, the main goal of MVMD is to decompose the input signal into different band-limited intrinsic mode functions (IMFs). Due to the inherent characteristics of being insensitive to noise and very effective in decomposing the local features even with similar frequencies, it is very effective for fault detection in microgrid distribution systems. The proposed DCUSUM algorithm computes the differential cumulative energy for the remaining significant modes. A fault detection index is considered in this approach and applied for fault detection by adaptively through the threshold setting to accurately result in fault detection. To justify the proposed approach, a standard AC microgrid test system is considered and the approach is verified for fault detection under various fault conditions and resistances. The obtained results and the comparative analysis with other methods reflect the better accuracy, robustness, and reliability of the proposed approach.

Current sensor unanticipated fault detection and compensation of 14.5-meter optical/infrared telescope drive system based on algebraic transformations

This paper discusses an unanticipated fault detection, isolation, and compensation (FDIC) strategy for the arc splicing permanent magnet synchronous motor (PMSM) operating under a promising 14.5-meter optical/infrared telescope drive system, specifically focusing on current sensors. The application is based on algebraic transformations that allow not only failure detection but also location and isolation. Besides, detection is performed merely by means of the measured current sensors and does not require additional knowledge or estimators; isolation utilizes information provided by detection to locate where the faults originated and release isolation signals; and compensation is carried out through the remaining currents that are not affected by the faults. It is noted from the results that the performance of FDIC in faulty mode is very acceptable and uncompromising in terms of fault detection, current symmetry, speed tracking, load torque and robustness remedial measures. Therefore, the proposed method can effectively ensure the stable operation of the drive system in the presence of faults.

Comprehensive Survey on Fault Detection and Classification in Three-Phase Induction Motors

Electric motors have revolutionized the way of human living and resulted in the modern lifestyle. These motors often operate in corrosive and dusty places and are exposed to a variety of undesirable conditions and situations that result in the failure of the motor. The faults occurring in Induction Motors (IM) need to be detected at a proper time for avoiding losses and further consequences. A well-designed fault detection scheme not only reduces motor failure but also increases productivity and even sometimes avoids accidents. This paper presents a review of fault detection and classification techniques in three-phase induction motors (TPIM). The main theme of this paper is to revisit the conventional methods for fault detection in TPIM and compare them with recently published methods based on parameters to be sensed, and the type of fault that can be detected, with their advantages and drawbacks. Around a hundred papers are critically reviewed and studied from old and new regimes. Attention is also given to fault detection methods based on artificial intelligence (AI) and machine learning (ML). This paper concludes with brief remarks which will be very useful for new researchers who are willing to research in the domain of fault detection and classification.

Fault detection and isolation for multi-type sensors in nuclear power plants via a knowledge-guided spatial–temporal model

Sensor faults in nuclear power plants (NPPs) have the potential to propagate negative impacts on system stability, leading to false alarms and accident misdiagnosis. Existing methods seldom concurrently consider complex spatial–temporal correlations among multi-type sensors in the primary circuit. This study presents a novel sensor fault detection and isolation scheme named the knowledge-guided spatial–temporal model (KGSTM), using the knowledge-guided recurrent unit (KGRU) and the concurrent detection strategy. To organically express part and whole interdependencies from inherent sensor layout, several graphs are specifically designed with pertinent domain knowledge. KGRU consists of the multi-graph convolutional network (MGCN) for fusing various spatial information and the gate recurrent unit (GRU) for extracting dynamic temporal features, further obtaining precise reconstructed signals and residuals. The concurrent detection strategy can explicitly quantify abnormal behaviors to detect and isolate faulty sensors by characterizing spatial–temporal signal variation. Numerical results on two real-world datasets from a pressurized water reactor (PWR) with simulated faults illustrate that the KGSTM has superior performance over various state-of-the-art methods in terms of signal reconstruction and fault detection.

Robust multiple fault detection and estimation approaches based on adaptive finite‐time observer for nonlinear systems

AbstractThis article investigates robust multiple fault detection (FD) and fault estimation (FE) schemes, based on adaptive finite‐time observer (AFTO), applied to nonlinear systems with disturbances, actuator faults, and sensor faults. The AFTO is designed, and a parameter adaptive law is constructed to continuously optimize the residual based performance function to improve the AFTO robustness. The schemes are able to simultaneously detect and estimate unknown states, disturbances, and faults. The AFTO's design, finite‐time convergence and stability conditions are guaranteed in terms of linear matrix inequalities. Finally, simulation results, based on two typical practical systems, are given to illustrate the effectiveness of the proposed approaches. The simulation results also show that the designed AFTO can offer faster convergence speed, higher estimation accuracy, and stronger robustness than conventional observer and sliding mode observer.

Leveraging dynamic power benchmarks and CUSUM charts for enhanced fault detection in distributed PV systems

Faults in photovoltaic (PV) systems are common during their operational lifetime. Existing PV fault detection methods, which are primarily designed for large-scale PV fields, struggle with distributed systems due to their reliance on on-site weather sensors and inability to handle complex local shading effects on PV performance in the built environment. This paper presents a fault detection scheme specifically applicable to distributed PV systems that solely relies on monitored AC output and remote weather sources. Without requiring additional equipment or detailed information on the PV configuration and local shading conditions, the proposed method consists of two steps. First, historical monitoring data is used in a bootstrap fashion to develop machine learning models that establish baselines for normal PV output and corresponding estimation uncertainty. Then, a dynamic PV power benchmark constructed based on the 3-sigma rule, and cumulative sum (CUSUM) control charts are employed simultaneously to detect system malfunctions. Through experimental verification on actual distributed PV systems, the effectiveness of the proposed method is assessed for four categories of frequently observed malfunctions. The comprehensive experimental results indicate the considerable efficiency of the proposed sensorless method in detecting both serious PV malfunctions and minor faults with a severity as small as 4.2%.

A fast data-driven fault detection and location method for unknown distributed thermal processes

The distributed thermal processes are widely present in various industrial operations, but the presence of time delay effect and unknown model information make achieving fast and timely fault detection challenging. Furthermore, the involvement of spatiotemporal coupled data in measurement processes further exacerbates the situation. To solve these problems, this paper proposes a data-driven fault detection and location method for thermal processes described by distributed parameter systems (DPS). Firstly, a Time/Space separation is employed to decoupling the spatio-temporal data into spatial basis functions and temporal coefficients. Subsequently, dynamic partial least squares (DPLS) is employed to obtain the temporal information of the model with fault-free data. Monitoring statistics in both the temporal and spatial domains are then constructed, and their thresholds are estimated by kernel density function. Finally, an online strategy for fault detection and location is presented. The method was validated on the catalytic rod, and an experimental snap curing oven.

Kernel-based cumulative sum and differential cumulative sum approaches for resilient fault detection in LVDC microgrids

The energy industry is rapidly shifting from conventional to smart grid systems, involving advancements in microgrids and power management systems. However, achieving desired outcomes with the existing grid setup poses significant challenges. To address the above issues, adopting a Low Voltage Direct Current (LVDC) distribution system provides successful operational advantages. The primary objective of this work is to improve the system's security by developing a more resilient fault detection scheme. The presented work has proposed a Kernel-based Cumulative Sum (K-CUSUM) and Kernel-based Differential Cumulative Sum (K-DCUSUM) technique which is designed to identify changes in data under fewer assumptions. Instead of using the classical Maximum Mean Discrepancy (MMD) technique, the K-CUSUM and K-DCUSUM methods use a non-parametric MMD testing framework to evaluate incoming data by comparing it to reference distribution samples. Further, a comprehensive analysis is undertaken on an LVDC test system, considering fault scenarios involving pole-to-pole and pole-to-ground configurations. This research encompassed the determination of fault location, evaluation of fault resistance, examination of noise introduction, and an exploration of the influence of photovoltaic (PV) penetration. Moreover, the proposed approach is tested for a standard LVDC Microgrid. The results are compared with other contemporary techniques and recent literature.

Advanced impedance mismatch technique for detecting faults in photovoltaic systems

AbstractThis paper presents a comprehensive exploration of the Advanced Impedance Mismatch Technique (AIMT), a novel approach designed for the accurate detection of simultaneous and varied faults within photovoltaic (PV) systems. This investigation integrates a spectrum of fault detection strategies, pinpointing reflectometry as a notably effective tool. Despite its utility, conventional reflectometry applications face critical constraints, notably the limitation to identify only the primary fault location within a PV array and the inability to distinguish between different fault types. This work introduces an innovative mathematical model that estimates the impedance of PV modules, enhancing the reflectometry method to enable the precise identification and localization of multiple defective modules within a string. The proposed technique exhibits a remarkable sensitivity to detect slight impedance differences between a functional PV string and one with defective modules. The validity of the AIMT's mathematical model is corroborated through simulation experiments on a string of seven PV modules afflicted with multiple simultaneous faults. These experiments rigorously evaluate the technique's accuracy in pinpointing the locations of defective modules within a PV string. The outcomes of our proposed ‐times faster AIMT reveal a strong concordance between the simulated reflective signals and the impedance values forecasted by the model, highlighting the proposed method's proficiency in the detailed detection and diagnosis of progressive faults within PV systems.

Chemical process fault detection and trend analysis based on KESN

AbstractFault detection has great significance for chemical process safety with the development of science and technology. The conventional echo state network‐based fault detection method does not highlight key fault features, and cannot forecast future fault trend after the occurrence of faults. For the above problems, a chemical process fault detection and trend analysis strategy based on key feature enhanced echo state network (KESN) is proposed. First, dynamic features are extracted by a detecting echo state network. Then, a weighting strategy is designed to enhance key features and increase fault detection rates. After detecting a fault, independent component analysis is utilized to extract independent key features. Future fault trend is forecasted based on the forecasting multi‐KESN. Simulation results on the Tennessee Eastman process demonstrate the effectiveness of the proposed method.

Spacecraft attitude testbed

Spacecraft attitude determination and control systems (ADCS) stand as a pivotal factor for ensuring the success of missions. As satellite sizes continue to decrease, the need for more intelligent and efficient control algorithms becomes increasingly pronounced. In response to this imperative, we propose the development of a scaled-down testbed specifically designed for evaluating satellites (ACDS). This testbed will demonstrate effective control of a CubeSat through the use of Helmholtz coils, magnetic torquer solenoids, reaction wheels and inexpensive student-made components and available components off the shelf (COTS) in the San Diego State University SPACE Lab. Referred to as the Spacecraft Attitude Testbed (SAT), this custom-designed 3-DoF experimental attitude platform replicates the conditions experienced by a weightless satellite in space. The objective of this research paper is to improve the understanding of how to manufacture a testbed and components like those within this paper, expose some of the issues encountered and improve the outcomes for future academic testbeds. Within this study, the testbed employs torque application to magnetic torquers and reaction wheels to regulate attitude. In summary, this research endeavors to establish a comprehensive and cost-effective testbed platform dedicated to the assessment of spacecraft attitude control and determination systems. It not only seeks to enhance our understanding of advanced control algorithms but also aims to contribute valuable insights into fault detection and adaptive control strategies, ultimately bolstering the efficiency and robustness of satellite operations.

A Fault Detection and Compensation Strategy of the Resolver for Variable Leakage Flux Permanent Magnet Motor With Threshold Self-Regulation

A Fault Detection and Compensation Strategy of the Resolver for Variable Leakage Flux Permanent Magnet Motor With Threshold Self-Regulation

Advancing maintenance, monitoring, and control techniques for the optimization of power electronic systems

This paper explores the innovative methodologies and technologies in predictive maintenance, real-time monitoring, fault detection, and advanced control strategies for power electronic devices. Initially, we delve into data acquisition and preprocessing techniques crucial for ensuring the quality and reliability of data used in predictive models. These models, leveraging machine learning and time-series analysis, predict the remaining useful life of devices, guiding proactive maintenance strategies. Furthermore, we discuss the significance of risk assessment and decision support systems in prioritizing maintenance tasks and allocating resources efficiently. The narrative then shifts to real-time monitoring and fault detection, emphasizing the role of sensor integration, data fusion, anomaly detection, and diagnostics in maintaining system integrity. Condition-based maintenance strategies, underscored by real-time data analytics, are presented as a means to optimize maintenance activities and enhance operational performance. The paper concludes with a detailed examination of advanced control strategies, including model predictive control, reinforcement learning, and distributed control and optimization techniques, highlighting their potential to improve system efficiency, reliability, and adaptability. Through comprehensive research and analysis, this work aims to provide valuable insights into the development of sophisticated maintenance and control mechanisms for power electronic systems, ultimately contributing to their longevity and operational efficacy.

Robust fault detection and isolation for uncertain neutral time-delay systems using a geometric approach

This paper proposes a new geometric fault detection and isolation (FDI) strategy for uncertain neutral time-delay systems (UNTDS). Firstly, the concept of unobservability subspace is extended to the considered system. Subsequently, utilizing the geometric properties of factor space and canonical projection, the fault is divided into different unobservability subspaces. Therefore, an algorithm for constructing the subspace is developed for fault isolation. Finally, a set of observers is designed for the subsystems, and generates a set of structured residuals which is sensitive only to a specific fault. Additionally, the H∞ technique is utilized to suppress the disturbances and error signals due to time-varying delays on the residual. The simulation examples verify the effectiveness of the proposed approach.

Improved fault detection and diagnosis using graph auto encoder and attention-based graph convolution networks

A powerful fault detection and diagnosis (FDD) system plays a pivotal role in achieving operational excellence by maximizing system performance, optimizing maintenance strategies, and ensuring the longevity and resilience of process plants. In the context of FDD for multivariate sensor data, this study presents an improved FDD approach using graph-based neural networks. This graph neural network uses an adjacency matrix developed by extracting the expert domain knowledge and topological information of the multi-sensor system. This additional graph representation of the system is incorporated along with multivariate sensor data to capture the spatial and temporal information in neural networks efficiently. In this regard, we propose and evaluate: 1) A Graph Auto Encoder (GAE) based fault detection strategy and 2) An Attention-based Spatial Temporal Graph Convolution Network (ASTGCN) based fault diagnosis methodology. By leveraging the additional knowledge in the form of graphs, the GAE captures the complex relationships and dependencies among sensors, enabling effective anomaly detection, which identifies abnormal patterns and deviations from normal behavior, thus indicating potential faults in the system. The ASTGCN incorporates attention mechanisms to selectively focus on relevant sensor nodes and capture their spatial and temporal dependencies for fault diagnosis. The effectiveness of the proposed FDD approach is demonstrated using the benchmark Tennessee Eastman Process (TEP) problem. The results show that the proposed approaches outperform traditional methods and highlight the importance of leveraging graph-based knowledge for FDD in complex systems.

Sliding mode observer-based fault diagnosis and continuous control set fault tolerant control for PMSM with demagnetization fault

A fault detection, estimation and fault tolerant control scheme for demagnetization fault in the permanent magnet synchronous motor systems is proposed in this paper. When demagnetization fault occurs, the system will deviate from the normal state severely. To facilitate the calculation, the demagnetization fault is converted to the dq coordinate system for processing and the resistance change is modeled as the system uncertainty, and the effect of the uncertainty on the system is analyzed. Then a fault detection observer is designed to detect the fault occurrence time, and a sliding mode observer based fault estimation algorithm is designed to estimate the fault and disturbance information. The observer parameters are determined by linear matrix inequality. When the fault information is obtained, a continuous control set fault tolerant control scheme is presented to minimize the impact of fault as far as possible. Finally, experimental results provide evidence of the feasibility of the suggested algorithm.

Fault detection and classification scheme for power islands with inverter interfaced distributed generators

The low fault current contribution from inverter-interfaced distributed generators (IIDGs) and bidirectional flow of current in the power networks introduces protection challenges when islanded. The existing fault detection scheme based on the transient monitoring function (TMF) fails under such scenarios. A new decentralised machine learning (ML) based fault detection and classification scheme is proposed in this paper using terminal voltage information of IIDG (vo−abc). TMF of vo−abc, along with the zero sequence component of vo−abc, is utilised for training and testing of ensemble bagged trees ML algorithm. The proposed scheme is implemented inside each IIDG, and hence, it does not require communication. It is validated on a modified CIGRE test benchmark microgrid for all shunt faults at various locations with different loading, fault resistances, and inception angles. The accuracy of detection and classification of fault (DCF) during fault cases with unbalanced loads, noise, harmonic loads, and no-fault scenarios such as load and DG switching proves the effectiveness of the proposed method with varying scenarios. The performance of the proposed scheme is found to be superior when compared with the existing threshold based TMF and Hilbert–Huang transform (HHT) based techniques considering accuracy, dependability, security, and response time.

Evaluation of data imputation approaches for multi-stream building systems data1

Increasing advancements in building digitization, smart sensing, and metering technologies have allowed large amounts of timeseries data to be collected for monitoring, analyzing, and controlling building systems. However, due to sensor or communication failures, the data collected are often incomplete and poor in quality. Data imputation approaches to replace the missing values, specifically based on either statistical or predictive models have been widely adopted for multivariate datasets in other domains. It is hence of interest to find an effective way to impute timeseries data collected from a building system. In this paper, we evaluate multiple data imputation approaches using data collected from a medium sized building situated in Stockholm, Sweden and a small commercial building from the ASHRAE RP-1312 research project. Sensors with widely varying characteristics from the case study buildings were selected to evaluate the imputation methods. The imputation accuracy and the impact of each chosen imputation method on information entropy, short-term building forecasting model performance, and fault detection strategy were evaluated. Results demonstrate that incorporating time-lagged cross correlations within a k -nearest neighbor ( k NN) model provide the most accurate imputations without affecting the quality of subsequent data analysis.

Signal-based fault location method using time and frequency domain

This research introduces an innovative approach to fault detection in high-voltage alternating current (HVAC) systems by integrating temporal and frequency domain characteristics of signals. It utilizes Traveling Wave (TW) detection and analysis, focusing on a precise window of just 100 microseconds surrounding the TW's arrival at the measurement point. Mathematical Morphology is applied to identify time-domain properties, while the Stationary Wavelet Transform is employed to extract frequency-domain features. Additionally, the integration of a Dynamic Mode Decomposition-based technique enhances TW detection accuracy. The efficacy of this approach is extensively assessed for fault site classification and regression tasks, demonstrating superior performance compared to utilizing time or frequency-domain features separately. The study also examines the method's resilience to noisy measurements and the sufficiency of training data. Comparative analysis against existing signal-based approaches within the IEEE 34 node system underscores its heightened accuracy. Furthermore, the proposed technique showcases minimal fault location estimate errors along the feeder length and competes favorably with slower phasor-based fault detection methods in performance. Regarding the distance of the fault, its proximity to the generator, transmission line, or load significantly influences the parameters of positive, negative, and zero sequence components. When a fault occurs, these parameters undergo distinct changes, serving as indicators of the fault's severity and location. If the fault occurs near the generator or substation, there may be alterations in the voltage and current magnitudes, affecting the sequence components accordingly. In the case of a fault in the transmission line, impedance changes can lead to variations in sequence parameters. Similarly, a fault near the load can cause disturbances in the voltage and current profiles, impacting the sequence components. Analyzing these changes in positive, negative, and zero sequence parameters during a fault provides valuable insights into the fault's severity and location, aiding in effective fault detection and mitigation strategies. Key Words – fault detection, high-voltage alternating current (HVAC) systems, Traveling Wave (TW), temporal domain, frequency domain, Mathematical Morphology, Stationary Wavelet Transform, Dynamic Mode Decomposition, fault site classification,