Failure Detection Techniques Research Articles

Tie failure detection technique for total cross tied solar photovoltaic system

Partial Shading in the solar photovoltaic (PV) system leads to more power loss due to mismatch among cells or module’s power generation. The Total Cross Tied (TCT) array design approach is employed for Partial Shading Conditions (PSC) to reduce this issue and boost output power. A tie is used to connect the strings and modules in a solar-powered photovoltaicarray. However, external environmental conditions like strong winds and storms may influence this tie to break. The conventional protection and fault identification technique in the PV system failed to detect this fault. The Solar PV system's entire power generation is drastically impacted by this TCT tie connector problem. Furthermore, during PSC, there is a notable fall in power generation too. To address this, a new technique has been proposed in this work which is capable of precisely identifying the tie interconnection fault in the PV array. A line inspector circuit is used to identify and classify the partial shading and tie interconnection fault. This proposed technique is experimentally verified in the 4 × 4 PV array of 160 W. In all the cases, the proposed algorithm exactly finds the tie interconnection failure. When the diagnosed fault is fixed, the PV array's power output is found to rise by up to 4.5 times.

A Review of Photovoltaic Module Failure and Degradation Mechanisms: Causes and Detection Techniques

With the global increase in the deployment of photovoltaic (PV) modules in recent years, the need to explore and understand their reported failure mechanisms has become crucial. Despite PV modules being considered reliable devices, failures and extreme degradations often occur. Some degradations and failures within the normal range may be minor and not cause significant harm. Others may initially be mild but can rapidly deteriorate, leading to catastrophic accidents, particularly in harsh environments. This paper conducts a state-of-the-art literature review to examine PV failures, their types, and their root causes based on the components of PV modules (from protective glass to junction box). It outlines the hazardous consequences arising from PV module failures and describes the potential damage they can bring to the PV system. The literature reveals that each component is susceptible to specific types of failure, with some components deteriorating on their own and others impacting additional PV components, leading to more severe failures. Finally, this review briefly summarises PV failure detection techniques, emphasising the significance of electrical characterisation techniques and underlining the importance of considering more electrical parameters. Most importantly, this review identifies the most prevalent degradation processes, laying the foundation for further investigation by the PV research community through modelling and experimental studies. This allows for early detection by comparing PV performance when failures or degradation occur to prevent serious progression. It is worth noting that most of the studies included in this review primarily focus on detailing failures and degradation observed in PV operations, which can be attributed to various factors, including the manufacturing process and other external influences. Hence, they provide explanations of these failure mechanisms and causes but do not extensively explore corrective actions or propose solutions based on either laboratory experiments or real-world experience. Although, within this field of study, there are corresponding studies that have designed experiments to suggest preventive measures and potential solutions, an in-depth review of those studies is beyond the scope of this paper. However, this paper, in turn, serves as a valuable resource for scholars by confining PV failures to critically evaluate available studies for preventative measures and corrective actions.

Bridging the gap between AI and the industry — A study on bearing fault detection in PMSM-driven systems using CNN and inverter measurement

Deep-learning-based identification and diagnosis of faults in industrial assets has received much attention. It does not need a deep understanding of the target domain or complex signal-processing techniques for time-consuming feature identification and selection. The vast majority of approaches developed in this field rely on public data recorded in laboratory conditions with high-performance measurement equipment. Moreover, primarily vibration data is used for condition monitoring tasks, which is particularly sensitive to typical error patterns in rotating machinery. These conditions are difficult to maintain in industrial environments since using high-performance measurement systems would not be economically feasible. This paper demonstrates that modern deep learning can achieve extraordinary fault detection results with industrial standard low-cost measurement systems. We determine measurements taken from the industrial-sensory equipment that potentially capture the system’s fault patterns. Hereafter, we apply traditional failure detection and identification techniques to evaluate the suitability of the taken measurements for bearing fault detection. These techniques apply envelope analysis to reveal the fault patterns in the signals and a support vector machine to separate the fault classes. We can conclude that the used low-cost sensory equipment is capable to capture meaningful, fault-describing patterns. To improve the failure detection performance we are investigating a 2D, as well as 1D, convolutional neural network approach to identify the error patterns and classify the respective errors. We compare the deep-learning-based methods with the traditional methods. Furthermore, we assess which inverter signal carries the largest fault-describing information content. Experimental results indicate that the proposed deep learning methods outperform traditional fault diagnosis methods, hence, demonstrating the effectiveness in an industrial condition monitoring application.

Online Failure Detection using Deep Learning in FPGA PCB Interface

This research paper is aimed to present a real-time failure detection technique while working with Field Programmable Gate Arrays (FPGA) and interfaced Printed Circuit Boards (PCBs). In this research, we explored the feasibility of currently available innovative Deep Learning (DL) algorithms to detect the defects in variety of PCBs. In our proposed technique, we trained the YOLOv5 (You Only Look Once) algorithm with a few hundreds of defective PCBs’ images, which were obtained from Kaggle, an online community of data scientists and machine learning practitioners. The advantage of using YOLOv5 is that the detection is carried out in real-time. In the next phase, after training, the algorithm undergoes validation and testing, where we tested with different images. The obtained results are promising, as the Deep Learning process successfully detects the defects on the PCBs.

A new mooring failure detection approach based on hybrid LSTM-SVM model for semi-submersible platform

A reliable and stable mooring system is of vital significance to warrant the continuous operation and personnel heath on offshore floating platforms. The failure detection technique for the mooring line can provide an early warning and response for people so that action can be taken earlier to avoid more severe consequences. This paper proposed a novel approach based on Long short-term memory neural networks and support vector machine (LSTM-SVM) to detect the mooring line failure. Two combined failure signals can judge the actual mooring failure occurrence. The preliminary mooring failure signals can be identified based on the significant increase of errors in the LSTM prediction of three platform motions. Besides, a SVM classifier is constructed based on three parallel sea state lines to distinguish the impact of motions between sea state change and mooring failure and capture the secondary failure signal. Five time-domain features are extracted from heave motion, and well-trained SVM model with 132 group cases data from three sea state lines can achieve characteristic recognition of sea state change and mooring failure. According to four new cases with continuously changing sea states, the proposed method indicates the feasibility of mooring failure detection when platforms suffer the influence of rapid environmental changes under a broader range of sea states, and the LSTM-SVM can still distinguish the sea state change and actual mooring failure. In this study, motion data acquisition relies on the hydrodynamic simulation of a semi-submersible platform (SEMI). For practical application, the proposed method can utilize the mature Differential Global Positioning System (DGPS) to obtain the data and realize the mooring failure detection in real time.

ML Detection and Isolation of Functional Failures using Syndrome Diagnostics

Failure identification in complex engineering systems often needs to be in form of multilevel analysis, the first of which involves detection and isolation of functional disruptions across the system down to a discrete item. Considering that there is a functional flow(s) of operation in the machine, the loss or deviation of that functional flow(s) will imply a Functional Failure of the system. Most often, the disruption may propagate from where it may be first found and thus, the root issue can be isolated via the causal relationship between components and their flows. This paper references the idea of Causation-based AI: intelligence that brings together the fast estimations of machine learning and the domain-based physics of failure via qualitative models in the form of Syndrome Diagnostics (SD). There are three main routines in SD. Firstly, the current operating mode of the system is determined. Secondly, functional failure detection techniques are used to detect the existence of an anomaly. Thirdly, a functional failure isolation routine is executed to isolate the failing component, which is composed of two steps: using classification methods to generate a predicted syndrome and matching this generated syndrome with the failure syndrome pattern extracted from the Digital Risk Twin of the system (where the causation aspect is taken advantage of). A series of experiments are conducted and 90% of the failures in validation data have been correctly identified, which verifies the effectiveness of SD in terms of functional failure detection and isolation.

Failure Detection Methods for Pipeline Networks: From Acoustic Sensing to Cyber-Physical Systems.

Pipeline networks have been widely utilised in the transportation of water, natural gases, oil and waste materials efficiently and safely over varying distances with minimal human intervention. In order to optimise the spatial use of the pipeline infrastructure, pipelines are either buried underground, or located in submarine environments. Due to the continuous expansion of pipeline networks in locations that are inaccessible to maintenance personnel, research efforts have been ongoing to introduce and develop reliable detection methods for pipeline failures, such as blockages, leakages, cracks, corrosion and weld defects. In this paper, a taxonomy of existing pipeline failure detection techniques and technologies was created to comparatively analyse their respective advantages, drawbacks and limitations. This effort has effectively illuminated various unaddressed research challenges that are still present among a wide array of the state-of-the-art detection methods that have been employed in various pipeline domains. These challenges include the extension of the lifetime of a pipeline network for the reduction of maintenance costs, and the prevention of disruptive pipeline failures for the minimisation of downtime. Our taxonomy of various pipeline failure detection methods is also presented in the form of a look-up table to illustrate the suitability, key aspects and data or signal processing techniques of each individual method. We have also quantitatively evaluated the industrial relevance and practicality of each of the methods in the taxonomy in terms of their respective deployability, generality and computational cost. The outcome of the evaluation made in the taxonomy will contribute to our future works involving the utilisation of sensor fusion and data-centric frameworks to develop efficient, accurate and reliable failure detection solutions.

CalibrationTalk: A Farming Sensor Failure Detection and Calibration Technique

Sensor failure detection and calibration is an important issue for Internet of Things applications. Many smart applications fail due to inaccurate data produced by their sensors. To solve this issue, we propose the CalibrationTalk mechanism. We use smart agriculture as an example to show how CalibrationTalk automatically fixes the sensing accuracy problems in commercial farm operations. Besides manufacturing variation, hardware malfunction and aging, sensing inaccuracy may be caused by soil erosion of rain or irrigation, which changes the positions and the angles of the sensors inserted in the soil. The above sensor failure problems can be detected by CalibrationTalk. In this article, we first extend our previous work to detect sensor failures. If the sensors become inaccurate due to aging or similar reasons, CalibrationTalk automatically calibrates the sensors in the farm fields. We conduct measurements and analytic modeling to investigate the performance of CalibrationTalk, which suggests that to detect a potential moisture sensor failure, it is appropriate to set the detection time as 30 min for moisture sensors. With this setup, true failure is detected while false detections rarely occur. For a detected sensor failure due to aging, CalibrationTalk can automatically build a calibration table. During the calibration table establishment period, the sensors still produce correct measured data and do not interfere with the normal operation of smart farming. The value of CalibrationTalk lies in its power of automation for sensor failure detection and calibration processes developed in this article.

Diagnosis of Fault Modes Masked by Control Loops with an Application to Autonomous Hovercraft Systems

This paper introduces a methodology for the design, testing and assessment of incipient failure detection techniques for failing components/systems of critical engineered systems/processes masked or hidden by feedback control loops. It is recognized that the optimum operation of critical assets (aircraft, autonomous systems, industrial processes, etc.) may be compromised by feedback control loops, which mask severe fault modes while compensating for typical disturbances. Detrimental consequences of such occurrences include the inability to detect expeditiously and accurately incipient failures, loss of control, and inefficient operation of assets in the form of fuel overconsumption and adverse environmental impact. A novel control-theoretic framework is presented to address the masking problem. Major elements of the proposed approach are employed in simulation to develop, implement and validate how faults are distinguished from disturbances and how faults are detected and identified with performance guarantees, i.e., prescribed confidence level and given false alarm rate.The demonstration and validity of the tools/methods employed necessitates, in addition to the theoretical content, a suitable testbed. We have employed and describe briefly in this paper an autonomous hovercraft as the test prototype. We pursue a systems engineering process to design, construct and test the prototype hovercraft instrumented appropriately for purposes of fault injection, monitoring and the presence of control loops. We emphasize a general control-theoretic framework to the masking problem and utilize a simulation environment to derive results and illustrate the efficacy of the methodology.

Analysing Current Signature Data to Diagnose an In-Service Wind Turbine Generator

Condition-based maintenance (CBM) is currently seen as the preferred approach to optimise operation and maintenance (O&M) strategies by the early detection and diagnosis of critical failures for wind turbines (WT). WT reliability is highly affected by failures of the mechanical components from the drive train, although doubly-fed induction generators (DFIG) also contribute to high failure rates and downtime. As DFIG is the dominant technology employed for variable speed WTs, dedicated failure detection and diagnosis techniques are required. Current signature analysis (CSA) has emerged as a powerful tool for this purpose, despite some limitations. In this work, the variability of CSA through time by means of spectra waterfall and the wavelet transform are investigated to overcome these limitations. The analysis of a faulty in-service WT DFIG presented in this work raises the usefulness of the suggested approaches to identify patterns related to failure development through time.

An effective statistical fault detection technique for grid connected photovoltaic systems based on an improved generalized likelihood ratio test

This paper proposes an improved statistical failure detection technique for enhanced monitoring capabilities of PV systems. The proposed technique offers reduced false alarm and missed detection rates compared to the generalized likelihood ratio test (GLRT) by taking into consideration the nature variance of the GLRT statistics and applying a multiscale representation. The multiscale nature of the data provides better robustness to noises and better monitoring quality. The effectiveness of the proposed multiscale weighted GLRT (MS-WGLRT) method in detecting failures is evaluated using a set of synthetic and simulated PV data where the developed chart is used for detecting single and multiple failures (e.g., Bypass, Mix and Shading failures). Moreover, a set of real-data was used in order to prove the effectiveness of the proposed technique in detecting partial shading faults. All results show that the MS-WGLRT method offers better fault detection performances compared to the classical WGLRT and conventional GLRT charts.

Failure detection technique under random fatigue loading by machine learning and dual sensing on symmetric structure

To avoid the sudden failure of mechanical structures under repeated loading and to supplement conventional methods of managing life time, we propose a failure detection technique under random fatigue loading using machine learning and dual sensing on a symmetric structure. The state of a shackle, which is used to connect the cargo to the hoist efficiently, under fatigue loading was collected using two strain sensors of a dual system. The strains were preprocessed and labeled as normal or abnormal. Logistic regression machine learning was employed to determine the decision boundary line. Then, we gathered the decision boundary lines of each experiment for determining the time of failure, and we verified every experiment with the most conservative decision boundary line. The results indicate that failure was detected before the crack occurred and the time to notice maintenance could be controlled.

Structural feedback linearization based on nonlinearities rejection

In this paper, a structural feedback linearization technique is proposed. This is a quite simple and effective linear control scheme based on failure detection techniques. Our proposed linear control approach is intended to reject the nonlinearities, which are treated as failure signals affecting the systems dynamics. The proposed control methodology is illustrated via the attitude control of a quadrotor in hover flying.

Model Checking-based Software-FMEA: Assessment of Fault Tolerance and Error Detection Mechanisms

Failure Mode and Effects Analysis (FMEA) is a systematic technique to explore the possible failure modes of individual components or subsystems and determine their potential effects at the system level. Applications of FMEA are common in case of hardware and communication failures, but analyzing software failures (SW-FMEA) poses a number of challenges. Failures may originate in permanent software faults commonly called bugs, and their effects can be very subtle and hard to predict, due to the complex nature of programs. Therefore, a behavior-based automatic method to analyze the potential effects of different types of bugs is desirable. Such a method could be used to automatically build an FMEA report about the fault effects, or to evaluate different failure mitigation and detection techniques. This paper follows the latter direction, demonstrating the use of a model checking-based automated SW-FMEA approach to evaluate error detection and fault tolerance mechanisms, demonstrated on a case study inspired by safety-critical embedded operating systems.

Using a Hybrid Cost-FMEA Analysis for Wind Turbine Reliability Analysis

Failure mode and effects analysis (FMEA) has been proven to be an effective methodology to improve system design reliability. However, the standard approach reveals some weaknesses when applied to wind turbine systems. The conventional criticality assessment method has been criticized as having many limitations such as the weighting of severity and detection factors. In this paper, we aim to overcome these drawbacks and develop a hybrid cost-FMEA by integrating cost factors to assess the criticality, these costs vary from replacement costs to expected failure costs. Then, a quantitative comparative study is carried out to point out average failure rate, main cause of failure, expected failure costs and failure detection techniques. A special reliability analysis of gearbox and rotor-blades are presented.

Improving Displacement Number and Overheads of DRFN using Artificial Bee Colony Technique in WSNs

Failure in Wireless Sensor Networks is common due to deployment of sensor nodes in harsh or hostile environment with limited power backup. Node failures could degrade the efficiency of sensor networks. Thus, failure detection and recovery techniques are very crucial for effective performance of nodes in Wireless Sensor Networks. In this paper we presented an improved DRFN technique by improving its failover mechanism. The failure handling scheme will be improved by using Artificial Bee Colony (ABC) based optimization technique. We analyze existing DRFN technique and compare it with the proposed technique on the basis of performance metrics such as occurrence of displacements and displacement overheads. Experimental results show that the proposed failure detection and recovery technique outperform over the available technique. The presented scheme is implemented and analyzed in 2013 version of MATLAB simulation tool.

Adaptive failure detection in low power lossy wireless sensor networks

In this paper, we investigate the use of failure detectors (FD) in Wireless Sensor Networks (WSN). We provide a classification of FD with respect to some WSN criteria. The focus will be on energy depletion and lossy links. We then propose a new general FD model tailored to WSN constraints, called Adaptive Neighborhood Failure Detector for Low-power lossy WSN (AFDEL). AFDEL provides adaptive local failure detection robust against packet loss (intermittent failures) and saves the use of energy, bandwidth and memory storage. Furthermore, we introduce in AFDEL model an adaptive timer strategy. This strategy offers the possibility to customize the dynamic timer pattern with respect to application requirements in terms of completeness and accuracy. We illustrate the use of this strategy by proposing three failure detection techniques based on AFDEL general model. As a part of this work, we give a stochastic based approach for timer adaptation. We evaluate all techniques using implementation on MiXim/Omnet++ framework. The overall experiments show that AFDEL achieves better trade-off between accuracy/completeness and detection/recovery period compared to static timer approach FaT2D (Benhamida and Challal, 2010) and HeartBeat timer-free technique (Aguilera et al., 1997). Moreover, using adaptive timer strategy in local interaction/detection makes it possible to reduce energy consumption and the overall exchanged data overhead.

Cluster Based Failure Detection and Recovery Technique for Wireless Body Area Networks

In Wireless Body Area Networks (WBANs), the extremely sensitive data transmission mainly requires fault tolerance and consistent data transmission. In this study, we propose a Cluster Based Failure Detection and Recovery Technique for WBAN which presents a hierarchical architecture. Here, local nodes are connected to their Cluster Heads (CHs). Each CH is interconnected and also connected with a Wireless Local Gateway (WLG). Finally, WLG is connected to Hospital Gateway (HG). Each local sensor collects the fault related information and each node is assigned with priority to measure the fault tolerant level of individual nodes. Nodes with high priority are processed first. Further, node and CH level faulty node detection and recovery schemes are also proposed. Our technique provides both reliability and fault tolerance. The efficiency of our technique is proved through simulation results.

EEMD-based wind turbine bearing failure detection using the generator stator current homopolar component

Failure detection has always been a demanding task in the electrical machines community; it has become more challenging in wind energy conversion systems because sustainability and viability of wind farms are highly dependent on the reduction of the operational and maintenance costs. Indeed the most efficient way of reducing these costs would be to continuously monitor the condition of these systems. This allows for early detection of the generator health degeneration, facilitating a proactive response, minimizing downtime, and maximizing productivity. This paper provides then an assessment of a failure detection techniques based on the homopolar component of the generator stator current and attempts to highlight the use of the ensemble empirical mode decomposition as a tool for failure detection in wind turbine generators for stationary and non-stationary cases.

Asynchronous failed sensor node detection method for sensor networks

SUMMARYDetecting failed nodes is important in a sensor network, because a failed node can cause coverage and connectivity holes. Since sensors are resource‐limited devices, lack of a clocking mechanism imposes difficulties in implementing synchronous failure detection techniques. To counter this, asynchronous methods are needed for detecting failed sensor nodes. For example, in traditional distributed systems sent data packets and received acknowledgements are matched to identify any malfunctions. A similar approach can be adopted in sensor networks. An asynchronous failed sensor node detection (AFSD) method is proposed, where a numeric counter variable is used to track the received and sent data packets between active nodes. This variable acts as bounded for an active sensor node and unbounded for an out‐of‐communication or failed node. By observing the value of the counter, a decision can be made on a deployed sensor node. Whenever the counter value of a neighbour crosses a predefined threshold, that neighbour is suspected. However, measuring a counter value can always lead to a false positive. To avoid false suspicion, a consensus protocol is needed. Such a technique is described, which is proved as accurate and complete. The complexity analysis shows that the control, energy, and time overheads of the proposed method are linear and in the order of number of neighbours per node. Copyright © 2011 John Wiley & Sons, Ltd.