Fault Diagnostics Research Articles

Energy and battery management systems for electrical vehicles: A comprehensive review & recommendations

Electric vehicle technology has recently drawn a lot of interest on a global scale due to improved performance in its efficiency and the capability to solve the problems of carbon emission. As such, electric vehicles are the key to achieving sustainable development goals. This review article analyzes deeply the previous technical developments of electric vehicles, focusing on important topics like battery management systems, technologies of power electronics, techniques of charging, and the relevant algorithms and improvements. In addition, several critical problems, and difficulties are presented in order to pinpoint the gaps in the literature. To address the analysis of battery behavior, battery condition monitoring, real-time control design, temperature control, fault diagnostics, and efficiency of battery model are considered. This study highlighted the estimation techniques that predict the internal battery conditions such as internal temperature, state of health, and state of charge, which are difficult to be directly monitored and determined. A lithium-ion battery, a super-capacitor, and related bidirectional DC/DC converters constitutes the infrastructure of a hybrid power system. This review offers useful and practical recommendations for the future development of electric vehicle technology which in turn help electric vehicle engineers to be acquainted with effective techniques of battery storage, battery charging strategies, converters, controllers, and optimization methods to satisfy the requirements of sustainable development goals. Accordingly, this review article will be a platform and future guide for those who are interesting in the field of energy management and its development.

Intelligent faults diagnostics of turbine vibration’s via Fourier transform and neuro-fuzzy systems with wavelets exploitation

ABSTRACT Gas turbines play a vital role in gas transportation and power generation, but they are prone to instability phenomena that can lead to vibrations, shorten equipment lifespan, and result in catastrophic failures. To tackle these challenges, a paper introduces an integrated approach that leverages advanced techniques like Fourier transform, Neuro-Fuzzy systems, and wavelet analysis for continuous monitoring of the MS5002C turbine’s condition. The proposed method begins by collecting operational data and utilizing the Fourier transform to measure vibratory quantities, accurately representing their evolution through spectral data obtained from the analyzed signals. Adaptive inference-based algorithms of neuro-fuzzy systems are then employed to generate turbine failure indicators. This approach enables the development of a model-based fault detection method that compares the actual turbine operation with the estimated operation derived from a pre-established model, enabling the classification of detected faults. To enhance decision-making quality, evaluation, and validation of the diagnostic strategy’s performance, a multi-resolution analysis based on the wavelet transform is applied. The presented results from various implementation and validation tests demonstrate the effectiveness of this intelligent diagnostic approach in detecting and analyzing gas turbine vibrations. The paper exhibits promising outcomes in real-time monitoring, ensuring the operational safety of the turbine.

Quantitative Fault Diagnostics of Hydraulic Cylinder Using Particle Filter

Condition-based hydraulic cylinder maintenance necessitates quantitative fault diagnostics. However, existing methods are characterized by either qualitative or limited quantitative capabilities. In this paper, a quantitative fault diagnostic method using a particle filter for hydraulic cylinders is proposed. The problem of quantitative fault diagnostics is formally formulated in a stochastic framework to assess the health/fault state, and an architecture based on joint state-parameter estimation is proposed. Through the establishment and analysis of a nonlinear dynamic model of the hydraulic cylinder, the impact of time-varying parameters on the state variables is revealed. Three fault modes of the cylinder, including friction, internal leakage, and external leakage, are theoretically identified. The proposed method allows for a simultaneous quantitative diagnosis of these three fault modes. The performance of the proposed method is evaluated using meticulously designed experiments. The results demonstrate that the mean absolute percentage errors in the parameter estimations are below 9% (accuracy exceeding 91%), thus validating its feasibility and effectiveness.

Utilizing MATLAB machine learning models to categorize transient events in a nuclear power plant using generic pressurized water reactor simulator

Enhancing safety and dependability within nuclear power facilities holds paramount importance in safeguarding both individuals and the environment. The adoption of machine learning for diagnosing faults in these plants is steadily gaining interest, driven by its capacity to detect faults, alleviate human errors in high-pressure scenarios, and ensure the secure and consistent operation of these facilities swiftly and accurately.This paper examines the use of machine learning models for fault diagnostics, specifically, the identification of transient events in a nuclear power plant to reduce human errors. The data was collected from WSC’s Generic Pressurized Water Reactor (GPWR) simulator and processed using MATLAB. The simulator encompasses models for both the primary and secondary systems of the nuclear power plant (NPP). Additionally, it incorporates models for the control systems and instrumentation responsible for monitoring and regulating the reactor, serving as integral components for data extraction and transient modeling. A total of 9 different transient events were simulated with 12 different initial conditions to create a dataset with 72,000 observations. Nine types of classification models (33 total preset models) were trained and validated using the classification learners application. Among them, the Neural Network Classifiers (NNC) displayed the highest average accuracy of 90%. The Fine Tree, Ensemble Bagged Trees, and Medium Neural Network models were the best-performing individual models with validation accuracies above 90% and a maximum training time of 8 min. These models were further analyzed using accuracy, confusion matrix, precision, recall, and F1 score. To optimize these models, techniques such as different validation schemes and feature selection were utilized to further reduce their training time and improve their prediction accuracy. The optimized models boasted comparable accuracies with a maximum training time of under 1.5 min. The results of this study exhibited favorable comparisons with other machine-learning endeavors in the field of reactor transient detection and diagnostics. Notably, the study maintained low execution and computation times while preserving high levels of accuracy. This study offers insightful information on how AI and machine learning can be used to improve nuclear power plant diagnostics, enhance safety, and provide support to the operator.

A comparative investigation between rule- and inverse model-based fault detection and diagnostics for HVAC control systems

Fault detection and diagnostics (FDD) tools provide valuable information regarding system faults and deviation from expected operation. Most existing FDD tools apply rule-based fault detection algorithms that generate an alarm when a rule is met; however, these tools cannot evaluate the overall performance of a system. Inverse-model-based FDD algorithms can be deployed to complement the fault alarms triggered by rule-based building energy management systems (BEMS). This paper examines the faults detected by rule- and inverse model-based algorithms used to detect faults in multiple zone variable air volume air handling unit systems. The capability of the rule- and inverse model-based algorithms in detecting and diagnosing faults is demonstrated through illustrative examples using data from three commercial buildings in New Brunswick, Canada. The results show that inverse model-based algorithms could diagnose faults that were not detected by the rule-based FDD algorithms implemented in a commercially available BEMS tool.

A new method to perform Lithium-ion battery pack fault diagnostics – Part 1: Algorithm development and its performance analysis

The successful integration of Li-ion batteries in the automotive industry paved the way for their integration into a new generation of aircraft with zero emission. However, potential hazards associated with Li-on batteries are a major concern for the safety-critical aerospace sector. Removing or reducing the hazards from Li-ion batteries while preferable, requires fundamental changes to the battery chemistry. Another way to address this is the implementation of fault diagnostics and prognostics for forecasting or detecting the presence of faults before an aircraft is airborne. This article, part of a three-part series, presents a novel algorithm exploiting the data collected while charging a battery pack for diagnosing faults. The algorithm works off-board, thus not adding any computational burden or weight to the aircraft while minimising aircraft turn-around time, which is crucial for commercial flight operation. This paper presents the underlying method of the algorithm involving the computation of resistance, charge energy and incremental capacity curve using the charging data followed by a combined horizontal and vertical comparative scheme for diagnosis of faults. The performance of the proposed algorithm was validated via module-level experiments considering three cell inactivation faults. Results indicated that the algorithm successfully segregated faulty and unfaulty cells in multiple test scenarios with different number of faults in a laboratory-assembled module. Upon testing the algorithm in experiments conducted using a real-life aerospace battery module, there was a slight reduction in the performance, attributed to greater thermal variation. Nevertheless, the results indicated promising prospects for practical implementation in eVTOL batteries.

Transfer learning methodology for machine learning based fault detection and diagnostics applied to building services

Machine Learning (ML) models for Fault Detection and Diagnosis (FDD) can automatically detect anomalies in the operation in large facilities or district heating networks and can help tackling energy wastes. Nevertheless, the development of ML-models is a costly and tedious task requiring large amounts of labelled data. Setting up ML-models for a high number of systems is effort and know-how intensive. However, assets like commercial buildings and district heating networks are constituted of systems with similar topologies. Transferring a ML model initially trained on a source system to a multitude of similar target systems, can help reducing the training costs and facilitating the scalability of ML-based FDD in those assets. To enable this, we have developed a methodology that assesses the potential for Transfer Learning (TL) from a source system to target systems by determining the covariate and concept shifts between the source and target domains and integrating the source model into the target system if the TL assessment is positive. We used a patented method for the model development, that combines two ML-models, that are initially trained on a source system by means of a feedback system. We implemented this methodology on district heating (DH) substations, as DH systems typically contain this kind of subsystems with similar topologies and have thus a high scalability potential for TL. Initial findings showed the effectiveness of TL in adapting the source model to the target domain, resulting in enhanced FDD capabilities with significantly reduced training efforts.

Incrementally Contrastive Learning of Homologous and Interclass Features for the Fault Diagnosis of Rolling Element Bearings

Bearing condition is a non-negligible part of mechanical equipment health monitoring. Most of the existing bearing fault diagnosis methods are based on the premise that all data classes are known and lack the capability of incremental diagnosis of fault modes. However, in engineering practice, the initial monitoring data only provide normal condition, and the subsequent data of different classes of faults are collected gradually. To address this practical problem, we propose incremental contrastive learning of homologous and interclass features for bearing to achieve incremental diagnosis of bearing fault modes from single to multiple classes. Important homologous and interclass features of bearings are first extracted by contrastive learning. The obtained features are then employed to establish a distance threshold for the anomaly diagnosis of subsequent samples. Upon appearing anomalies incrementally up to a certain amount, novel classes are upgraded and fed back to the model. In this way, new class data are treated as incremental learning resources. The proposed method was evaluated using both benchmark bearing and gearbox bearing experiments. Results show supreme diagnostic performance compared to peer state-of-the-art approaches. The present method is intrinsic in extracting homologous and interclass features for practical bearing fault diagnostics.

Vehicle State Monitoring and Fault Detection System for Unmanned Ground Vehicles (UGV) using Markov Models

This research presents a novel fault detection and diagnostics system for unmanned ground vehicles (UGVs) by combining Markov models representing the vehicle's navigation, kinematic behavior, and vehicle dynamics systems. Existing studies do not specifically address the challenges related to UGVs and their complex subsystems or the incorporation of weather and environmental condition data. The proposed system leverages environmental and weather condition data to monitor the UGV's state and detect anomalies in its behavior. By predicting the probability of faults such as collisions, sensor damage, and other malfunctions, the system aims to enhance the safety, reliability, and performance of UGVs. The research will demonstrate the effectiveness of the proposed methodology through case studies and performance evaluation, highlighting its potential application in various real-world scenarios. This work contributes to the ongoing research in prognostics and health management, particularly for autonomous systems, by providing a new approach to fault detection and diagnostics in UGVs.

Increasing Robustness of Data-Driven Fault Diagnostics with Knowledge Graphs

In the realm of PHM, it is common to possess not only process data but also domain knowledge, which, if integrated into data-driven algorithms, can aid in solving specific tasks.This paper explores the integration of knowledge graphs (KGs) into deep learning models to develop a more resilient approach capable of handling domain shifts, such as variations in machine operation conditions.We present and assess a KG-enhanced deep learning approach in a representative PHM use case, demonstrating its effectiveness by incorporating domain-invariant knowledge through the KG.Furthermore, we provide guidance for constructing a comprehensive hierarchical KG representation that preserves semantic information while facilitating numerical representation.The experimental results showcase the improved performance and domain shift robustness of the KG-enhanced approach in fault diagnostics.

Data Augmentation of Sensor Time Series using Time-varying Autoregressive Processes

This work presents a novel data-centric solution for fault diagnostics and failure prognostics consisting of a data-augmentation method which is well suited for non-stationary mutivariate time-series data. The method, based on time-varying autoregressive processes, can be employed to extract key information from a limited number of samples and generate new artificial samples in a way that benefits the development of diagnostics and prognostics solutions. The proposed approach is tested based on three real-world datasets associated with failure diagnostics problems using two types of machine learning methods. Results indicate the proposed method improves performance in all tested cases.

Can pretrained networks be used in fault diagnosis of monoblock centrifugal pump?

Monoblock centrifugal pumps are used daily and play an essential role in agriculture, wastewater treatment, food, paper, gas and oil industries. Extended and endless use of monoblock centrifugal pump can affect the desired pump characteristics, compromising the efficiency of the pump. The condition of a pump is judged based on the high vibration, unusual noise, leakage, etc. Maintaining pump health is essential to increase the productivity in industry. Countless experiments were carried out by researchers to maintain the efficiency of the pump. The main shortcomings of traditional fault diagnosis are the need for a high degree of human intellect and professional knowledge. Such drawbacks have made the researchers provide autonomous and intelligent diagnostic tools. The objective of this article is to investigate the possibility of use of deep learning technique to identify the type of fault present in the monoblock centrifugal pump. First, vibration signals from the monoblock centrifugal pump were acquired and plotted as images that serve as an input for deep learning (DL) algorithms. The deep learning system can analyse the status of the monoblock centrifugal pump and learn from the vibration signal plots. In this paper, pre-trained networks such as ResNet-50, AlexNet, GoogleNet and VGG-16 were used to diagnose the fault in the monoblock centrifugal pump. The outcome of the hyperparameters namely learning rate (LR), batch size (BS), split ratio (SR) and solver were considered and the best performing network was recommended for the monoblock centrifugal pump fault diagnostics. Findings: A higher classification accuracy of 100% with lower training time of 177 s was obtained while using ResNet-50.

Efficient channel attention residual learning for the time-series fault diagnosis of wind turbine gearboxes

Due to harsh and variable working conditions, the wind turbine gearbox may be damaged during the operation, resulting in an extended downtime with reduced productivity and economic loss. This calls for efficient fault diagnostics for the wind turbine gearboxes. Commonly-used diagnostics based on classical deep learning networks cannot guarantee good performance with time series signals due to the weakness of feature extraction. For this reason, an efficient channel attention residual learning approach is proposed to enhance the feature extraction and fault diagnosis for wind turbine gearboxes, leading to the development of a channel attention residual network (CAResNet). The collected time series signals are directly employed as the input. The efficient channel attention is embedded into the residual network to capture features and improve fault diagnosis capability. Experiments are carried out on a real wind turbine gearbox. The results showed that the average diagnostic accuracy of the CAResNet model was 94.41%. CAResNet has better diagnostic accuracy than other methods proposed in this paper.

A new solving method for ball bearing dynamics model based on optimization with nonlinear constraints: Contact angles as iterative variables

Bearing dynamics model is the premise for bearing vibration analysis and fault diagnostics. This paper proposed a novel modeling and solving method for angular contact ball bearing dynamics. It transformed the problem of solving nonlinear model equations into an optimization issue with equality constraints by choosing contact angles rather than deformations as the iteration variables. Additionally, to accelerate the solving convergence and reduce dependence on the initial values, a polynomial model was adopted to approximate and define the contact angle initials by analyzing their value distribution characteristics at different ball positions. Furthermore, a genetic algorithm was applied to reset the initial values at every iteration to speed up the solving process and improve the global search capability. Simulations under various conditions were used to validate the proposed method and compared with the traditional Newton–Raphson solving method. The influence of contact angles on bearing fatigue life was also analyzed. Results showed that the accuracy, efficiency, and dependence on the initial values were thoroughly improved.

Empirical analysis of the prevalence of HVAC faults in commercial buildings

Commercial building HVAC systems experience many sensing, mechanical, and control-related faults that increase energy consumption and impact occupant comfort. Fault detection & diagnostics (FDD) software has been demonstrated to identify and help diagnose these types of faults. Several studies have demonstrated FDD energy savings potential, but there is limited empirical data characterizing the quantity and type of faults reported by FDD tools. This paper presents results of an FDD fault reporting study, employing multi-year monitoring data for over 60,000 pieces of HVAC equipment, covering over 90 fault types, and using new metrics that we developed to characterize fault prevalence. Study results offer an unprecedented accounting of the quantity of faults reported, the most commonly occurring faults, and fault persistence. We find that 21 air handling unit (AHU) faults were reported on 20% or more AHUs in our dataset, and 18 AHU faults persisted for more than 20% of the time period covered by the data. On any given day, 40% of AHUs and 30% of air terminal units saw a reported fault of some kind. Based on in-depth analysis of these results we provide recommendations for building operators, FDD software developers, and researchers to enable more efficient commercial building operation.

Line electrical transmission damage identification tool in distributors electrical networks

In the paper showing that one of the main way for Ukraine electric power industry evolution is electric power networks and systems improvement and construction considering Smart Grid concept. The main idea of Smart grid is the reliable, energy-efficient and high-quality energy supply system. To implement the idea, it is necessary to create a high-performance information and computing infrastructure. The main components of Smart Grid is emergency mode diagnostics and damage monitoring. Providing fault diagnostics and fault monitoring can improve the power supply reliability and quality to consumers. So, the task of quickly and accurately determining the place of damage is important. In the article showing that emergency detection tools in sections of both cable and overhead electrical networks it’s a way to improve efficiency of the networks. The diagram of the placement of damage indicators on the section of the electrical distribution network is presented, which allows determining the direction of the search for the location of the damage. A comparative analysis of measuring current transformers of optical and electromagnetic type was carried out. It is shown that a significant number of advantages of optical measuring current transformers, which can be used in damage indicators, can be provided by measuring current transformers of the electromagnetic type. Creating fault indicator based on the Smart Grid concept lets do to reduce searching time for the cause and location of an emergency situation to a minimum. In this concept application fault indicator in energy grid ensure it connection to operating overhead and cable lines without removing voltage by connecting the output of the secondary winding of the measuring current transformer of the detachable design to the measuring converters directly placed near the current measuring transformers with the help of a mechanical spring fastener. A block diagram of a specialized information-measuring system with a damage indicator was created, taking into account the requirements of the Smart Grid concept, which allows to reduce the time of searching for the cause to a minimum.

Domain Discrepancy-Guided Contrastive Feature Learning for Few-Shot Industrial Fault Diagnosis Under Variable Working Conditions

Recent advances in data-driven methods have significantly promoted intelligent fault diagnostics for varied industrial applications. However, due to the limitations of machine fault data and the varied scenarios in the context of industrial working conditions, existing diagnostic models can hardly achieve satisfactory results. In this study, we propose a domain discrepancy-guided contrastive feature learning framework for few-shot fault diagnosis under varied working conditions. Unlike the conventional contrastive learning paradigm using manually augmented data, a sample pairs construction is implemented based on the differences between domain distributions for data acquired under different working conditions. The similarity contrast learns the domain-invariant features from a small number of sample pairs. The learned fault features can then be used for fault identification without parameter fine-tuning. In two case studies, we validated the performance of the proposed framework with small training samples under varying speeds, loads, and significant noises. Compared with the state-of-the-art methods, the proposed solution achieved higher diagnostic accuracy for the targeted applications.

Fault detection and diagnosis in AHU system using deep learning approach

Energy consumption in buildings increases with the failures of equipment involved in the energy exchange, and control networks in buildings. One of the ways to remedy this issue is to offer high-performance fault detection systems. This article proposes a Fault Detection and Diagnostics (FDD) system based on Convolutional Neural Network (CNN) and Long Term Short Memory (LSTM) neural networks, applied to an AHU and using an hybrid database containing data from simulation and real-world on an actual physical building. The proposed system is designed to effectively identify and categorize faults, whether they occur in the sensors or in the mechanical equipment responsible for critical functions such as heat exchanges, air transfer, and system control. The FDD system provides results with an overall accuracy of around 96.88 %.

A Review of Gas Pumps Fault Behavior and Measurement, Diagnosis and Identification Methods Including Virtual Sensors

Almost half of all electrical pump capacity consume in three-phase induction motors and most of them use in gas pumps and compressor in processes of industries for heating, cooling, pumping, conveyors, etc. The most important role of fault detection and detection (FDD) for manufacturing equipment is an effective indicator that can identify the faulty state of a process and then take appropriate action against future failures or adverse events. The impact of proper maintenance is reflected on an especially costly type of industrial machine. To avoid reducing the efficiency of the gas station, automated fault detection and diagnostics is very important and need to study. This article aims to establish a criterion for selecting which faults can be tested under laboratory conditions or by simulation with a virtual model and to determine the features that identify those faults. This efficiency identify of faults leads to saving of energy, service and operating costs. Virtual sensors applied to gas pumps to reduce the cost associated with FDD implementation are described. Finally, several areas of improvement for the aspects reviewed have been identified: increase the use of performance indicators for FDD, new and updated studies about the health status of field heat pumps, testing methods that take into account the gradual and probabilistic nature of heat pump faults and further research in the use of virtual sensors in FDD systems.

A fault diagnosis method for nuclear power plant rotating machinery based on adaptive deep feature extraction and multiple support vector machines

Rotating machinery is the essential component in nuclear power plants (NPPs). Effective fault detection and diagnosis is a main challenge in the operation and maintenance of NPPs rotating machinery. The performance of traditional fault diagnosis methods mainly depends on complex manual feature extraction and sufficient expert prior knowledge. This study proposes a fault diagnosis method based on adaptive feature extraction and multiple support vector machines (ResNet-SVMs) to overcome the limitations of the traditional intelligent fault diagnostics. First, the vibration information from different locations of the rotating machinery is collected and used as input data for the algorithm model. Then, the deep residual neural network adaptively extracts fault features from input data to obtain feature data of different depths. Finally, multiple support vector machines identify the feature data to realize the fault diagnosis. The effectiveness verification is carried out based on the experimental cases of induction motor and rolling bearing fault diagnosis. Compared with other advanced intelligent fault diagnosis methods, ResNet-SVMs model provides better diagnostic performance, which demonstrates its potential value for NPPs rotating machinery fault diagnosis.