Early Fault Detection Research Articles

Early detection of PMSM faults at static and dynamic operating points by using artificial neural network for automated electric vehicle

Early detection of PMSM faults at static and dynamic operating points by using artificial neural network for automated electric vehicle

An improved detection method of GNSS faults with fractional information divergence

Early detection of faults guarantees the integrity of navigation systems. The standard receiver autonomous integrity monitoring (RAIM) incorporated with the Kullback-Leibler divergence (KLD) has been proposed to improve detection performance. However, the KLD-RAIM cannot ensure a high detection rate in the presence of small or growing faults. This paper broadens the existing KLD-RAIM to further enhance the detection ability for satellite navigation-based aviation applications. The Cumulative residual KL information (CRKL) replaces the KLD to quantify the dissimilarity between the current and normal cumulative distribution functions for the CRKL-RAIM construction. Then CRKL is extended to fractional orders using Riemann–Louville (RL) calculus to form fractional CRKL-RAIM for detection sensitivity enhancement. Simulations using static GNSS station data were conducted to compare four methods: the standard RAIM, existing KLD-RAIM, the proposed CRKL-RAIM, and fractional CRKL-RAIM. When dealing with small step faults, the detection sensitivity increases progressively, with the fractional CRKL-RAIM capable of detecting 10-meter faults. Regarding growing faults, the standard RAIM fails to detect and KLD-RAIM barely detects faults around 600 s with 50% missed detection rate. By contrast, CRKL-RAIM and fractional CRKL-RAIM detect faults around 100 and 200 s in advance, respectively, without any missed detection events.

Wideband Vibro-Acoustic Modulation for Crack Detection in Wind Turbine Blades

Wind turbines (WT) are a popular method used in energy production, but blade failure and maintenance costs pose significant challenges for the industry. Early detection of blade defects is vital to prevent collapse. This paper examines the modulation of blade vibrations via low-frequency blade rotation, mirroring the vibro-acoustic modulation (VAM) method. Specifically, we study the modulation of blade vibrations, which are generated via blade interactions with air turbulence and have a wide frequency range. These vibrations are modulated by the alternating bending stress experienced during blade rotation. For the simulation of VAM, we employ a simple breathing crack model, which considers a mechanical oscillator with parameters that are periodically changed in response to low-frequency blade rotation. The modulation of the wideband signal by blade rotation can be extracted using the detection of envelope modulation on noise (DEMON) algorithm. This model was applied for the estimation of the modulation of a large (52-m-long) WT blade. Steel specimens have been used in laboratory experiments to demonstrate the feasibility of VAM using a probe broadband noise signal. This paper presents the first work to experimentally and theoretically apply wideband signals in VAM. It further explores the analysis of the use of natural vibrations within VAM for the SHM of WT blades.

Reduction of False Positives for Runtime Errors in C/C++ Software: A Comparative Study

In software development, early defect detection using static analysis can be performed without executing the source code. However, defects are detected on a non-execution basis, thus resulting in a higher ratio of false positives. Recently, studies have been conducted to effectively perform static analyses using machine learning (ML) and deep learning (DL) technologies. This study examines the techniques for detecting runtime errors used in existing static analysis tools and the causes and rates of false positives. It analyzes the latest static analysis technologies that apply machine learning/deep learning to decrease false positives and compares them with existing technologies in terms of effectiveness and performance. In addition, machine-learning/deep-learning-based defect detection techniques were implemented in experimental environments and real-world software to determine their effectiveness in real-world software.

Early Bearing Fault Detection Using EEMD and Three-Sigma Rule Denoising Method

Rotating electrical machines have several physical phenomena. Vibration is one of the important phenomena in the operation of rotating electrical machines. In addition, the vibration signal is considered an important source to have good information on the state of rotating electrical machinery. But this signal is rich in noise, especially under the presence of the bearing fault. This paper proposes a bearing fault diagnosis method based on EEMD and a denoising method based on three-sigma rule. In the first step, the EEMD decomposed the vibration signal into several components called Intrinsic Mode Functions (IMFs). After the calculation of the kurtosis of each IMF component, the signal is reconstructed by choosing components with higher values. To enhance periodic impulses, the three-sigma rule de-noising is applied to the reconstructed signal. As a final step, the envelope spectrum is used to determine the fault characteristic frequency. As a result of testing the bearing with inner race fault and the bearing with outer race, it was verified that the proposed approach suppressed noise effectively and extracted rich fault information from the vibration signals of bearings compared to the EEMD.

Guidelines for designing a digital twin for Li-ion battery: A reference methodology

The integration of digital technologies is causing a significant change in the energy sector. These innovations have transformed traditional energy grids into intelligent grids. As a result, the digital replica of Battery Energy Storage Systems (BESS) has become one of the most crucial components in the energy sector. Digital twin technology enables the seamless integration of BESS into intelligent grids and offers numerous benefits, such as easy identification and prediction of faults, real-time system monitoring, optimization, temperature regulation, and estimation of parameters. As a result, the overall performance of BESS is improved by the digital twin technology. Consequently, this paper discusses the general guidelines that must be followed to develop a digital twin for a Li-ion BESS successfully. The main function is to define how to design a digital twin able to optimize the system and facilitate early and predictive fault detection and diagnosis.

A novel semi-supervised fault detection and isolation method for battery system of electric vehicles

The detection and isolation of early and minor faults in vehicle battery systems is vital to safe driving and improving power utilization. This paper proposes a data-driven model to achieve accurate, early, and economical fault detection and isolation. The model is based on kernel principal component analysis (KPCA), which maps complex nonlinear data from the input space into a high-dimensional feature space to gain a detection model with good performance. To overcome the difficulty of hyperparameter selection, KPCA is trained using Bayesian Optimization (BO) iterations with a small amount of labeled data and a large amount of unlabeled data. This step can obtain the optimal hyperparameter to greatly improve the model fault detection capability, which is beneficial for detecting both early faults and minor faults. In addition, a unified contribution graph based on the partial differentiation of KPCA was adopted to build a reasonable isolation scheme. The semi-supervised model of KPCA based on Bayesian Optimization and contribution graph is developed to reveal the relationship between fault and variable. Finally, the proposed method is fully tested on four fault datasets and the results prove the excellent detection capability in the early stage of faults compared with other methods and the accurate fault isolation capability from the occurrence to the end of the fault.

Time-Series Transfer Learning: An Early Stage Imbalance Fault Detection Method Based on Feature Enhancement and Improved Support Vector Data Description

Early stage fault detection plays a pivotal role in Industrial equipment accidents avoidance and scientific maintenance. While limited by the complex operation background, its application encounters with the conundrum of fault feature indistinctness. To address the challenge, a time-series transfer learning (TSTL) method is proposed, which contains two phases: first, early stage series are transferred to their corresponding serious stage for fault feature enhancement. Moreover, due to the improvement of model structure and loss function, the limitation of mismatched working condition is well-weaken. Second, a transferred fault mode recognition model is trained by using transferred normal series that provides a novel solution for data imbalance. Finally, the TSTL method is verified by actual vibration datasets of power pole tower bolts. Its superiority in feature transfer and fault detection is confirmed by several groups of comparative experiments and results demonstrate TSTL outperforms mainstream methods.

Revolutionizing condition monitoring techniques with integration of artificial intelligence and machine learning

The field of condition monitoring has made great strides in the development of innovative techniques that facilitate early detection and diagnosis of faults in equipment. Present article provides a comprehensive overview of condition-monitoring technologies, namely, ultrasound technique and radiography. By providing detailed insight into the strengths and weaknesses of mentioned techniques, present work highlights their suitability for various industrial applications and how artificial intelligence and machine learning can be integrated to form multimodal systems that improve overall performance. Additionally, it evaluates developments in artificial intelligence and machine learning that enable accurate and reliable condition monitoring systems.

Crack detection of a modified high contact ratio gear

For the purpose of vibration-based fault detection and to avoid gear failure consequences it is necessary to investigate the possibility of early fault detection. A reduction in the gear time-varying mesh stiffness (TVMS) due to the existence of a crack can be used to detect and evaluate tooth damage. The reduction in TVMS has an impact on the system’s dynamic response signal, and then the used fault detection indicators are affected. The current article sheds light on the fault detection possibility of high contact ratio (HCR) gears as compared to those standard ones of low contact ratio (LCR). The influence of modifying the gear tooth addendum dimension for obtaining a high contact ratio on the possibility of early fault detection is studied. Twenty different crack cases are examined by using both the LCR and HCR design cases. For every crack case, the mesh stiffness is evaluated and introduced to the used gear dynamic model for obtaining the system’s dynamic response. The TVMS evaluation is presented for both the LCR and HCR gear cases. Four fault detection indicators, namely the RMS, kurtosis, peak value and crest factor are applied as different measures to enhance the research outcome. The results show that the studied indicators will have less increase when the crack size increases in the HCR design than in the LCR one. Therefore, the HCR design will be less sensitive or more difficult to detect from the fault detection perspective. Lowering the amount of signal noise is applied to the HCR case to examine the impact on enhancing the obtained indicators. The four applied indicators show different behaviours with reducing noise.

Data-driven condition monitoring of two-stroke marine diesel engine piston rings with machine learning

ABSTRACT Maintaining the condition of a vessel and its equipment guarantees the scheduled completion of voyages and the safety of the crew. This paper presents condition monitoring techniques for early detection of faults related to piston rings in remote cylinders of two-stroke marine diesel engines. Operational sensor data from the main engine of a container ship are provided by a shipping company. A graphical approach complimented by correlation heatmaps and feature importance from gradient boosting trees are used for feature selection. Support Vector Machine, Random Forest and Extreme Gradient Boosting Trees are tested for residual generation from the nominal behavior. The residual time series gives a good indication of the degradation of the system and can be used for alarm raising under strict rules. It is proven that the proposed method could alert the engine crew of a change in the condition of the piston rings much earlier than existing methods.

Wind Turbine Gearbox Diagnosis Based on Stator Current

Early detection of faults in wind energy systems can reduce downtime, operating, and maintenance costs while increasing productivity. This paper proposes a method based on the analysis of generator stator current signals to detect faults in a wind turbine gearbox equipped with a doubly fed induction generator (DFIG). A localized parameter model was established to simulate the vibratory response of a two-stage gear system under healthy and faulty conditions. The simulation was performed in the MATLAB/Simulink environment. The results include a detailed analysis of the mechanical part of the gearbox, highlighting mesh stiffness, output speed, and accelerations. Additionally, the electrical part was evaluated based on the current supplied by the doubly fed induction generator. The results were presented in the case of healthy gears and in the presence of faults such as a broken or cracked tooth. Fast Fourier transform (FFT) analysis was employed to detect gear defects in the stator current signal. The presence of a crack or broken tooth in the gearbox induces modulation of the DFIG stator current signals according to the shaft frequencies corresponding to the faulty gear. These findings provide a preliminary basis for the detection and diagnosis of this type of failure.

A Software Verification Method for the Internet of Things and Cyber-Physical Systems

With the proliferation of the Internet of Things devices and cyber-physical systems, there is a growing demand for highly functional and high-quality software. To address this demand, it is crucial to employ effective software verification methods. The proposed method is based on the use of physical quantities defined by the International System of Units, which have specific physical dimensions. Additionally, a transformation of the physical value orientation introduced by Siano is utilized. To evaluate the effectiveness of this method, specialized software defect models have been developed. These models are based on the statistical characteristics of the open-source C/C++ code used in drone applications. The advantages of the proposed method include early detection of software defects during compile-time, reduced testing duration, cost savings by identifying a significant portion of latent defects, improved software quality by enhancing reliability, robustness, and performance, as well as complementing existing verification techniques by focusing on latent defects based on software characteristics. By implementing this method, significant reductions in testing time and improvements in both reliability and software quality can be achieved. The method aims to detect 90% of incorrect uses of software variables and over 50% of incorrect uses of operations at both compile-time and run-time.

Wind Turbine Blade Early Fault Detection With Faulty Label Unknown and Labeling Bias

In practical industrial applications, the need for a large number of accurately labeled training samples is a significant challenge for fault detection tasks. However, labeling all training samples is expensive and prone to labeling errors, especially for early fault detection of wind turbine blades. This paper proposes a labeling bias (LB) hypothesis. Assuming the labeler only needs to label parts of normal samples that are easy to judge, we design a probability ratio least-squares importance fitting (PRL-SIF) method based on variable homogeneity. Unlike other state-of-the-art positive unlabeled (PU) learning methods, PRL-SIF does not require knowledge of the class priors to achieve training. Furthermore, to better handle the multi-dimensional time-series data of wind turbines, we provide a data preprocessing method based on functional analysis to achieve time series feature extraction and dimensionality reduction. The effectiveness and robustness of the proposed method are verified on 23 real-world wind turbine datasets. Experimental results show that the proposed method can achieve nearly 90% accuracy while only labeling 20% of normal samples.

The Assessment of White Matter Integrity Alteration Pattern in Patients with Brain Tumor Utilizing Diffusion Tensor Imaging: A Systematic Review.

Alteration in the surrounding brain tissue may occur in the presence of a brain tumor. The present study aims to assess the characteristics and criteria of the pattern of white matter tract microstructure integrity alteration in brain tumor patients. The Scopus, PubMed/Medline, and Web of Science electronic databases were searched for related articles based on the guidelines established by PRISMA. Twenty-five studies were selected on the morphological changes of white matter tract integrity based on the differential classification of white matter tract (WMT) patterns in brain tumor patients through diffusion tensor imaging (DTI). The characterization was based on two criteria: the visualization of the tract-its orientation and position-and the DTI parameters, which were the fractional anisotropy and apparent diffusion coefficient. Individual evaluations revealed no absolute, mutually exclusive type of tumor in relation to morphological WMT microstructure integrity changes. In most cases, different types and grades of tumors have shown displacement or infiltration. Characterizing morphological changes in the integrity of the white matter tract microstructures is vital in the diagnostic and prognostic evaluation of the tumor's progression and could be a potential assessment for the early detection of possible neurological defects that may affect the patient, as well as aiding in surgery decision-making.

Early fault diagnosis in rolling element bearings: comparative analysis of a knowledge-based and a data-driven approach

The early identification of a defect that is developing in a bearing is crucial for avoiding failures in rotating machinery. Frequency domain analysis of the vibration signals has been shown to contribute to a better understanding of the nature of a developing defect. Early signs of degradation might be more noticeable in certain frequency bands. The advantages in identifying and monitoring these bandwidths are several: prevention of serious machinery damages, reduction of the loss of investments, and improvement of the accuracy in failure predicting models. This paper presents and compares two approaches for the diagnosis of bearing faults. The first approach was knowledge-based. It relied on principles of mechanics to interpret the measured vibration signals and utilized prior knowledge of the bearing characteristics and testing parameters. The second approach was data-driven whereby data were acquired exclusively from the vibration signal. Both approaches were successfully applied for fault diagnosis by identifying the frequencies of the vibration spectra characteristic for the bearing under study. From this, bandwidths of interest for early fault detection could be determined. The diagnostic abilities of both approaches were studied to analyze and compare their individual strengths regarding the aspects of implementation time, domain knowledge, data processing associated knowledge, data requirements, diagnostic performance, and practical applicability. The advantages, apparent limitations as well as avenues for further improvement of both approaches are discussed.

Wasserstein GAN-Based Digital Twin-Inspired Model for Early Drift Fault Detection in Wireless Sensor Networks

In this Internet of Things (IoT) era, the number of devices capable of sensing their surroundings is increasing day by day. Based on the data from these devices, numerous services and systems are now offered where critical decisions depend on the data collected by sensors. Therefore, error-free data are most desirable, but due to extreme operating environments, the possibility of faults occurring in sensors is high. So, detecting faults in data obtained by sensors is important. In this paper, a digital twin inspired detection approach is proposed, and its ability to detect a single type of fault in several sensor is analyzed. The digital equivalent of the sensor is developed using a Generative Adversarial Network (GAN). As GANs inherently performs well with images, Gramian Angular Field (GAF) encoding is used to convert timeseries data to image. The GAF encoding preserves the temporal relations of the timeseries data. The GAN is trained with the GAF images. The trained GAN model acts as the virtual representation of the sensor, and the discriminator network of the GAN model, once it has learned the pattern of normal data, is used as the fault detector. The performance of the virtual sensor is promising because it successfully generates data for normal conditions. The best fault detection accuracy achieved by the proposed model is 98.7%, which makes this GAN-based digital twin inspired approach a promising candidate for sensor fault detection.

Developing a Hybrid Algorithm Based on an Equilibrium Optimizer and an Improved Backpropagation Neural Network for Fault Warning

In today’s rapidly evolving manufacturing landscape with the advent of intelligent technologies, ensuring smooth equipment operation and fostering stable business growth rely heavily on accurate early fault detection and timely maintenance. Machine learning techniques have proven to be effective in detecting faults in modern production processes. Among various machine learning algorithms, the Backpropagation (BP) neural network is a commonly used model for fault detection. However, due to the intricacies of the BP neural network training process and the challenges posed by local minima, it has certain limitations in practical applications, which hinder its ability to meet efficiency and accuracy requirements in real-world scenarios. This paper aims to optimize BP networks and develop more effective fault warning methods. The primary contribution of this research is the proposal of a novel hybrid algorithm that combines a random wandering strategy within the main loop of an equilibrium optimizer (EO), a local search operator inspired by simulated annealing, and an adaptive learning strategy within the BP neural network. Through analysis and comparison of multiple sets of experimental data, the algorithm demonstrates exceptional accuracy and stability in fault warning tasks, effectively predicting the future operation of equipment and systems. This innovative approach not only overcomes the limitations of traditional BP neural networks, but also provides an efficient and reliable solution for fault detection and early warning in practical applications.

Fault Detection and Identification with Kernel Principal Component Analysis and Long Short-Term Memory Artificial Neural Network Combined Method

A new fault detection and identification approach is proposed. The kernel principal component analysis (KPCA) is first applied to the data for reducing dimensionality, and the occurrence of faults is determined by means of two statistical indices, T2 and Q. The K-means clustering algorithm is then adopted to analyze the data and perform clustering, according to the type of fault. Finally, the type of fault is determined using a long short-term memory (LSTM) neural network. The performance of the proposed technique is compared with the principal component analysis (PCA) method in early detecting malfunctions on a continuous stirred tank reactor (CSTR) system. Up to 10 sensor faults and other system degradation conditions are considered. The performance of the LSTM neural network is compared with three other machine learning techniques, namely the support vector machine (SVM), K-nearest neighbors (KNN) algorithm, and decision trees, in determining the type of fault. The results indicate the superior performance of the suggested methodology in both early fault detection and fault identification.

Early Fault Detection in Particle Accelerator Power Electronics Using Ensemble Learning

Early fault detection and fault prognosis are crucial to ensure efficient and safe operations of complex engineering systems such as the Spallation Neutron Source (SNS) and its power electronics (high voltage converter modulators). Following an advanced experimental facility setup that mimics SNS operating conditions, the authors successfully conducted 21 early fault detection experiments, where fault precursors are introduced in the system to a degree enough to cause degradation in the waveform signals, but not enough to reach a real fault. Nine different machine learning techniques based on ensemble trees, convolutional neural networks, support vector machines, and hierarchical voting ensembles are proposed to detect the fault precursors. Although all 9 models have shown a perfect and identical performance during the training and testing phase, the performance of most models has decreased in the next test phase once they got exposed to realworld data from the 21 experiments. The hierarchical voting ensemble, which features multiple layers of diverse models, maintains a distinguished performance in early detection of the fault precursors with 95% success rate (20/21 tests), followed by adaboost and extremely randomized trees with 52% and 48% success rates, respectively. The support vector machine models were the worst with only 24% success rate (5/21 tests). The study concluded that a successful implementation of machine learning in the SNS or particle accelerator power systems would require a major upgrade in the controller and the data acquisition system to facilitate streaming and handling big data for the machine learning models. In addition, this study shows that the best performing models were diverse and based on the ensemble concept to reduce the bias and hyperparameter sensitivity of individual models.