Accuracy Of Fault Prediction Research Articles

Fault prediction of electronic devices based on attention mechanism time-series point process

In recent years, with the development of high-precision sensors, computer technology and artificial intelligence, it makes the data-driven intelligent fault warning and diagnosis method gradually become a research hotspot. Aiming at the low utilization of fault diagnosis information of complex electronic devices and the problems of convergence and training time of traditional neural network fault diagnosis models, a point-in-time process generation model without using intensity function is proposed. The model uses the Wasserstein distance to construct the loss function, which facilitates the measurement of the deviation between the model distribution and the true distribution, and uses a self-concern mechanism to describe the degree of influence of the historical events on the current events, which makes the model interpretable and more capable of generalization. Comparative experiments show that without a priori information about the intensity function, the method reduces the relative error rate by 3.59 %, improves the fault prediction accuracy by 3.91 %, and has a better overall fit than the RNN-like generative model and the great likelihood model. Example analysis shows that the model has good prediction accuracy and provides a feasible solution for real-time fault diagnosis of complex electronic devices.

Intelligent agents for advanced power system protection schemes

This study explores the incorporation of intelligent agents to improve power system safety, using several computational models including machine learning, rule-based systems, neural networks, and fuzzy logic. The research assesses the effectiveness and efficiency of these agents in promptly identifying, categorizing, and responding to faults in the power system architecture using empirical analysis. The results demonstrate the higher performance of agents based on neural networks, with an average improvement in fault prediction accuracy of 38% compared to systems based on rules. Furthermore, the evaluation of power system devices demonstrates a direct relationship between greater voltage ratings and increased expenses for both installation and maintenance, underscoring their crucial importance within the system. An examination of fault severity reveals that greater severity failures have a direct and significant influence on system downtime. These problems lead to longer interruptions, which emphasizes the need of implementing effective fault management systems. Intelligent agents' actions have different costs and reaction times. Actions based on neural networks have lower average costs and shorter response times, demonstrating their cost-effectiveness and efficiency in addressing faults. The study of percentage change highlights the importance of using various kinds of intelligent agents and higher-rated devices. This research offers insights into performance differences and the consequences for optimizing protection measures. This research provides a thorough understanding of how intelligent agents may enhance power system protection. It also offers guidance for future improvements in creating power grid infrastructures that are robust, dependable, and adaptable.

Fault Prediction of Molecular Pump Based on DE-Bi-LSTM

A molecular pump is a high vacuum acquisition piece of equipment that provides a clean vacuum environment for the Experimental Advanced Superconducting Tokamak (EAST) device. Its running state affects the smooth development of the EAST experiment. Because of fatigue degradation of internal components of the molecular pump, vacuum leakage may occur during long-term operation, causing secondary hazards to the device. In order to improve the accuracy of molecular pump fault prediction, based on the long short-term memory network (LSTM), the deep long short-term memory network (DE-LSTM) and the bidirectional long short-term memory network (Bi-LSTM) are combined. The deep bidirectional long short-term memory network (DE-Bi-LSTM) algorithm is proposed, and the piecewise linear degradation model is introduced to predict fault of the molecular pump. By collecting the vibration signals leaked in the atmosphere and running to the fault time series on the destructive test platform simulating molecular pump fault, data were extracted in the time domain. Finally, the obtained feature vector set was used as the input of the DE-Bi-LSTM algorithm through data standardization to train the model and realize the prediction of molecular pump fault. The experimental results show that the proposed method is optimal to LSTM, DE-LSTM, and Bi-LSTM in predicting performance.

Neural network-based fiber optic cable fault prediction study for power distribution communication network

Abstract As the foundation of communication networks, optical fiber carries huge network traffic, so the prediction of fiber optic cable faults is an important guarantee for the operation of communication networks. Based on the combination of fiber optic system networking technology and network management data, this study constructs an alarm correlation analysis method by using data mining technology to obtain the data set of the fault prediction model for the problem of low fault prediction accuracy of traditional communication networks. The dataset is used to balance the sample data by generating a small number of new samples through the generative adversarial network. The memory-based feature generation convolutional network is proposed to enhance the feature interaction to realize fault prediction in communication networks. The prediction model has a high prediction accuracy of 98.68%, which saves about 160 min for repair work through the application of fiber optic cable fault prediction, which compares well with other models. Fault prediction based on neural networks can provide assistance in the operation and maintenance of distribution communication networks.

Design and application of gearbox vibration testing system

Judging the working status of a gearbox through the characteristics of vibration signals was an effective fault diagnosis method. In order to improve the efficiency and reliability of vibration detection, a multi node testing system based on a microcontroller was designed and validated. The system mainly consists of signal processing equipment, signal collector, upper computer, sensors, and experimental workbench. Under different types of gear faults, the vibration signals of ten sets of gearboxes were synchronously collected and processed. In order to ensure the resolution of frequency conversion and accurately locate frequency conversion, the low-pass filtered spectrum with a filtering frequency of 1000 was verified and used to calculate the amplitude of frequency conversion. The normalization of amplitude was more conducive to feature recognition, and the frequency amplitude was proportional to its corresponding energy. According to standard spectral data and spectral analysis, wear, pitting, fracture, and bonding faults in the gearbox were identified and classified. Through the verification of three order feature spectrum, it can be seen that the accuracy of fault prediction for gearbox through vibration signals is high, which can effectively reduce maintenance and repair costs.

A Comprehensive Hybrid Model for Language-Independent Defect Prediction in Microservices Architecture

The transformation of software development from monolithic frameworks to microservices-based architectures, focusing on the challenges of creating a unified defect prediction model that spans various programming languages in practice of automating integration of code modification into a single codebase. It proposes a hybrid machine learning approach to enhance defect prediction accuracy by integrating different data sources and algorithms. The goal is to create a language and project-independent model. The hybrid model combines Bi-Directional LSTM (BiD-LSTM) networks and Attention mechanisms, static code metrics, and BERT-based language models. BiLSTM-Attention captures temporal dependencies within Abstract Syntax Trees (ASTs), static code metrics provide insights into software complexity, and BERT interprets textual context for a holistic understanding of code snippets. The research methodology involves quantitative techniques, starting with a literature review to establish the theoretical foundation. An empirical study follows, encompassing data gathering, feature crafting and pre-processing, model building, training and evaluation, validation and analysis and conclusions. The research’s insights aim to improve defect prediction techniques, contributing to software engineering’s pursuit of better quality and reliability.

Intelligent diagnosis and prediction of turbine digital electro-hydraulic control system faults: Design and experimentation.

A physical modeling approach was adopted to build a Digital Electro-Hydraulic Control (DEH) system simulation model and the fault models using the SIMULINK tool. This research combined the advantages of the gray system and neural network to build a multi-parameter gray error neural network fault prediction model for the first time. Furthermore, an embedded platform for intelligent fault diagnosis and prediction was developed using an Application Specific Integrated Circuit chip. The results show that the simulation model of the DEH system has good performance. A jam fault, internal leakage, and a device fault could be accurately identified through the fault diagnosis model. The multi-parameter gray error neural network prediction model improves the accuracy of fault prediction. The embedded platform developed by the Application Specific Integrated Circuit chip solves the problem of transmission limitation and insufficient computing power. It realizes the intelligent diagnosis and prediction of DEH system faults and guarantees the regular operation of the DEH system.

A novel quantum genetic algorithm with the application in LS-SVR

The characteristics of LS-SVR are analyzed. LS-SVR is fitted for modeling small samples and high dimensional data, but the performance of LS-SVR is related to the specific data distribution, the kind of kernel function, its related kernel parameter, and the penalty coefficient. In this paper, the radial basis function is applied as the kernel function of LS-SVR, and the real double-chain coding target gradient quantum genetic algorithm (DCQGA) is applied to optimize the kernel parameter and penalty item coefficient of LS-SVR, then the regression prediction model DCQGALSSVR is proposed. It is of great significance to build an accurate and reliable fault prediction model for the health monitoring and fault diagnosis of liquid rocket engines. The thrust of a liquid rocket engine is an important factor in its health monitoring. By predicting the thrust change value and comparing the predicted value with the engine thrust threshold, it can be predicted whether the engine will fail at a certain time. In this paper, the proposed DCQGALSSVR model is used to model the thrust of a liquid rocket engine. The simulation results show that the average relative error is 0.37% using LSSVR for modeling on 12 test samples, and is 0.3186% using DCQGALSSVR on the same samples. It can be seen that DCQGALSSVR is effective for the health monitoring of liquid rocket engines, so it has a certain promotion significance.

Research on Fault Detection for Microservices Based on Log Information and Social Network Mechanism Using BiLSTM-DCNN Model

The microservice architecture breaks through the traditional cluster architecture mode based on virtual machines and uses containers as carriers to interact through lightweight communication mechanisms to reduce system coupling and provide more flexible system service support. With the expansion of the system scale, a large number of system logs with complex structures and chaotic relationships are generated. How to accurately analyze the system logs and make efficient fault prediction is particularly important for building a safe and reliable system. By studying neural network technology, this paper proposes an Attention-Based Bidirectional Long Short-Term Memory Network (Bi-LSTM). Combined with the dual channel convolutional neural network model (DCNN), it uses the attention mechanism to explore the differences between dimensional features, realizes multi-dimensional feature fusion, and establishes a BiLSTM-DCNN deep learning model that integrates the attention mechanism. From the perspective of social network analysis, a data preprocessing method is proposed to process fault redundant data and improve the accuracy of fault prediction under Microservices. Compare BiLSTM-DCNN with the mainstream system log analysis machine learning models SVM, CNN and Bi-LSTM, and explore the advantages of BiLSTM-DCNN in processing microservice system log text. The model is applied to simulation data and HDFS data set for experimental comparison, which proves the good generalization ability and universality of BiLSTM-DCNN.

Application of multivariate time-series model for high performance computing (HPC) fault prediction.

Aiming at the high reliability demand of increasingly large and complex supercomputing systems, this paper proposes a multidimensional fusion CBA-net (CNN-BiLSTAM-Attention) fault prediction model based on HDBSCAN clustering preprocessing classification data, which can effectively extract and learn the spatial and temporal features in the predecessor fault log. The model can effectively extract and learn the spatial and temporal features from the predecessor fault logs, and has the advantages of high sensitivity to time series features and sufficient extraction of local features, etc. The RMSE of the model for fault occurrence time prediction is 0.031, and the prediction accuracy of node location for fault occurrence is 93% on average, as demonstrated by experiments. The model can achieve fast convergence and improve the fine-grained and accurate fault prediction of large supercomputers.

Effective fault module localization in substation critical equipment: an improved ant colony optimization and back propagation neural network approach

AbstractThe rapid development of substations has increased the demand for accurate and fast fault prediction systems. In order to achieve rapid localization and autonomous decision‐making of fault modules and types in substations, the article proposes a fault autonomous localization algorithm based on improved ant colony optimization (IACO) and back propagation neural network (BPNN). The fault data of the substation secondary equipment for training and testing the BPNN model is based on the actual operating equipment of the substation, which can significantly improve the reliability of the model results. In addition, the IACO is used to globally optimize the weights and thresholds of BPNN, and the number of hidden layer nodes in BPNN was analyzed to further improve the accuracy of the established fault prediction algorithm. The test results show that the fault prediction accuracy of the BPNN model optimized by IACO is 93.67%, which is significantly improved compared to the traditional BPNN and BPNN with ant colony optimization method (with an accuracy of 82.98% and 91.04%). The above results effectively demonstrate the high accuracy and effectiveness of the established prediction algorithm in processing data and locating faults, which can improve the maintenance and operational efficiency of substations.

Precision fault prediction in motor bearings with feature selection and deep learning

In the disciplines of industrial machinery, mechanical engineering is beneficial to recognize motor performance for motors with HP power, torque transducer, dynamometer, and control electronics. The motivation is to address the need for more accurate and efficient fault prediction in machinery to prevent breakdowns, reduce maintenance costs, and improve overall reliability. In this work, deep learning classifiers used to classify ball defect inner race fault, outer race fault and normal motor performance in testing. With the aid of three distinct classifiers CNN, FFNN, and RBN; these suggested relative characteristics are assessed. In comparison to other current algorithms, the suggested methodology for classifying motor performance achieved maximum accuracy in each CNN test at 95.4% and 97.7%. The correlation and chi-square algorithms are used to find out the added characteristics and rank of features. The correlation technique provides relations between attributes, and the chi-square offers the optimal balance between precision and feature space. We discovered that the performance is enhanced overall by relative power characteristics. The suggested models might offer rapid responses with less complexity.

SD2SDFNet: A novel deep fusion network based on statistical denoising and dual-dimension self-attention model for power transformer hybrid-space fault prognosis

Intelligent fault prognosis plays an essential role in ensuring the stability and safety of power systems. However, the prediction of power transformer faults encounters significant challenges, including ultra-high noise interference, imbalanced samples, and difficulties in extracting sensitive fault features. To address these problems, this paper proposes SD2SDFNet, a novel deep fusion network that utilizes statistical denoising and a dual-dimension self-attention network. The proposed approach incorporates a statistical denoising approach called successive variational mode decomposition and Cramer Von Misses (SVMD-CVM). This technique aims to decompose signals into intrinsic mode functions (IMFs) and eliminate noise from each corresponding mode. By employing SVMD-CVM, the scheme effectively mitigates the impact of noise on fault prediction. Furthermore, a dual-dimension self-attention model is developed to selectively capture salient features from different IMF and time step dimensions. This adaptive feature selection process enhances the network's ability to focus on relevant information. To integrate the extracted features from both dimensions, a deep fusion network is adopted. This fusion process combines the weighted features, resulting in a more comprehensive set of features that facilitates accurate fault prediction. Experimental evaluations conducted on hybrid-space datasets demonstrate that SD2SDFNet achieves outstanding performance in predicting inter-turn faults. The network exhibited a Mean Squared Error (MSE) of 3.25E-05 and a Relative Absolute Error (RAE) of 1.58% for sensor 1, surpassing state-of-the-art prediction methods.

CaFANet: Causal-Factors-Aware Attention Networks for Equipment Fault Prediction in the Internet of Things.

Existing fault prediction algorithms based on deep learning have achieved good prediction performance. These algorithms treat all features fairly and assume that the progression of the equipment faults is stationary throughout the entire lifecycle. In fact, each feature has a different contribution to the accuracy of fault prediction, and the progress of equipment faults is non-stationary. More specifically, capturing the time point at which a fault first appears is more important for improving the accuracy of fault prediction. Moreover, the progress of the different faults of equipment varies significantly. Therefore, taking feature differences and time information into consideration, we propose a Causal-Factors-Aware Attention Network, CaFANet, for equipment fault prediction in the Internet of Things. Experimental results and performance analysis confirm the superiority of the proposed algorithm over traditional machine learning methods with prediction accuracy improved by up to 15.3%.

A Deep Learning-Based Fault Warning Model for Exhaust Temperature Prediction and Fault Warning of Marine Diesel Engine

Marine diesel engines are essential for safe navigation. By predicting the operating conditions of diesel engines, the performance of marine diesel engines can be improved, failures can be prevented to reduce maintenance costs, and emissions can be controlled to protect the environment. To this end, this paper proposes a hybrid neural network (HNN) prediction model (CNN-BiLSTM-Attention) based on deep learning (DL) for predicting the exhaust gas temperature (EGT) of marine diesel engines. CNN is used to extract features from time-series data, BiLSTM is used to predict the time series through modeling, and Attention is used to improve the accuracy and robustness of fault prediction. Moreover, through comparison experiments with other neural network prediction models, it has been proven that the CNN-BiLSTM-Attention method is more accurate. This article also presents an approach to fault prediction by integrating the Mahalanobia distance and the mathematical model. Based on the Mahalanobia distance between the prediction result and the actual value, the function mapping method combined with the criterion is used to set the alarm value and threshold of the monitoring indicators, and the failure data set is used for experimental verification. The results indicate that the approach presented in this article can accurately realize the operating condition monitoring and fault early warning of marine diesel engines, which provides a new way of thinking for the research of fault early warning and health management of marine diesel engines.

A multi-stage fault prediction method of continuous casting machine based on Weibull distribution and deep learning

When predicting the failure of large complete equipment such as continuous casting machines, it is usually difficult to obtain full life cycle failure data of core equipment such as continuous casting rollers. The strategy based solely on reliability distribution cannot guarantee the accuracy of fault prediction, while the strategy based solely on deep learning cannot predict medium and long-term fault trends. Aiming at the problem, a multi-stage fault prediction model TCN-BiGRU-WD for continuous casting rollers is built. The fault data of continuous casting roll is firstly input into TCN for feature extraction, and then the extracted features are input into BiGRU. Not only the short-term fault rate with high accuracy is output, but also the shape parameter(k) and scale parameter(λ) of Weibull distribution (WD) are output, so as to obtain the medium and long-term fault rate function (f(t;λ,k)) of continuous casting roll in the future. An example shows that the proposed multi-stage fault prediction model can not only improve the prediction accuracy of the short-term fault rate of continuous casting roll, but also expand the ability to predict the long-term fault trend in the future.

Fault diagnosis of 5G networks based on digital twin model

Fault diagnosis of 5G networks faces the challenges of heavy reliance on human experience and insufficient fault samples and relevant monitoring data. The digital twin technology can realize the interaction between virtual space and physical space through the fusion of model and data, providing a new paradigm for fault diagnosis. In this paper, we first propose a network digital twin model and apply it to 5G network diagnosis. We then use an improved Average Wasserstein GAN with Gradient Penalty (AWGAN-GP) method to discover and predict failures in the twin network. Finally, we use XGBoost algorithm to locate the faults in physical network in real time. Extensive simulation results show that the proposed approach can significantly increase fault prediction and diagnosis accuracy in the case of a small number of labeled failure samples in 5G networks.

PREDICTIVE ANALYSIS OF FAULTS IN ELECTRIC VEHICLES

Electric vehicle (EV) development has produced a number of advantages in terms of sustainability and energy efficiency. The dependability and safety of EV systems must still be ensured, though. In-depth study of predictive modeling methods for EV fault finding is provided in this research. The work focuses on the creation and use of defect prediction models utilizing real-time data gathered from EV systems, building on the foundations of predictive analysis and machine learning. The approach for data collecting, preprocessing, feature extraction, and model training is described in detail, with an emphasis on the necessity of reliable and representative data for successful fault diagnosis. Additionally, it analyzes the performance criteria employed for assessing the precision and dependability of these models and illustrates the difficulties in incorporating predictive analytic models into EV designs. The research results illustrate how predictive analysis may be successfully applied to problem detection and diagnosis, highlighting its potential to improve the safety and effectiveness of EV systems. In order to achieve more robust and accurate defect prediction in EVs, the article finishes with insights into future research areas, highlighting the need for more breakthroughs in data gathering techniques, model optimization, and real-time monitoring systems. KEYWORDS—Electric vehicles; predictive analysis; fault detection; machine learning; data collection; data preprocessing; feature extraction; model training; performance metrics; real-time monitoring;

Gaussian kernelized feature selection and improved multilayer perceptive deep learning classifier for software fault prediction

Software fault prediction is the significant process of identifying the errors or defects or faults in a software product. But, accurate and timely detection is the major challenging issue in different existing approaches to predicting software defects. A novel Gaussian linear feature embedding-based statistical test piecewise multilayer perceptive deep learning classifier (GLFE-STPMPDLC) is introduced to improve software fault prediction accuracy and minimize time consumption. First, the input data are collected from the dataset. Next, the software metrics selection is carried out to select the significant metrics using Gaussian kernelized locally linear embedding with lesser software fault prediction. Then classification is carried out by Kaiser Meyer piecewise multilayer perceptive deep learning classifier for software fault prediction. The novelty of Kaiser–Meyer–Olkin (KMO) correlation test analyzes testing and training instances. The innovation of the Heaviside step activation function is applied for analyzing the KMO correlation test results and providing the final software fault prediction results. Finally, accurate fault prediction outcomes are achieved at the output layer with lesser error. Simulation of proposed GLFE-STPMPDLC technique achieves better 5%, 3%, 3% and 3% enhancement of fault prediction accuracy, precision, recall, and f-measure and 13% faster prediction time compared to conventional methods.

Enhancing Power Grid Resilience through Real-Time Fault Detection and Remediation Using Advanced Hybrid Machine Learning Models

Ensuring a reliable and uninterrupted supply of electricity is crucial for sustaining modern and advanced societies. Traditionally, power systems analysis was mostly dependent on formal commercial software, mathematical models produced via a mix of data analysis, control theory, and statistical methods. As power grids continue to grow and the need for more efficient and sustainable energy systems arises, attention has shifted towards incorporating artificial intelligence (AI) into traditional power grid systems, making their upgrade imperative. AI-based prediction and forecasting techniques are now being utilized to improve power production, transmission, and distribution to industrial and residential consumers. This paradigm shift is driven by the development of new methods and technologies. These technologies enable faster and more accurate fault prediction and detection, leading to quicker and more effective fault removal. Therefore, incorporating AI in modern power grids is critical for ensuring their resilience, efficiency, and sustainability, ultimately contributing to a cleaner and greener energy future. This paper focuses on integrating artificial intelligence (AI) in modern power generation grids, particularly in the fourth industrial revolution (4IR) context. With the increasing complexity and demand for more efficient and reliable power systems, AI has emerged as a possible approach to solve these difficulties. For this purpose, real-time data are collected from the user side, and internal and external grid faults occurred during a time period of three years. Specifically, this research delves into using state-of-the-art machine learning hybrid models at end-user locations for fault prediction and detection in electricity grids. In this study, hybrid models with convolution neural networks (CNN) have been developed, such as CNN-RNN, CNN-GRU, and CNN-LSTM. These approaches are used to explore how these models can automatically identify and diagnose faults in real-time, leading to faster and more effective fault detection and removal with minimum losses. By leveraging AI technology, modern power grids can become more resilient, efficient, and sustainable, ultimately contributing to a cleaner and greener energy future.