A Hierarchical Category Embedding Based Approach for Fault Classification of Power ICT System

This paper presents a modular approach for detection, classification and location of transmission line faults using wavelet transform along with intelligent techniques. By using wavelet Multiresolution Analysis (MRA), summation of detail coefficients is extracted for three-phase fault currents. These detail coefficients constitute the edifice for classification and location of faults. The classification of different types of faults is done using Fuzzy Inference System (FIS). The algorithm proceeds with classification of arcing and non-arcing faults, and then locates the fault on the transmission line using Modular Artificial Neural Networks (MANNs). The results indicate that such a modular approach can be used for supporting high-speed protective relaying systems during both arcing and non-arcing faults.

A systematic fuzzy rule based approach for fault classification in transmission lines

This study presents an approach for fault detection and classification in a DC drive system. The fault is detected by a classical Luenberger observer. After the fault detection, the fault classification is started. The fault classification, the main contribution of this paper, is based on a representation which combines the Subctrative Clustering algorithm with an adaptation of Particle Swarm Clustering.

This study presents an approach for fault detection and classification in a DC drive system. The fault is detected by a classical Luenberger observer. After the fault detection, the fault classification is started. The fault classification, the main contribution of this paper, is based on a representation which combines the Subctrative Clustering algorithm with an adaptation of Particle Swarm Clustering.

This paper presents an approach for the fault classification in transmission line using multi-class support vector machine (SVM). This approach uses information obtained from the wavelet decomposition of post fault current signals as input to SVM for classification of various faults that may occur in transmission line. Using MATLAB Simulink, dataset has been generated with different types of fault and system variables, which include fault resistance, fault distance and fault inception angle. The proposed method has been extensively tested on a 240-kV, 200-km transmission line under variety of fault conditions. The results indicate that the proposed technique is accurate and robust for a variation in system parameter and fault conditions.

A fault classification approach is presented which considers the advantages of Time Encoded Signal Processing (TESP) of dynamic signals combined with the ability of Artificial Neural Networks (ANNs) to classify changes in TESP codes. This is demonstrated using a new TESP code approach applied to a pressure control system exhibiting both leakage at the actuator and a servovalve fault. It was found that the use of both pressure transducer voltage and servovalve drive voltage, when entered into the ANN in a parallel data structure manner, resulted in an excellent fault classification capability. In addition the inherent classification approach gave very good leakage discrimination for arbitrarily-set, and low, levels of 0, 2, 4, 6 l/min. A range of 16 different ANNs were investigated and the classification results indicate a preferred topology for this application.

Deep learning-based automated defect classification in Electroluminescence images of solar panels

Switching LDS-based approach for process fault detection and classification

This study aims to develop a real-time model for detecting and classifying various fault types in power distribution networks to enhance reliability and operational efficiency. A hybrid DWT–SVM approach is essential for accurately detecting and classifying faults in modern power distribution systems, especially under non-stationary and dynamic conditions. The proposed method addresses a range of fault scenarios, including single-phase-to-ground, line-to-line, double-line-to-ground, and three-phase faults, within a 33 kV distribution network. A hybrid approach combining Discrete Wavelet Transform (DWT) and Support Vector Machine (SVM) is introduced. Using the Debauchies-4 (Db4) wavelet, DWT effectively decomposes transient fault currents at the source terminal, capturing critical time-frequency domain features. Fault classification is performed using an SVM optimized with a Radial Basis Function (RBF) kernel and Particle Swarm Optimization (PSO), enabling precise mapping of data into higher-dimensional spaces for optimal separation. Validation conducted with MATLAB R2023b demonstrates a detection accuracy of 100% and a classification accuracy of 99%. Comparative analyses against models such as PSO-based Support Vector Machine (PSO-SVM), DWT-DNN and wavelet transform with artificial neural networks (WT-ANN) on the IEEE 13-bus system highlight the proposed method's superior performance. This innovative approach proves to be robust, adaptable, and highly effective across diverse fault scenarios, offering significant improvements in accuracy and reliability for fault detection and classification in power distribution networks. The method supports real-time operation by using high-speed simulation platforms MATLAB/Simulink to process signals and classify faults within milliseconds using a pre-trained SVM model.

This article presents a new approach for fault classification in a two-terminal overhead transmission line using a support vector machine classifier. Wavelet transform is used for the decomposition of measured signals and for extraction of the most significant features (feature extraction), which facilitates training of the SVM, particularly in terms of getting better classification performance (high accuracy). After extracting useful features from the measured signals, a decision of fault or no-fault on any phase or multiple phases of a transmission line is carried out using three SVM classifiers. The ground detection task is carried out by a proposed ground index. Two kernel functions—polynomial and Gaussian radial basis function (RBF)—have been used, and performances of classifiers have been evaluated based on fault classification accuracy. In order to determine the optimal parametric settings of an SVM classifier (such as the type of kernel function, its associated parameter, and the regularization parameter C), five-fold cross-validation has been applied to the training set. It is observed that an SVM with an RBF kernel provides better fault classification accuracy than that of an SVM with polynomial kernel. One of the key points of this article is the development of an automatic fault data generation model using PSCAD and its application for training and testing of SVMs. To illustrate the effectiveness of the proposed scheme, extensive simulations have been carried out for different fault conditions with wide variations in the operating conditions and source impedances. It has been found that the proposed scheme is very fast and accurate, and it proved to be a robust classifier for digital distance protection.

A fast and accurate distance relaying scheme using an efficient radial basis function neural network

A new multiway discrete hidden Markov model (MDHMM)‐based approach is proposed in this article for fault detection and classification in complex batch or semibatch process with inherent dynamics and system uncertainty. The probabilistic inference along the state transitions in MDHMM can effectively extract the dynamic and stochastic patterns in the process operation. Furthermore, the used multiway analysis is able to transform the three‐dimensional (3‐D) data matrices into 2‐D measurement‐state data sets for hidden Markov model estimation and state path optimization. The proposed MDHMM approach is applied to fed‐batch penicillin fermentation process and compared to the conventional multiway principal component analysis (MPCA) and multiway dynamic principal component analysis (MDPCA) methods in three faulty scenarios. The monitoring results demonstrate that the MDHMM approach is superior to both the MPCA and MDPCA methods in terms of fault detection and false alarm rates. In addition, the supervised MDHMM approach is able to classify different types of process faults with high fidelity. © 2011 American Institute of Chemical Engineers AIChE J, 2012

Condition classification of rolling element bearings in rotating machines is important to prevent the breakdown of industrial machinery. A considerable amount of literature has been published on bearing faults classification. These studies aim to determine automatically the current status of a roller element bearing. Of these studies, methods based on compressed sensing (CS) have received some attention recently due to their ability to allow one to sample below the Nyquist sampling rate. This technology has many possible uses in machine condition monitoring and has been investigated as a possible approach for fault detection and classification in the compressed domain, i.e., without reconstructing the original signal. However, previous CS based methods have been found to be too weak for highly compressed data. The present paper explores computationally, for the first time, the effects of sparse autoencoder based over-complete sparse representations on the classification performance of highly compressed measurements of bearing vibration signals. For this study, the CS method was used to produce highly compressed measurements of the original bearing dataset. Then, an effective deep neural network (DNN) with unsupervised feature learning algorithm based on sparse autoencoder is used for learning over-complete sparse representations of these compressed datasets. Finally, the fault classification is achieved using two stages, namely, pre-training classification based on stacked autoencoder and softmax regression layer form the deep net stage (the first stage), and re-training classification based on backpropagation (BP) algorithm forms the fine-tuning stage (the second stage). The experimental results show that the proposed method is able to achieve high levels of accuracy even with extremely compressed measurements compared with the existing techniques.

The objective of this project is to conduct a comparative study of sentimental analysis of textual data using machine learning techniques, specifically employing an Early Classification Based Approach for Fault Classification. The emotions under consideration are sadness, joy, love, anger, fear, and ego, which are commonly expressed in text data. To achieve this goal, three machine learning algorithms, namely Stochastic Decision Tree (DT), Random Forest (RF), and Long Short-Term Memory (LSTM), are utilized. The study utilizes a dataset containing diverse text samples expressing different emotions, which is pre-processed to remove irrelevant information and then split into training and testing sets. The DT, RF, and LSTM models are trained on the training set and evaluated on the testing set using various evaluation metrics. The experimental results provide insights into the performance of the three algorithms for sentimental analysis through text. The findings reveal that DT and RF achieve comparable accuracy levels, while LSTM outperforms them in terms of overall classification accuracy However, Random Forest exhibits better performance in capturing sequential patterns and contextual information in text data, making it particularly effective for emotions expressed through longer and more complex text samples, such as ego. On the other hand, DT and RF may perform better for emotions expressed through shorter and more straightforward text samples, such as joy and sadness, due to their relative simplicity. This research contributes to the field of emotion classification and provides valuable insights for researchers and practitioners in selecting appropriate machine learning algorithms for emotion classification tasks. The early classification-based approach for fault classification in the context of emotion classification from text opens up avenues for further research in this area. Keywords— Decision Tree, Random Forest, LSTM.

A rough membership neural network approach for fault classification in transmission lines