Search

Dissolved gas analysis (DGA) is commonly used to identify the fault type in power transformers. However, the available DGA methods have certain limitations because every method depends on the concentration of the dissolved gases. Therefore, in this work, hybrid feature selection–artificial intelligence–gravitational search algorithm (GSA) techniques were proposed to determine the fault type of power transformers based on DGA data. The artificial intelligence (AI) methods applied include support vector machine and artificial neural network. Both AI methods were optimised by GSA to enhance the accuracy of the results. Feature selections using stepwise regression and robust regression were applied to utilise only significant gases. The accuracy of the results was tested with various ratios of testing and training data. Comparison of the results using the proposed method with other optimisation methods and the previous works was performed to validate the performance of the proposed technique. It was observed that the proposed hybrid feature selection–AI–GSA technique yields reasonable accuracy although fewer types of dissolved gases were used. Therefore, the proposed method can be recommended for the application of automated power transformer fault type detection based on DGA data in practice. [ABSTRACT FROM AUTHOR]

Various fault location methods have been developed in the past to identify the faulty phase, fault type, faulty section, and distance. However, this identification is commonly conducted in a separate manner. An effective fault location should be able to identify all of these at the same time. Therefore, in this work, a method using a support vector machine (SVM) to identify the fault type, faulty section, and distance considering the faulty phase is proposed. The proposed method uses voltage sag magnitude of the distribution system as the main feature for the SVM to identify faults. The fault type is classified using a directed acyclic graph SVM. The possible faulty sections are identified by estimating the fault resistance using support vector regression and matching the voltage sag data in the database with the actual voltage sag data. The most possible faulty sections are ranked using ranking analysis. The fault distance for the possible faulty sections is then identified using support vector regression analysis and its overfitting or underfitting issues are addressed by the proper selection of a regularization parameter. The feasibility of the proposed method was tested on an actual Malaysian distribution system. The results of faulty phase, fault type, faulty section, and fault distance are analyzed. The performance of the proposed method is compared with various other intelligent methods such as the artificial neural network, deep neural network, extreme learning machine, and kriging method. The test results indicate that the faulty phase and fault type yield 100% accurate results. All the faulty sections are identified and the proposed method obtains reliable fault location. [ABSTRACT FROM AUTHOR]

Support vector machine (SVM) is a novel machine for data analysis and has advantageous characteristic of good generalization. Because of this characteristic, SVM is used in this work for fault classification and diagnosis in distribution systems. This work proposes an effective fault location method using SVM to identify the fault type, faulty section, and fault distance. The classification and regression analysis of the SVM are performed to locate a fault. The proposed method utilizes the voltage sag magnitude and angle measured at the primary substation of a distribution system. First, the fault type is identified using one- versus-one concept of support vector classification. The next step identifies the faulty section by calculating fault resistance, finding possible faulty sections and ranking the possible sections. Finally, the fault distance is identified using support vector regression analysis. The performance of the proposed method is tested using SaskPower distribution system from Canada having 20 line sections. Test cases are carried out under various fault scenarios considering the fault type and fault resistance. The results of fault distance are compared for different kernel functions, and the most accurate kernel is chosen. Test results show that the proposed method obtains reliable fault location. © 2017 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

Fault location identification is an important task to provide reliable service to the customer. Most existing artificial intelligence methods such as neural network, fuzzy logic, and support vector machine (SVM) focus on identifying the fault type, section, and distance separately. Furthermore, studies on fault type identification are focused on overhead transmission systems and not on underground distribution systems. In this paper, a fault location method in the distribution system is proposed using SVM, addressing the limitations of existing methods. Support vector classification (SVC) and regression analysis are performed to locate the fault. The method uses the voltage sag data during a fault measured at the primary substation. The type of fault is identified using SVC. The fault resistance and the voltage sag for the estimated fault resistance are identified using support vector regression (SVR) analysis. The possible faulty sections are identified from the estimated voltage sag data and ranked using the Euclidean distance approach. The proposed method identifies the fault distance using SVR analysis. The performance of the proposed method is analyzed using Malaysian distribution system of 40 buses. Test results show that the proposed method gives reliable fault location. [ABSTRACT FROM AUTHOR]

Cable joints are the weakest point in cross-linked polyethylene (XLPE) cables and are susceptible to insulation failures. Partial discharge (PD) analysis is a vital tool for assessing the insulation quality in cable joints. Although many works have been done on PD pattern classification, it is usually performed in a noise-free environment. Also, works on PD pattern classification are mostly done on lab fabricated insulators, where works on actual cable joint defects are less likely to be found in literature. Therefore, in this work, classifications of cable joint defect types from partial discharge measurement under noisy environment were performed. Five XLPE cable joints with artificially created defects were prepared based on the defects commonly encountered on-site. A high noise tolerance principal component analysis (PCA)-based feature extraction was proposed and compared against conventional input features such as statistical and fractal features. These input features were used to train the classifiers to classify each defect type in the cable joint samples. Classifications were performed using Artificial Neural Networks (ANN) and Support Vector Machine (SVM). It was found that the proposed PCA features displayed the highest noise tolerance with the least performance degradation compared to other input features under noisy environment. [ABSTRACT FROM AUTHOR]

Fast and effective fault location in distribution system is important to improve the power system reliability. Most of the researches rarely mention about effective fault location consisting of faulted phase, fault type, faulty section and fault distance identification. This work presents a method using support vector machine to identify the faulted phase, fault type, faulty section and distance at the same time. Support vector classification and regression analysis are performed to locate fault. The method uses the voltage sag data during fault condition measured at the primary substation. The faulted phase and the fault type are identified using three-dimensional support vector classification. The possible faulty sections are identified by matching voltage sag at fault condition to the voltage sag in database and the possible sections are ranked using shortest distance principle. The fault distance for the possible faulty sections isthen identified using support vector regression analysis. The performance of the proposed method was tested on an unbalanced distribution system from SaskPower, Canada. The results show that the accuracy of the proposed method is satisfactory. [ABSTRACT FROM AUTHOR]