Abstract-This paper presents an artificial neural network (ANN) and support vector machine (SVM) approach for locating faults in radial distribution systems. Different from the traditional Fault Section Estimation methods, the proposed approach uses measurements available at the substation, circuit breaker and relay statuses. The data is analyzed using the principal component analysis (PCA) technique and the faults are classified according to the reactances of their path using a combination of support vector classifiers (SVCs) and feedforward neural networks (FFNNs). A practical 52 bus distribution system with loads is considered for studies, and the results presented show that the proposed approach of fault location gives accurate results in terms of the estimated fault location. Practical situations in distribution systems, such as protective devices placed only at the substation, all types of faults, and a wide range of varying short circuit levels, are considered for studies. The results demonstrate the feasibility of applying the proposed method in practical distribution system fault diagnosis.

No abstract

No abstract

No abstract

Post-fault studies of recent major power failures around the world reveal that maloperation and/or improper coordination of protection system were responsible to some extent. When a major power disturbance occurs, protection and control action are required to stop the power system degradation, restore the system to a normal state and minimise the impact of the disturbance. However, this has indicated the need for improving protection coordination by additional post-fault and corrective studies using intelligent/knowledge-based systems. A process to obtain knowledge-base using support vector machines (SVMs) is presented for ready post-fault diagnosis purpose. SVMs are used as Intelligence tool to identify the faulted line that is emanating and finding the distance from the substation. Also, SVMs are compared with radial basis function neural networks in datasets corresponding to different fault on transmission system. Classification and regression accuracies are is reported for both strategies. The approach is particularly important for post-fault diagnosis of any mal-operation of relays following a disturbance in the neighbouring line connected to the same substation. This may help to improve the fault monitoring/diagnosis process, thus assuring secure operation of the power systems. To validate the proposed approach, results on IEEE 39-Bus New England system are presented for illustration purpose.