Supervised Learning for Fault Classification Using Hybrid Training Datasets
Archana Ranganathan (Alliander)
Simon H. Tindemans (TU Delft - Intelligent Electrical Power Grids)
Frans Provoost (Qirion)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Electrical faults in the distribution system can lead to interruptions in customer power supply resulting in penalties that are borne by the distribution system operator. Accurate fault classification is an important step in locating the fault to achieve faster network restoration times. This paper presents a classification model in two parts: one determines the degree of stability in the fault waveforms and the second uses a machine learning model to classify real-world faults based on the number of fault phases. A set of business rules are developed to characterise instability by performing a windowed Fourier analysis and studying the strength of the fundamental frequency component of fault waveforms. Results show that the developed SVM model can differentiate between real-world instances of single-phase, two-phase and threephase stable faults with a classification accuracy of 95%. Additionally, we show that adding a small subset of synthetically developed faults to the training data improves classification accuracy.