Novelty Detection on Power Quality Disturbances Monitoring

Abstract

Complex disturbance patterns take place over the corresponding power supply networks due to the increased complexity of electrical loads at industrial plants. Such complex patterns are the result of a combination of simpler standardized disturbances. However, their detection and identification represent a challenge to current power quality monitoring systems. The detection of disturbances and their identification would allow early and effective decision-making processes towards optimal power grid controls or maintenance and security operations of the grid. In this regard, this paper presents an evaluation of the four main techniques for novelty detection: k-Nearest Neighbor, Gaussian Mixture Models, One-Class Support Vector Machine, and Stacked Autoencoder. A set of synthetic signals have been considered to evaluate the performance and suitability of each technique as an anomaly detector applied to power quality disturbances. A set of statistical features have been considered to characterize the power line. The evaluation of the techniques is carried out throughout different scenarios considering combined and single disturbances. The obtained results show the complementary performance of the considered techniques in front of different scenarios due to their differences in the knowledge modelization.

Methodology for the Detection and Classification of Power Quality Disturbances Using CWT and CNN

“…Another important aspect is that the signals analyzed in [19,20] are free of noise, which is an important issue that significantly affects the performance of classification when real signals are under evaluation. For the remaining works, such as in [22][23][24], the overall percentages show variations within the range of 80% to 96.67%, which are lower than the accuracy reached by the proposed approach. Therefore, it seems that conventional methods provide high percentages of classification accuracy, but such results are reached only because the analyzed signals are ideally perfect without noise.…”

“…On the deep learning side, several works have been developed. For instance, in [22,23] the authors use an auto-encoder neural network to find the most representative features of PQ-related events in an automated way. Also, in [24] the large number of images and information required for the correct training of the network takes at least 87 min; thus, it requires a high computational effort and hinders its application in continuous monitoring systems.…”