Highlighting of local Power Quality states with the new QuEEN system, enhanced with Deep Learning and Machine Learning algorithms

Abstract

The enhancement of the new QuEEN system, with the integration of Deep Learning and Machine Learning algorithms for the evaluation of the validity and origin of voltage dips, has allowed to highlight specific situations concerning Power Quality. It was in fact possible to make comparisons, also at a regional level, between the results of these algorithms and those of the "official" QuEEN criteria, by cross-referencing the results with the network data. If at national level the results of the extended comparison only provide confirmations of previous assumptions, we observe specific local situations regarding the number of occurrences of false voltage dips, and therefore of single-phase earth faults. These situations are analyzed with reference to network characteristics.