Affiliations: [a] Computer Architecture and Networks Group, School of Computer Science, University of Castilla-La Mancha, Ciudad Real, Spain | [b] Institute of Energy Research and Industrial Applications, University of Castilla-La Mancha, Ciudad Real, Spain | [c] Institute of Technology and Information Systems, University of Castilla-La Mancha, Ciudad Real, Spain
Correspondence:
[*]
Corresponding author: Maria J. Santofimia, Computer Architecture and Networks Group, School of Computer Science, University of Castilla-La Mancha, Ciudad Real, Spain. E-mail: mariajose.santofimia@uclm.es.
Abstract: This research is intended to evaluate the suitability of a common-sense-based approach for providing causal explanations to power quality disturbances and, more specifically, to voltage-sage events, with the aim of improving the security and reliability of the electrical grid. Since voltage sags are one of the most common power-quality disturbance and may have a severe impact on sensitive loads and users, the case study presented is dedicated to find their causes as a prerequisite to prevent them. The main contribution presented in this work is in the knowledge management domain. However, the proposed architecture adopts a multi-layer approach that comprehensively faces all the necessary stages, comprising: measurement and data collection, filtering, distribution, homogenization, and integration up to the process of inference of the possible cause of the event. The architecture of the proposed system comprises three layers, namely: the information gathering, the information modeling and the information understanding layers. The proposed system has been evaluated using a synthetic dataset of simulated voltages sags. Information about weather conditions and different location features and environmental circumstances is aggregated to the electrical features of the voltage sag, in order to improve the inference of the external cause of the event. Even though a real-world dataset would be required to fully validate the proposed system and assess the benefits in a real scenario, the results obtained with the synthetic dataset are satisfactory and an overall accuracy above 90% of the cause identification is achieved.
Keywords: Smart grid, power quality, common sense, voltage sag
