Analysis method of voltage quality in distribution station area based on high‐speed power line carrier and GA‐ANN in smart grid

Abstract

Aiming at the problem that most of the existing methods cannot obtain and accurately analyse the voltage quality of distribution network in time, a method for monitoring and analysing the voltage quality of distribution station area based on high-speed power line carrier under the power Internet of Things (PIoT) architecture in smart grid is proposed. Firstly, based on the PIoT architecture, a monitoring system for the distribution station area is designed, which realises the reliable monitoring of the voltage in the station area through the information interaction of the sensing layer, the network layer, the platform layer and the application layer. Then, the high-speed power line carrier is used to realise the rapid monitoring and collection of the station data, which improves the quality of basic data, and the principal component analysis method is used to extract the voltage quality characteristics and reduce the dimension. Finally, genetic algorithm (GA) is used to optimise the training of artificial neural network (ANN) to obtain the best improved GA-ANN network model. It is used to analyse the voltage quality feature set, which further improves the accuracy of obtaining voltage quality anomalies and related reasons. Based on the established distribution station area, the proposed method is experimentally demonstrated. The results show that the accuracy of voltage quality anomaly monitoring and analysis exceeds 99.5% and 97% respectively, and the average accuracy of low-voltage cause analysis reaches 97.83%, laying a theoretical foundation for building a strong distribution network. K E Y W O R D SArtificial intelligence and data analytics, Communications and networking, distribution networks, genetic algorithms, power grids, Smart grid devicesThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.