Design of Intelligent Low Voltage Station System Based on Edge Calculation

Abstract

The low-voltage distribution network field wiring is very complicated and there are many changes in the relationship between households and changes, which brings great difficulties to the topology identification and line loss management of the transformer area. This paper proposed an intelligence low-voltage power system based on edge computing in order to improve the automation level of low-voltage distribution area, and to solve problems such as identification inaccuracy of topological relations in the low-voltage distribution area at present and singleness of condition monitoring parameters in station area, etc. The terminal is implanted with two edge calculation models based on artificial intelligence technology, which can realize topology identification and state parameters monitoring. The multistate parameter monitoring and decision analysis provides functions of event monitoring and early warning of abnormal state for the low voltage distribution system. In the topology identification analysis, the Markov random field was applied to establish the mathematical model of non-oriented graph and the joint probability distribution to describe the correlation between the nodes in the distribution network, thus realizing topology-relation identification of nodes in respective layers of the distribution area. On this basis, taking a low-voltage station as an example, a case analysis was undertaken. It was verified via the effectiveness and accuracy of the intelligence low-voltage station, which lays a foundation for research of the intelligent and fine management of the low voltage station system.