On this page

A Methodology for Multi-dimensional Mining and Analysis of Power Grid Operation and Maintenance Data through the Synergy of Computation and Information Theory

By: Songyao Feng 1, Zhengyan Huang 1, Junhao Song 1, Xuexia Quan 1
1The Information Center of Guangxi Power System Co., Ltd., Nanning, Guangxi, 530012, China

Abstract

With the rapid development of smart grid and the increasing growth of electric power equipment, operation and maintenance intelligence gradually turns into an important way for power grid enterprises to improve productivity. The research proposes a smart grid operation and maintenance system using ExtJs+Spring+iBatis architecture. It first improves the weighted fusion rule based on the D-S evidence theory of virtual union, proposes a grid diagnostic model with multi-source information fusion, and then establishes a grid state evaluation model using AR model and SOM neural network model. The results show that the fusion model based on the improved D-S evidence theory fuses the results of switching quantity analysis and electrical quantity analysis for diagnosis, and the diagnosis results are more accurate compared to a single source of information, and at the same time, the grid state evaluation method can quickly and effectively detect the state of power grid operation and maintenance. The combination of big data analysis technology and power equipment evaluation will be a useful attempt in the construction of smart grid, which improves the requirements for equipment testing parameters.