Transformer state prediction based on CSSA-BP neural network algorithm

Abstract

Abstract As an important part of the power system, it is particularly important for the monitoring of the status of the power transformer in real time opearation. A transformer winding temperature prediction method through swarm intelligent optimization algorithm (CSSA) and BP neural network is proposed. Firstly, the CSSA-BP prediction model of the winding’s temperatur of dry-type transformer is established, and the existing 300 groups of data on the cloud platform are used for simulation analysis. The results show that, through the optimization by sparrow algorithm, the proposed model has a good prediction performance and is of great significance for realizing transformer winding temperature monitoring.