Anomalies analysis and prediction method research of converter transformer winding temperature in HVDC project based on BP neural network

Operating experience shows that the converter transformer winding temperature in HVDC project is mainly affected by ambient temperature, current load, number of cooler groups,etc. In this paper, a converter transformer oil temperature prediction method based on BP neural network is proposed by introducing the principle of the converter transformer winding temperature measurement device, which proves the reliability of the performance index of the method. Meanwhile through the analysis of the fault causes of the converter transformer winding temperature measurement device, an operation treatment method for winding temperature abnormality is proposed.