Transformer Bushing Temperature Measurement Model Based on Infrared Temperature Measurement
Download as PDF
Chen Gu, Yu Yin, Qinyu Pan, Min Xu, Wangwang Shi, Guifeng Wu, and Donglei Chen
This paper proposes a new method of temperature measurement and predict model for transformer bushing based on support vector regression (SVR) algorithm. In this model, infrared image is used to measure the temperature of bushing porcelain. Aiming at the phenomenon that the temperature of infrared thermal image is easily disturbed by environment, a SVR filtering method is proposed to correlate the load current with the temperature of infrared thermal image, which maintained the correlation between bushing porcelain temperature and load current, reduced infrared temperature measurement noise. Compared with RBF neural network, BP neural network and SVR method, the training and prediction results of 110kV transformer bushing in a sample city show that the model has better prediction ability.
Transformer bushing, early warning, infrared temperature measurement, support vector regression (SVR)