Prediction of the Breakdown Voltage of Transformer Oil Based on a Backpropagation Network

摘要

Prediction of the breakdown voltage of transformer oil facilitates the early fault diagnosis of transformers,and provides a scientific basis for the prevention of faults in transformer oil.In this paper,based on the correlation between performance parameters of transformer oil,along with the excellent fault-tolerant ability,prominent non-linear approximation capability and self-learning capacity of backpropagation (BP) networks,a BP network with a BP algorithm and a BP network with an improved BP algorithm are developed to simulate the correlation between breakdown voltage and four relevant parameters,using the monitoring data of transformer oil.The results show that the latter algorithm gives more accurate predicted values,which proves to be of high application value.