Abstract:Current transformers are important equipments in substations. Their operating conditions directly affect the safe operation of substations. There are a large number of current transformers, and they often meet various defects of current transformers during operation, such as heat generation and oil leakage. , low oil level and so on. Through the huge data of the current transformer on the PMS, the one-factor graph method is used to analyze the relationship between the current transformer fault and its equipment model, equipment manufacturer, equipment operation time, and multi-factors are considered to establish the BP neural network model. Considering the factors of equipment model, equipment manufacturer and equipment operation time, the probability prediction of its operating conditions is carried out. At the same time, all the current transformers that meet the requirements of the model in each substation are predicted, and the current transformer failure is easy to occur. Substations are alerted.The software called dituwuyou is used to visually display the results of the BP model calculation, which is convenient for the operation and maintenance personnel to grasp the operating conditions of the current transformer, strengthen the live detection of the current transformers that are prone to failure, arrange maintenance in advance, and ensure the reliability of power supply.