Research on Prediction of Dynamic Current Carrying Capacity of 110kV oil Immersed Transformer

In order to alleviate the problem of high short-term load rate in summer, improve the power transmission and transformation capacity of urban power system and improve the utilization rate of transformer equipment, the dynamic current carrying capacity prediction of 110kV oil immersed transformer is studied in this paper. This paper first calculates the hot spot temperature, top oil temperature and 24-hour life loss percentage of the transformer by using GB 1094.7. Secondly, the long-term and short-term memory neural network model (LSTM) is trained and verified. Compared with ARIMA model, it highlights the high accuracy and rapid convergence of LSTM, and predicts the power load data in the next 24 hours according to the historical data of power load in one month. Finally, set the three constraints of hot spot temperature limit, top oil temperature limit and life loss limit, calculate the normal, long-term and short-term current carrying capacity of oil immersed transformer according to the predicted power load data and real-time ambient temperature, and draw a conclusion: hot spot temperature is the limiting factor affecting the long-term and short-term load capacity of transformer, and the long-term current carrying capacity of this transformer is 1.61 times that of the original, The short-term current carrying capacity is 1.64 times of the original, which provides a certain reference for the power dispatching department.