Forecasting for the Risk of Transmission Line Galloping Trip Based on BP Neural Network

Due to the strong randomness and nonlinear characteristics of the transmission line galloping, the prediction of the intensity and the characteristics (amplitude, frequency, trip rate, etc.) of the galloping cannot reach a high precision. A BP neural network model is employed to map three main meteorological factors and galloping trip-out risk. Three main meteorological factors, temperature, humidity and wind speed were used as input parameters and the risk of galloping trip as the output parameter of the model. Typical galloping data from State Grid Corporation were used to verify the validity of the model. In order to counteract random factors, the operations were performed for 20 times with the same training and testing data. All of the network results had more than 90% accuracy and the average rate was 92.3%. The results show that it is feasible to use this model to predict the risk of transmission line galloping trip. The research results can provide support for the transmission line galloping prediction and early warning technology, so as to improve the level of intelligent operation and maintenance of power grid.