BP Neural Network-based Transmission Line Galloping Prediction And Grid Risk Early Warning Method

China is one of the most serious countries in the world that affected by the transmission line galloping disaster,and there is a traditional galloping belt extending from the northeast to the south as Hunan.Galloping disaster might cause the line flashover and trips,at worst,might cause the wear and tear of insulators and metal fittings,even other serious power accidents such as tower fall,which will force the important transmission channel outage for a long time.However,only through the construction and transformation of transmission lines to prevent galloping disaster would be negative economics and operability,and therefore we urgent need to forecast the galloping through information methods,and optimize the traditional power grid security evaluation and galloping early warning system,that will strengthen the capacity of the grid to cope with bad weather conditions economically and reliably.In this paper,the formation and evolution of the galloping disaster of transmission lines are analyzed,and the forecasting model of transmission line galloping based on BP neural network is established,and the short term operation risk of the line galloping is evaluated.The main contents of the thesis are:(1)Taking the history galloping samples of a provincial power grid as an example,the temporal and spatial distribution characteristics of meteorological factors before galloping disaster are studied,and the meteorological factors that will affect galloping are extracted.We found there are strong correlations among the wind speed,wind direction,temperature,relative humidity and galloping,and the construction method of transmission line galloping prediction and short term power grid risk early warning is put forward.(2)Consider to adopt a supervised machine learning method as a basic means of galloping prediction.Based on BP neural network,the forecasting model of transmission line galloping is constructed.The regional meteorological factors and the regional ID(County)which the transmission line is seat of are selected to are used to train the model.The validity of the technical method of the transmission line galloping prediction model based on the BP neural network is illustrated by the regional and linear examples of a province.(3)According to the current safety assessment criteria,defines the possibility of line galloping and short-term operational safety indicators,establishes a short-term power grid risk assessment method and risk early warning system considering line galloping forecasting,and proposed galloping risk level,galloping warning information and precautions to reduce risk operation measures.Through the actual case of Henan power grid,the proposed method is described in detail.