Research On Grid Security Early Warning Model Based On Meteorological Factors

As modern society’s dependence on power levels getting higher and higher, growing power grid size to ensure reliable supply of electricity in the socio-economic importance of higher and higher. A short big area power cut creates the loss consequence and the influence are incalculable. However, due to the complexity of the production and supply of electricity and the production process transient, large-area power outages often occur. Therefore, establishment of electrical safety warning to quickly and efficiently solve the large area blackout accidents that may occur, is safeguards the social essential for electricity demand and power grid security.In recent years, various factors resulting in high temperature, typhoons, thunderstorms, strong convection, wind, fog haze weather phenomena such as anomalous multiple, compared with the previous meteorological disasters more frequent, resulting in power grid accidents. s global climate change and global warming, unusual weather disasters, as well as increased grid size continues to expand, meteorological disasters of grid affect the frequency and extent of which will continue to increase. conduct electricity weather in-depth research, building grid meteorological disaster monitoring and warning to build strong power grid is essential.In this paper, based on the theory sufficient to collect large amounts of data through the analysis of grid security operation worldwide and the safe operation of the national grid in2010and found that meteorological factors are the main factors leading to grid security incidents. In response to the factors, the option3of the early-warning model, elected logistic model, for a southern region grid incidents for the empirical analysis. The paper has collected this local continual48month-long electric power accident number as well as the corresponding15meteorological element, first used the factor analytic method to eliminate15meteorological element multiple collinearities, has withdrawn the temperature factor, the precipitation factor, the humidity factor, the wind power factor four kind of primary factors, next has established the climatic factor using the Logistic return to the electric power accident influence model. Model research on weather conditions and incidents of internal relations and2010year test to verify the accuracy of model fitting, for the power has laid a good foundation for the establishment of early warning mechanisms, provide theory and methods for network security management support.