Fitting Of The Icing Forecasting Model And Method Of Icing Process Identification

By using the icing data and the related meteorological data over the same period of Guizhou Province during the1951-2007winter season, the methods of correlation analysis, regression analysis and discriminatory analysis are used to study the meteorological factors and icing index associated with the icing of electric transmission line. The main conclusions are as follows:(1) The phenomenon of wire icing has affected by many kinds of meteorological factors, such as atmospheric temperature, humidity, velocity of wind and so on. Analyzed meteorological factors of research region, it is found that the phenomenon of electric transmission line icing exist at the high altitude area of Guizhou Province. The optimal condition of icing are as follows:temperature range is-6℃-0℃, relative humidity is higher than90%, velocity range of wind is0-4m/s.(2) Mechanism of icing is complex, has many influence factors. This paper analysis process of wire icing weight growth, according to the theoretical framework of ice model, then establishes the icing weight prediction model. Based on the historical icing observational data and common meteorological data in the same period from1951to2007in Guizhou Province, this study fitted the experience function of some parameters of icing forecasting model, and then established icing weight prediction models. After the simulation tests, the fitting effect is more ideal. The technology of meteorological Observation in icing period and the design of icing model have some shortcomings and deficiencies, these may lead to the calculation error. The result obtained in this paper is an attempt in the existing conditions.(3) The key of icing process identification is the identification of start time and end time of process. Whether the beginning or ending of ice process, its essence is the moment of the combination of environmental meteorological factors reaches a certain critical value. Icing phenomenon easily appeared in the weather conditions of rainfall, low temperature, high relative humidity, low wind speed. Too high or too low temperature is not conducive to icing. Temperature decreases, relative humidity increases, icing process begin, temperature rises, relative humidity decreases, icing process is over.(4) Discriminatory analysis of ice-coating day has accomplished by using the Fisher method, selected day mean total cloud cover, daily minimum temperature, daily average water vapor pressure, daily mean relative humidity, daily mean wind speed for the impact factor, established the discriminant equation. Test the effect of discriminant equation:the correct rate of historical data fitting icing day events is82.94%, the correct rate of the historical data fitting without icing events is75.34%, and the correct rate of total historical data fitting is79.32%. The discriminant equation passed the significance test, it can forecast the real situation of icing appearance. (5) This paper established the beginning and end discriminant equations of icing process respectively, according to the meteorological factors in different stage, selected main influence factors. The fitting correct rate of historical data of icing process start events is76.3%, icing process end events’ historical data fitting accuracy rate is70.1%. With analysis of the meteorological conditions at the beginning and end of the icing process, correct discrimination of the start and end events, we can determine the start node and end node of icing process, and then can determine the icing process and process duration. It can provide some reference for the reproducing of icing process or icing forecasting.