Intelligent Identification Of Hydrometeor Phase State And Its Application In Weather Analysis

Hydrometeor Classification Algorithm(HCA)is an important research direction for analyzing weather characteristics from the perspective of microphysics.The study on the Classification of Hydrometeor phase states is of great significance for the observation of hail,rainfall and snowfall.In this paper,the phase state recognition algorithm of hydrocondensate is studied by using dual polarization weather radar.Its main work is as follows:1.The polarization characteristics of hydrocondensate,the physical meaning of each polarization parameter and the echo analysis,as well as the microphysical characteristics of meteorological echo are analyzed.The polarization characteristics of precipitation samples in meteorological echo can be explained by studying the polarization parameters,and the shape,size and orientation of different precipitation particles can be obtained.2.Studied the progress of phase state identification technology of dual-polarization weather radar,summarized the advantages of the classical Fuzzy logic Hydrometeor Classification(FHC)and the application of phase state identification of hydrometeors in the weather process.The existing problems of FHC are analyzed,including the parameter setting of membership function and the recognition effect of ice phase particles.3.Aiming at the parameter setting of membership function of traditional fuzzy logic Algorithm,a precipitation particle classification Algorithm based on sass-fcm(Simulate Anneal Genetic algorithm-fuzzy c means)was proposed.The method USES the polarization parameters of dual-polarization weather radar echo to identify and classify the phase states of precipitation particles adaptively.In this paper,the particle phase state classification of KTFG radar algorithm applied in different temperature conditions is listed.It is proved that the classification results based on sg-fcm algorithm are basically consistent with the data results provided by the national oceanic and atmospheric administration(NOAA),and the algorithm can effectively classify the phase state of precipitation particles.4.Through weather analysis of a downburst process in the north of greeley(Colorado),USA,the state of simulated atmospheric environment,quantitative analysis of radar detection data,and observation of particle phase change characteristics.Compared with the traditional FHC,the phase recognition algorithm of SAGA-FCM hydrocondensate proposed in this paper reduces the computation and the learning times and time complexity in the model design.Compared with FCM algorithm,the optimized algorithm can reduce the recognition error,more accurately identify graupel particle region,and accurately determine the hail region.