Research And Implementation Of Evaporation Duct Inversion And Prediction Method Based On Evolutionary Optimization

In the field of meteorological sciences,evaporation ducts have an important influence on the propagation of electromagnetic waves,which can increase radar clutter,cause over-thehorizon propagation and even cause radio communication signal interruption.Evaporation duct inversion problem is a complex nonlinear problem,there is no analytical solution,it can be modeled as a multi-objective optimization problem.Generally,methods such as traditional optimization algorithms and evolutionary multi-objective optimization can be used to solve the cross-sectional structure and propagation characteristics of the air duct,and then the forward propagation of the air duct can be modeled to simulate the forward modeling process of the air duct.Evaporation ducts in evaporation ducts often appear in the near-surface layer close to the sea surface,are ubiquitous above the ocean surface,and have an important influence on the propagation of electromagnetic waves,affecting the performance of shore-based,shipborne or low-altitude airborne radar and communications systems Important factor.This thesis is aimed at the inversion and prediction of evaporation ducts and related research using evolutionary algorithms and machine learning methods.Specifically,the main work of this thesis is as follows:(1)The Global Navigation Satellite System(GNSS)signal receiving antenna can record the propagation loss and phase delay of the electromagnetic wave in the actual propagation process when receiving the electromagnetic wave signal.This thesis takes the phase delay and propagation loss of GNSS as the optimization goals,and calculates the target value through forward modeling(Evaporation Duct Simulation Propagation Calculation program)and establishes a multi-objective optimization problem model.On this basis,design and implement an evolutionary optimization algorithm for a single GNSS received signal and solve the optimization problem.At the same time,this thesis carries out related experiments on the standard problem test set and the evaporation duct inversion problem for the evolutionary multi-objective optimization algorithm for a single GNSS signal source.The experimental results show that the algorithm proposed in this thesis has the uniformity of the population distribution.Compared with similar algorithms,the accuracy is significantly improved.(2)When multiple GNSS signal sources receive the same electromagnetic wave signal at the same time,the GNSS phase delay and propagation loss of multiple signal sources can be simultaneously used for Evaporation duct inversion,which can further improve the prediction accuracy in theory.This thesis takes the phase delay and propagation loss of multiple GNSS as the optimization goals,calculates the target value through forward modeling and establishes a multi-objective optimization problem model.Based on the design of the evolutionary optimization algorithm for a single GNSS signal source,the corresponding high-dimensional heuristic strategy is introduced,and an evolutionary multi-objective optimization algorithm for multiple GNSS signal sources is designed.At the same time,this thesis has carried out related experiments on the evolutionary multi-objective optimization algorithm for multiple GNSS signal sources in the standard problem test set and the Evaporation duct inversion problem.The experimental results show that the algorithm designed in this thesis has better solution performance.(3)The height of the evaporation duct is also an important parameter in the expression of the Evaporation duct profile.Duct inversion can only infer the profile parameters of the duct based on the propagation loss and phase delay of the received signal source.In order to predict the evaporative duct height in the future based on historical offshore surface meteorological data,this thesis designs and implements an Evaporation duct height prediction method based on evolutionary optimization and support vector machines,which combines parameter tuning and time series data prediction.Solve the problem of time series prediction of evaporation duct height.The experimental results show that the prediction method proposed in this thesis has obtained good results.To sum up,this thesis mainly carried out problem modeling,algorithm design and experimental verification for the inversion and prediction of Evaporation duct.Aiming at the Evaporation duct inversion problem,this thesis designs and implements an evolutionary multi-objective optimization algorithm to solve such problems;for the evaporation duct height prediction problem,this thesis designs and implements an SVR prediction method based on EMD decomposition and parameter optimization.To predict the height of the short-term evaporation waveguide.The algorithms proposed in this thesis have been validated through experiments.They are a systematic study of newer ideas for solving problems related to Evaporation duct inversion and prediction,as well as a useful attempt to apply artificial intelligence methods to practical applications.