Research On The Inversion Method Of Visibility Based On Micro Pulse Lidar

Atmospheric visibility is an important meteorological parameter.It is of great significance to obtain its information in real time and accurately for transportation support,environmental monitoring,weather forecasting and so on.Lidar visibility meter represents the new direction of the development of visibility measurement technology due to the advantages of wide measurement range,high temporal,spatial resolution and strong anti-interference capability.However,traditional Mie scattering lidar measurement methods need to assume or approximately treat about the aerosol extinction backscattering ratio(AEBR)and the extinction coefficient boundary value(AEC-BV)in the inversion process,which limits the accuracy of the inversion results.In this thesis,a micro-pulse lidar visibility system is developed based on the Mie scattering principle and a high-precision aerosol extinction coefficient inversion algorithm is proposed to achieve accurate measurement of visibility.The specific research contents are as follows:1.Firstly,the design idea of the entire micro-pulse lidar visibility system is elaborated from the aspects of the selection of lidar's important components and the main parameters.The signal preprocessing scheme is given,and the noise in the echo signal is filtered by the empirical mode decomposition(EMD)method based on the principle of correlation coefficient.The signal to noise ratio of the echo signal increases to 20.31dB after de-noising.2.This thesis proposes a parameter determination method for the AEBR and AEC-BV based on the secant method,and generates the aerosol extinction coefficient profiles by the Fernald method accurately and obtains the atmospheric visibility value.The experimental results show that the lidar visibility meter and inversion algorithm have good stability and measurement accuracy.In the continuous experiment,the average absolute error of visibility is only 5.48%.3.In addition,in order to avoid the problem of parameters assumption in the aerosol extinction coefficient inversion process,we propose a novel method,a feedback radial basis function(RBF-FB),for retrieving high precision aerosol extinction coefficient profiles based on a RBF neural network.We choose a set of lidar signals and their corresponding aerosol extinction coefficient profiles as learning samples for network training and establish an RBF network model for aerosol extinction coefficient retrieval.Next,we correct the network output by introducing a feedback mechanism that uses the aerosol optical depth(AOD)measured by a sun photometer as the error criterion.Tests on measured signals confirm that the outputs of the proposed RBF-FB model are consistent with the desired output and have the advantages of speed and robustness.