| In recent years,smog and ozone pollution have increased throughout the country,and lidar technology has also been widely used to monitor aerosols and ozone.However,for climate prediction,atmospheric model verification,and process analysis of regional transmissions,only have local aerosol and ozone information is insufficient.Therefore,some new ground-based laser radar networks in China have been applied to the study of particulate matter and ozone in regional transmission.However,due to the short application time of these new lidar observation networks,there is currently no industry standard for reference.The design concept and structural specifications of each manufacturer are not uniform.Moreover,the design concept and structural specifications of each manufacturer are not uniform,and the same manufacturer also has slight differences in the process of equipment assembly,resulting in uneven overall data quality.At the same time,due to the complexity of the laser radar system itself,the factors affecting the quality of the laser radar data are various,and there is no unified and effective data quality control system for the various fault sources that cause data errors.Radar data is quality controlled,and the quality of the data directly affects the efficiency and accuracy of radar data in various business applications.At the same time,hardware failures are also difficult to avoid due to factors such as the working environment and the quality of accessories.When a hardware failure occurs in the radar,it will directly lead to abnormal echo patterns and echo strengths,and even cause equipment downtime.The emergence of this situation has a great impact on data quality,and data quality control is the fundamental guarantee for the promotion and in-depth study of the laser radar business.Therefore,in order to improve the quality of the lidar data as a whole,it is necessary to study the data quality control of the laser radar network.For the ground-based scatter laser radar data and differential lidar networking data,this paper mainly analyzes the possible hardware error sources of rice scattering laser radar and differential laser radar,and studies the impact of various error sources on radar data quality.Then,this paper mainly analyzes the different types of hardware fault data appearing at different sites for rice scattering lidar and differential lidar.Firstly,combined with the field maintenance records of different sites,the fault types of the two types of lidar are separately counted,the fault data is extracted,and the corresponding data without hardware faults is extracted,and the training set and the test set are respectively established;Then,the characteristics of abnormal echo data and hardware without hardware faults are analyzed.Due to the complexity of the laser radar system,the automatic identification method of hardware fault data based on fuzzy logic is established.The method of reducing the false positive rate and testing the method in the test set.After testing,the data recognition rate of the rice scatter laser radar hardware failure data reached 94.6%,and the false positive rate was only 1.5%.The hardware fault data recognition rate of differential lidar reached 94.8%,and the false positive rate was 2.3%.Finally,the application of this algorithm and the identification of unknown hardware fault data have achieved good results.This proves the feasibility of the hardware fault data automatic identification algorithm.At the same time,the method can also monitor the radar running status in real time,and provide support for timely detection of faults,troubleshooting,and ensuring the normal operation of the instrument and data integrity.Finally,a method of data correction using Raman radar in the process of particle lidar networking is proposed,and a linear function of radar ratio and extinction coefficient,pollutant data,and meteorological data is constructed.By comparison and verification,this method is proved The data correction in radar network observation has good results. |
PDF Full Text Request