Research On Wind Measurement System Based On Non-doppler Lidar

In atmospheric observations,the surface wind speed and its direction are the main concerns.In view of the problems existing in the current wind measuring devices on the market—the traditional mechanical wind measuring devices are easily damaged due to long-term exposure to the outdoor environment,and the Doppler lidar wind measuring devices have too high results and cannot be widely used.Two non-Doppler lidar-based wind measurement systems for wind speed measurement.It can be seen from the experimental results that the proposed wind measurement system has a certain degree of reliability,but there are still many areas that can be improved.The specific research is as follows:(1)The background of the selected topic is introduced,several common methods and principles for measuring wind speed are introduced,and their advantages and disadvantages are analyzed.The research progress at home and abroad is described.(2)It mainly introduces the relevant theoretical basis of the experimental research based on the non-Doppler lidar wind measurement system.First,it briefly describes the composition of atmospheric aerosols,and then introduces the characteristics of several traditional Chinese medicines that use non-Doppler lidar wind measurement systems for atmospheric measurement.Combining the research content of this article,through the Mie scattering characteristics and combining the basic principles of the two proposed non-Doppler wind measurement methods,the theoretical basis of using non-Doppler lidar for wind measurement is explained.(3)A non-Doppler lidar wind measurement system based on convolutional neural network is proposed for the measurement of low-speed wind fields.The system uses CCD to capture the lidar side scatter map of aerosol particles,and realizes quantitative wind speed measurement according to the trajectory characteristics of aerosol particles at different wind speeds.In the experiment,there are three wind speeds,0m/s,1m/s,and 3m/s,respectively,and 300 pictures are taken under each wind speed.The convolutional neural network extracts the features of the aerosol particle motion trajectory map after image preprocessing and generates the wind speed measurement model.The accuracy of the 100 th training sample and verification sample is 0.92 and 0.93,respectively.Using the generated wind speed measurement model to conduct experiments on the test samples,the accuracy rate reaches0.84.(4)A non-Doppler lidar wind measurement system based on dual laser beams is proposed.The system uses an industrial camera to capture the light intensity map of the forward scattered echo signal,and obtains the echo signal variation map of the two laser beams after image processing.According to the similarity relationship between the two pictures,quantitative wind speed measurement is realized.In the experiment,the wind speed was controlled at 3m/s,and 1000 forward scattered echo signal pictures were collected by an industrial camera.After a series of processing,the SSIM change map and difference map were obtained.It can be seen from the SSIM change graph and difference graph that when the number of moving frames is 4,the echo signal change graphs of the two laser beams have the highest similarity.That is to say,the time it takes for the wind to move from the left laser to the right laser is 0.04 s.The distance between the two laser beams is 12 cm,so the wind speed measured by the wind measuring system can be obtained as 3m/s,which is consistent with the wind speed set in the experiment.It shows that the wind measurement system has a certain degree of accuracy.