A power spectrum signal processing method to improve the accuracy of lidar wind measurement

Abstract

The invention discloses a power spectrum signal processing method for improving the wind measurement accuracy of lidar, which is used for fitting power spectrum signals. The method includes: step 1, constructing a background noise curve for the original power spectrum; step 2, according to the The noise power spectrum is obtained by deducting the background noise from the power spectrum of the single distance gate one by one near and far; step 3, single Gaussian fitting is performed on the denoising power spectrum of the single distance gate, and the initial judgment parameters are obtained; step 4, the even order derivative is used Sharpen the peak-seeking method to obtain the fitted wind speed and carrier-to-noise ratio; Step 5, judge whether the single power spectrum has been processed, if so, go to Step 6, if not, go to Steps 3 to 4 to process the next distance gate; Step 6, judge Whether all power spectra have been fitted, if so, go to step 7, if not, go to steps 1 to 4 to process the next power spectrum; step 7, output the fitted wind speed and carrier-to-noise ratio, which can realize high range resolution coherent measurement Precise extraction of Doppler frequencies from wind lidar power spectra.

Description

technical field [0001] The invention relates to the field of laser radar, in particular to a power spectrum signal processing method for improving the wind measurement accuracy of laser radar. Background technique [0002] Low-altitude atmospheric wind farms are of great significance in urban pollution monitoring, wind energy development and optimization, aviation security and weather forecasting. Doppler wind Lidar is an active measurement system, mainly using the Doppler effect generated by the movement of tiny particles (aerosols or atmospheric molecules) in the atmosphere to measure the structure of the atmospheric wind field. A modern photoelectric detection technology, its main characteristics It is a non-contact real-time measurement of the wind field in the measurement area by optical methods. When measuring the wind field, the lidar scans the atmosphere in space, and determines the radial wind speed in the direction of the lidar line of sight through the Doppler fr...

AI Technical Summary

Problems solved by technology

In the past Doppler wind lidar signal processing, there are different methods to find the secondary peak or hidden peak. Double Gaussian fitting is realized by frequency difference; some researchers also use the difference between single Gaussian fitting and original power spectrum to fit a new Gaussian peak as the secondary peak, but these methods have certain limitations when extracting wind speed from power spectrum fitting. Insufficient, mainly including two reasons: one is that the secondary peak or hidden peak in the power spectrum signal is caused by uncertain noise, and its position is uncertain, so the method of fixed frequency difference cannot be used; the other is that it is caused by noise The power spectrum signal amplitude is much smaller than the peaks produced by physical phenomena such as rainfall, and the residual method is not applicable

Images