CN-121276478-B - Laser radar wind field inversion method based on filtering speed volume processing algorithm

CN121276478BCN 121276478 BCN121276478 BCN 121276478BCN-121276478-B

Abstract

The embodiment of the disclosure provides a laser radar wind field inversion method based on a filtering speed volume processing algorithm, which is applied to the technical field of laser radars. The method comprises the steps of obtaining observation data base data of each voxel in a laser radar wind field, inverting the observation data base data, wherein the observation data base data comprise signal to noise ratios at observation points in the voxel and are used for calculating average signal to noise ratios in the voxel, inverting a three-dimensional wind vector in the voxel based on a speed volume processing algorithm when the average signal to noise ratio is greater than or equal to a preset signal to noise ratio switching threshold value, and otherwise inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm. In this way, the three-dimensional wind vector can be inverted by adopting a speed volume processing algorithm for the near-field wind field data inversion with higher signal to noise ratio to improve the calculation efficiency, and the three-dimensional wind vector is inverted based on a preset filtering speed volume processing algorithm after the signal to noise ratio of a far field is reduced to a preset signal to noise ratio switching threshold value, so that the laser radar wind field inversion precision is improved.

Inventors

BU ZHICHAO
CHEN YUBAO
DAI YARU
SHAO NAN

Assignees

中国气象局气象探测中心

Dates

Publication Date: 20260505
Application Date: 20250822

Claims (7)

1. The laser radar wind field inversion method based on the filtering speed volume processing algorithm is characterized by comprising the following steps of: Obtaining observation data base data of each voxel inverted by a laser radar wind field, wherein the observation data base data comprises signal to noise ratios at each observation point in the voxel, and the signal to noise ratios are used for calculating average signal to noise ratios in the voxel; inverting the three-dimensional wind vector in the voxel based on a speed volume processing algorithm when the average signal-to-noise ratio is greater than or equal to a preset signal-to-noise ratio switching threshold value; inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm when the average signal-to-noise ratio is smaller than the preset signal-to-noise ratio switching threshold value; Inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm comprises inverting the three-dimensional wind vector in the voxel based on the preset filtering speed volume processing algorithm, synchronously calculating the radial wind speed bad estimation duty ratio, and stopping the wind field inversion in the current radial direction until the radial wind speed bad estimation duty ratio in the voxel is greater than or equal to a preset bad estimation duty ratio threshold value; the optimization objective function constructed in the preset filtering speed volume processing algorithm comprises the following steps: Wherein, the Represents an optimized objective function inverted based on a preset filtering speed volume processing algorithm in the voxel, An optimized variable matrix representing 12 components, Representing the three-dimensional wind vector to be inverted in the voxel, Represents the measured radial wind velocity at each observation point in the voxel, Representing the radial direction vector corresponding to each observation data point in the voxel, Representing Doppler signal peak standard deviation spectrum width on echo signal power spectrum; the calculation process of the radial wind speed estimation duty ratio in the voxel comprises the following steps: Wherein, the Represents the radial wind speed estimation duty cycle in the voxel, Represents the measured radial wind velocity at each observation point in the voxel, Representing inverted three-dimensional wind vectors Radial wind speed, N, resolved along the laser beam direction at each observation point within the voxel Representing satisfaction within a voxel The number of observation points for the condition, N total , represents the total number of observation points in the voxel, Representing the standard deviation spectrum width of the signal peaks on the signal power spectrum.
2. The method of claim 1, wherein the determining of the preset signal-to-noise ratio switching threshold comprises: establishing a speed volume processing algorithm inversion error model through a numerical simulation algorithm, and determining a first preset error threshold; Determining a preset signal-to-noise ratio switching threshold according to the first preset error threshold; Wherein the velocity-volume processing algorithm inverts an error model comprising: E1 =f1 (SNR, other configuration parameters); b1 =f2 (SNR, other configuration parameters); The method comprises the steps of E1 representing root mean square error of inversion wind speed in a speed volume processing algorithm inversion error model, B1 representing systematic deviation of inversion wind speed in a speed volume processing algorithm inversion error model, SNR representing signal to noise ratio, other configuration parameters including equipment operation parameters, atmosphere state parameters and algorithm configuration parameters, and inverting a three-dimensional wind vector by applying the speed volume processing algorithm in a voxel.
3. The method of claim 1, wherein the determining of the preset bad estimate duty cycle threshold comprises: Establishing a filtering speed volume processing algorithm inversion error model through a numerical simulation algorithm, and determining a second preset error threshold; determining a preset bad estimation duty ratio threshold according to the second preset error threshold; the filtering speed volume processing algorithm inversion error model comprises the following steps: e2 =f3 (b, other configuration parameters); b2 =f4 (b, other configuration parameters); wherein E2 represents the root mean square error of the inversion wind speed in the inversion error model of the filtering speed volume processing algorithm, B2 represents the systematic deviation of the inversion wind speed in the inversion error model of the filtering speed volume processing algorithm, The method comprises the steps of representing the radial wind speed estimation duty ratio in the voxel, and inverting the three-dimensional wind vector by applying a filtering speed volume processing algorithm in the voxel, wherein other configuration parameters comprise equipment operation parameters, atmosphere state parameters and algorithm configuration parameters.
4. A method according to any one of claims 1 to 3, further comprising: When inversion is carried out on the wind field in the voxel by using the preset filtering speed volume processing algorithm, a search interval is defined by taking an inversion three-dimensional wind vector of a radially previous distance library as a center, the three-dimensional wind vector in the voxel is inverted only in the search interval, and the range width of the search interval is determined by calculating inversion three-dimensional wind vector component difference values of all adjacent distance libraries within the radial distance of the current observation point position to be inverted, solving the maximum value of the absolute value of the inversion three-dimensional wind vector component difference values and multiplying the maximum value by an amplification factor.
5. A laser radar wind field inversion device based on a filtering speed volume processing algorithm is characterized by comprising: the acquisition module is used for acquiring observation data base data of each voxel inverted by the laser radar wind field, wherein the observation data base data comprises signal to noise ratios at each observation point in the voxel, and is used for calculating the average signal to noise ratio in the voxel; The inversion module is used for inverting the three-dimensional wind vector in the voxel based on a speed volume processing algorithm when the average signal-to-noise ratio is greater than or equal to a preset signal-to-noise ratio switching threshold value; The inversion module is further used for inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm when the average signal-to-noise ratio is smaller than the preset signal-to-noise ratio switching threshold value; The inversion module is further specifically configured to invert the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm, and synchronously calculate a radial wind speed bad estimation duty cycle until the radial wind speed bad estimation duty cycle in the voxel is greater than or equal to a preset bad estimation duty cycle threshold value, and terminate wind field inversion in the current radial direction; the optimization objective function constructed in the preset filtering speed volume processing algorithm comprises the following steps: Wherein, the Represents an optimized objective function inverted based on a preset filtering speed volume processing algorithm in the voxel, An optimized variable matrix representing 12 components, Representing the three-dimensional wind vector to be inverted in the voxel, Represents the measured radial wind velocity at each observation point in the voxel, Representing the radial direction vector corresponding to each observation data point in the voxel, Representing Doppler signal peak standard deviation spectrum width on echo signal power spectrum; the calculation process of the radial wind speed estimation duty ratio in the voxel comprises the following steps: Wherein, the Represents the radial wind speed estimation duty cycle in the voxel, Represents the measured radial wind velocity at each observation point in the voxel, Representing inverted three-dimensional wind vectors Radial wind speed, N, resolved along the laser beam direction at each observation point within the voxel Representing satisfaction within a voxel The number of observation points for the condition, N total , represents the total number of observation points in the voxel, Representing the standard deviation spectrum width of the signal peaks on the signal power spectrum.
6. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-4.
7. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-4.

Description

Laser radar wind field inversion method based on filtering speed volume processing algorithm Technical Field The disclosure relates to the technical field of data processing, in particular to the technical field of laser radar, and specifically relates to a laser radar wind field inversion method based on a filtering speed volume processing algorithm. Background In a speed-volume processing algorithm (Volume Velocity Processing, VVP) for inverting a three-dimensional wind field from radial wind speed data of a single radar, a system of overdetermined linear equations is solved by adopting a least square method for data points in analysis voxels, and the goal is to find a set of values of wind field components so that the sum of squares of residuals of radial data calculated values and actual observed values of all observed points is minimum, namely an optimized objective function constructed by the least square method is the sum of squares of residuals of radial wind speeds. However, through numerical simulation analysis, the least square method based on such an optimized objective function described above continuously increases the wind field inversion error when the Signal-to-Noise Ratio (SNR) is small and continuously decreases. This is because as SNR decreases, the narrowband-CNR also decreases, and noise spur amplitudes that are uniformly distributed over the signal power spectrum have a greater probability of exceeding the true doppler signal peak (the spectral pattern is gaussian) and are erroneously identified as signal peaks by the wind speed estimation algorithm, resulting in a false estimate of the radial wind speed, which is known as a "bad estimate". The smaller the SNR, the higher the radial wind speed bad estimation duty ratio in the analysis voxel, the larger the wind field error inverted by the least square method, which is the disadvantage of the least square inversion method based on the residual square sum as the optimization objective function. Disclosure of Invention The disclosure provides a laser radar wind field inversion method, a device, equipment and a storage medium based on a filtering speed volume processing algorithm. According to a first aspect of the present disclosure, a laser radar wind field inversion method based on a filtering speed volume processing algorithm is provided. The method comprises the following steps: Obtaining observation data base data of each voxel inverted by a laser radar wind field, wherein the observation data base data comprises signal to noise ratios at each observation point in the voxel, and the signal to noise ratios are used for calculating average signal to noise ratios in the voxel; inverting the three-dimensional wind vector in the voxel based on a speed volume processing algorithm when the average signal-to-noise ratio is greater than or equal to a preset signal-to-noise ratio switching threshold value; And inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm when the average signal-to-noise ratio is smaller than the preset signal-to-noise ratio switching threshold value. In the aspect and any possible implementation manner described above, there is further provided an implementation manner, where inverting the three-dimensional wind vector in the voxel based on the preset filtering speed volume processing algorithm includes: And inverting the three-dimensional wind vector in the voxel based on a preset filtering speed volume processing algorithm, and synchronously calculating the radial wind speed bad estimation duty ratio until the radial wind speed bad estimation duty ratio in the voxel is greater than or equal to a preset bad estimation duty ratio threshold value, and terminating the wind field inversion in the current radial direction. In the aspect and any possible implementation manner as described above, there is further provided an implementation manner, where the optimizing objective function constructed in the preset filtering speed volume processing algorithm includes: Wherein Q FVVO (V) represents an optimized objective function inverted based on a preset filtering speed and volume processing algorithm in the voxel, u represents an optimized variable matrix of 12 components, V i represents a three-dimensional wind vector to be inverted in the voxel, Representing the measured radial wind speed at each observation point in the voxel, s i representing the radial azimuth vector corresponding to each observation point in the voxel, σ V representing the Doppler signal peak standard deviation spectrum width on the echo signal power spectrum. In the aspect and any possible implementation manner as described above, there is further provided an implementation manner, where the determining process of the preset signal-to-noise ratio switching threshold includes: establishing a speed volume processing algorithm inversion error model through a numerical simulation algorithm,