A method for inverting the height of a mixing layer based on the spectral width of coherent lidar

Abstract

The present invention discloses a method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar, comprising the following steps: (1) reading a single signal power spectrum that has been preprocessed; (2) calculating the signal spectral width of each range gate projection height; (3) calculating the turbulence spectral width; (4) calculating the mean and variance of the turbulence spectral width within a given preset time window; and (5) identifying the height of the mixed layer based on the mean and variance profiles of the turbulence spectral width, respectively. The present invention is independent of the laser radar elevation angle, and detection results at different elevation angles can be used simultaneously to calculate and invert the height of the mixed layer, thereby having the advantages of high flexibility and high temporal and spatial resolution.

Description

Technical Field

[0001] The present invention relates to the field of laser radar technology, and in particular to a method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar. Background Art

[0002] The atmospheric mixing boundary layer refers to a layer of atmosphere where temperature, humidity, and pollutants are evenly distributed in the vertical direction due to turbulent mixing. It plays a vital role in the material and energy exchange process between the free atmosphere and the surface. The height of the mixing layer usually reflects the vertical range of the near-surface transport and diffusion process, usually between hundreds of meters and thousands of meters, depending on factors such as terrain, weather conditions, and solar radiation. The characteristics of the mixing layer height have an important impact on weather forecasting, pollution transmission, and aviation safety. Based on the assumption of uniform vertical mixing, the height of the mixing layer can be determined by the height at which aerosol density, potential temperature, and humidity decrease significantly. The traditional detection method based on sounding balloons has low temporal resolution and it is difficult to obtain the fine change characteristics of the mixing layer height. The detection method based on aerosol lidar uses the characteristics of the vertical distribution profile of aerosols to determine the height of the mixing layer, but it will be seriously disturbed when encountering complex boundary layer structures with clouds or residual aerosol layers.

[0003] Coherent wind laser radar has been widely used to detect atmospheric wind fields. Based on the inverted atmospheric motion parameters such as vertical wind speed variance and turbulence dissipation rate, the vertical distribution of atmospheric turbulence can be reflected, and the height of the mixing layer can be further determined using the threshold method. However, the calculation of vertical wind speed variance and turbulence dissipation rate is easily disturbed by large-scale vertical motion, precipitation echoes, etc., and the radar needs to work in a fixed mode. The Doppler spectral width indicates the degree of dispersion of the wind speed distribution in the laser radar detection space, which directly reflects the intensity of turbulent activity. The turbulence in the mixing layer is strong, resulting in a wide spectral width of the echo signal; while in the stable layer (i.e., the area above the mixing layer where the stratification is more stable), the turbulence intensity is weak and the spectral width of the echo signal is narrow. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a method for inverting the height of a mixed layer based on the spectral width of a coherent lidar, which can utilize the power spectrum width parameters of a coherent Doppler wind lidar to invert the atmospheric turbulent motion and determine the height of the mixed layer.

[0005] Technical solution: The method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar described in the present invention comprises the following steps:

[0006] (1) Read the power spectrum of a single signal after preprocessing: , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection;

[0007] (2) Calculate the signal spectrum width at each range gate projection height;

[0008] (3) Calculation of turbulence broadening ;

[0009] (4) Calculate the mean and variance of the turbulence spectrum width within a given preset time window;

[0010] (5) Identify the mixing layer height based on the mean and variance profiles of the turbulence spectrum width.

[0011] Furthermore, step (2) is as follows: First, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows:

[0012] ;

[0013] in, is the laser radar detection wavelength, The signal integration range is preset, which is set based on the spectral width of the typical signal. Usually, 3 times of the typical signal spectral width is extracted to reduce the integration error caused by random noise. Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows:

[0014] ;

[0015] in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction.

[0016] Furthermore, the formula of step (3) is as follows:

[0017] ;

[0018] in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset time domain width of the transmitting pulse and the time window length of the signal truncation.

[0019] Furthermore, step (4) is as follows:

[0020] For the same altitude , in the set time window The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h:

[0021] ;

[0022] ;

[0023] in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of the time period depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion.

[0024] Furthermore, step (5) includes the following steps:

[0025] (51) Use the set threshold to determine the height of the mixing layer; set the threshold of the spectral width variance and use the spectral width variance profile to identify the height of the mixing layer; the thresholds of the spectral width mean and variance can be set to 1m / s;

[0026] (52) Use the gradient method to identify the height of the mixing layer; use the gradient changes of the mean and variance profiles of the turbulence spectrum width to find the inflection point height corresponding to the fastest decrease in the mean and variance of the turbulence spectrum width with height, that is, the point of maximum gradient value, and determine it as the height of the mixing layer.

[0027] The present invention provides a mixed layer height inversion system based on coherent laser radar spectral width, comprising:

[0028] Reading module: reads the preprocessed single signal power spectrum as , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection;

[0029] Signal spectrum width module: calculates the signal spectrum width of each range gate projection height;

[0030] Turbulence Broadening Module: Calculates turbulence broadening ;

[0031] Mean and variance module: calculates the mean and variance of the turbulence spectrum width within a given preset time window;

[0032] Identification module: Identify the height of the mixing layer based on the mean and variance profiles of the turbulence spectrum width.

[0033] Furthermore, in the signal spectrum width module, the details are as follows: first, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows:

[0034] ;

[0035] in, is the laser radar detection wavelength, The signal integration range is preset, which is set based on the spectral width of the typical signal. Usually, 3 times of the typical signal spectral width is extracted to reduce the integration error caused by random noise. Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows:

[0036] ;

[0037] in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction.

[0038] Furthermore, in the turbulence broadening module, the formula is as follows:

[0039] ;

[0040] in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset time domain width of the transmitting pulse and the time window length of the signal truncation.

[0041] Furthermore, in the mean and variance modules, the details are as follows:

[0042] The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h:

[0043] ;

[0044] ;

[0045] in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of the time period depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion.

[0046] Furthermore, the identification module includes the following steps:

[0047] (51) Use the set threshold to determine the height of the mixing layer; set the threshold of the spectral width variance and use the spectral width variance profile to identify the height of the mixing layer; the thresholds of the spectral width mean and variance can be set to 1m / s;

[0048] (52) Use the gradient method to identify the height of the mixing layer; use the gradient changes of the mean and variance profiles of the turbulence spectrum width to find the inflection point height corresponding to the fastest decrease in the mean and variance of the turbulence spectrum width with height, that is, the point of maximum gradient value, and determine it as the height of the mixing layer.

[0049] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the spectral width of the coherent laser radar power spectrum is used to characterize the atmospheric turbulent motion, and then the height of the mixed layer is determined. This method is not affected by aerosol stratification, is not affected by large-scale vertical motion, does not rely on the calculation of velocity, and has the advantages of high inversion accuracy; it does not rely on the laser radar elevation angle, and the detection results of different elevation angles can be used to calculate and invert the height of the mixed layer at the same time, which has the advantages of high flexibility and high temporal and spatial resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of the present invention;

[0051] Figure 2 Is a signal power spectrum example diagram of the present invention;

[0052] Figure 3 Is a diagram showing an example of the spatiotemporal distribution of the signal spectrum width of the present invention;

[0053] Figure 4 This is an example diagram of the mean turbulence spectrum width and the mixing layer height based on the turbulence spectrum width of the present invention;

[0054] Figure 5 It is an example diagram of the turbulence spectrum width variance and the mixing layer height based on the turbulence spectrum width variance of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0056] like Figure 1 As shown, an embodiment of the present invention provides a method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar, comprising the following steps:

[0057] S1 reads the preprocessed single signal power spectrum as , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection; Figure 2As shown in the figure, the atmospheric aerosol echo power spectrum of the 50th range gate obtained by vertical detection, where the peak position represents the average Doppler frequency shift of the wind speed, and the spectrum width represents the turbulent motion intensity in the detection space. The signal comes from a coherent wind laser radar with a pulse width of 600ns, and the corresponding spatial distance resolution is 90 meters. Within this small spatial distance, the change in the average wind speed can be ignored, and it is believed that the change in the signal spectrum width comes from the contribution of turbulence. In the method of the present invention, the laser radar detection mode can be vertical detection or oblique detection, including fixed detection or multi-directional scanning detection. The detection azimuth angle and pitch angle can be a single angle or a combination of multiple angles.

[0058] S2 calculates the signal spectrum width at each range gate projection height Since the traditional method of calculating the spectral width using the second-order central moment is very susceptible to far-end noise interference and shows great fluctuations under low signal-to-noise ratio conditions, the present invention uses the equal area principle to calculate the spectral width. The specific process is as follows:

[0059] First, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows:

[0060] ;

[0061] in, is the laser radar detection wavelength, The signal integration range is preset, which is set based on the spectral width of the typical signal. Usually, 3 times of the typical signal spectral width is extracted to reduce the integration error caused by random noise. Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows:

[0062] ;

[0063] in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction. Figure 3 As shown, the spectral width obtained by continuous 18 hours of detection is given.

[0064] S3 calculation of turbulence broadening ; The formula is as follows:

[0065] ;

[0066] in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset time domain width of the transmitting pulse and the time window length of the signal truncation.

[0067] S4 calculates the mean and variance of the turbulence spectrum width within a given preset time window. The details are as follows:

[0068] For the same altitude , in the set time window The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h:

[0069] ;

[0070] ;

[0071] in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion. Without loss of generality, it can be 5-30 minutes. Figure 4 and Figure 5 The distribution of the mean and variance of the turbulence spectrum width with height calculated in a 5-minute time window is given respectively. The figure shows obvious spatiotemporal fluctuations in turbulence activity.

[0072] S5 identifies the height of the mixing layer based on the mean value and variance profile of the turbulence spectrum width, including the following steps:

[0073] (51) The height of the mixed layer is determined by using the set threshold. Within the mixed layer, turbulent activity causes the spectral width to increase. Above the mixed layer, the atmospheric motion is stable and the spectral width is small. Therefore, by setting the spectral width threshold, when the turbulent spectral width decreases from the ground to the threshold, the corresponding height is the height of the mixed layer. At the same time, turbulent activity causes the spectral width fluctuation to increase, that is, the variance increases. Therefore, by setting the threshold of the spectral width variance, the spectral width variance profile can be used to identify the height of the mixed layer.

[0074] (52) The height of the mixing layer is identified using the gradient method. The gradient of the mean and variance profiles of the turbulence spectrum width is used to find the inflection point height corresponding to the fastest decrease of the mean and variance of the turbulence spectrum width with height, and this is determined as the height of the mixing layer.

[0075] like Figure 4 and Figure 5As shown, the curves of the mixing layer height changing with time obtained based on the threshold method using the mean and variance profiles of the turbulence spectrum width are respectively given, wherein the threshold of the spectrum width mean is 1 m / s, and the threshold of the spectrum width variance is 1 m / s.

[0076] The embodiment of the present invention further provides a mixed layer height inversion system based on coherent laser radar spectral width, comprising:

[0077] Reading module: reads the preprocessed single signal power spectrum as , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection;

[0078] Signal spectrum width module: Calculate the signal spectrum width of each range gate projection height; the details are as follows: First, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows:

[0079] ;

[0080] in, is the laser radar detection wavelength, The signal integration range is preset, which is set based on the spectral width of the typical signal. Usually, 3 times of the typical signal spectral width is extracted to reduce the integration error caused by random noise. Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows:

[0081] ;

[0082] in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction.

[0083] Turbulence Broadening Module: Calculates turbulence broadening ; The formula is as follows:

[0084] ;

[0085] in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset time domain width of the transmitting pulse and the time window length of the signal truncation.

[0086] Mean and variance module: Calculates the mean and variance of the turbulence spectrum width within a given preset time window; the details are as follows:

[0087] The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h:

[0088] ;

[0089] ;

[0090] in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of the time period depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion.

[0091] Identification module: Identify the height of the mixing layer based on the mean value and variance profile of the turbulence spectrum width, including the following steps:

[0092] (51) Use the set threshold to determine the height of the mixing layer; set the threshold of the spectral width variance and use the spectral width variance profile to identify the height of the mixing layer; the thresholds of the spectral width mean and variance can be set to 1m / s;

[0093] (52) Use the gradient method to identify the height of the mixing layer; use the gradient changes of the mean and variance profiles of the turbulence spectrum width to find the inflection point height corresponding to the fastest decrease in the mean and variance of the turbulence spectrum width with height, that is, the point of maximum gradient value, and determine it as the height of the mixing layer.

Claims

1. A method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar, characterized in that: The following steps are involved: (1) Read the power spectrum of a single signal after preprocessing: , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection; (2) Calculate the signal spectrum width at each range gate projection height; (3) Calculation of turbulence spectrum width ; (4) Calculate the mean and variance of the turbulence spectrum width within a given preset time window; (5) Identify the height of the mixing layer based on the mean and variance profiles of the turbulence spectrum width, including the following steps: (51) Use the set threshold to determine the height of the mixing layer; set the threshold of the spectral width variance and use the spectral width variance profile to identify the height of the mixing layer; the thresholds of the spectral width mean and variance are both set to 1 m / s; (52) Use the gradient method to identify the height of the mixing layer; use the gradient changes of the mean and variance profiles of the turbulence spectrum width to find the inflection point height corresponding to the fastest decrease in the mean and variance of the turbulence spectrum width with height, that is, the point of maximum gradient value, and determine it as the height of the mixing layer.

2. The method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 1, characterized in that: Step (2) is as follows: First, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows: ； in, is the laser radar detection wavelength, It is the preset signal integration range; according to different detection elevation angles Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows: ； in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction.

3. The method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 2, characterized in that: The formula for step (3) is as follows: ； in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset transmission pulse time domain width and the time window length of the signal truncation respectively; for The square of .

4. The method for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 1, characterized in that: Step (4) is as follows: For the same altitude , in the set time window The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h: ； ； in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of the time period depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion.

5. A mixed layer height inversion system based on coherent laser radar spectral width, characterized in that: include: Reading module: reads the preprocessed single signal power spectrum as , where f and k represent discrete frequency points and range gates, respectively. Indicates the elevation angle of laser radar detection; Signal spectrum width module: calculates the signal spectrum width of each range gate projection height; Turbulence Spectral Width Module: Calculates the turbulence spectral width ; Mean and variance module: calculates the mean and variance of the turbulence spectrum width within a given preset time window; Identification module: Identify the height of the mixing layer based on the mean and variance profiles of the turbulence spectral width, as follows: use the set threshold to determine the height of the mixing layer; identify the height of the mixing layer using the spectral width variance profile by setting the threshold of the spectral width variance; set the thresholds of the spectral width mean and variance to 1m / s; use the gradient method to identify the height of the mixing layer; use the gradient changes of the turbulence spectral width mean and variance profiles to find the inflection point height corresponding to the fastest decrease in the turbulence spectral width mean and variance with height, i.e., the gradient maximum value point, and determine it as the mixing layer height.

6. The system for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 5, characterized in that: In the signal spectrum module, the details are as follows: First, within a given bandwidth range From the power spectrum Extract the peak , and the frequency corresponding to the peak , calculate the spectrum width of each range gate k, the formula is as follows: ； in, is the laser radar detection wavelength, It is the preset signal integration range; according to different detection elevation angles Project the detection distance information onto the vertical height to obtain the spectrum width at different heights. The formula is as follows: ； in, , is the actual radial distance between the detection target and the laser radar corresponding to the range gate k, and h is the projection height of the radial distance in the vertical direction.

7. The system for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 6, characterized in that: In the turbulence broadening module, the formula is as follows: ； in, represents the spectrum width of the emitted laser pulse, The spectrum width of the window function representing the truncation of the beat frequency signal is calculated by the preset transmission pulse time domain width and the time window length of the signal truncation respectively; for The square of .

8. The system for inverting the height of a mixed layer based on the spectral width of a coherent laser radar according to claim 5, characterized in that: In the mean and variance modules, the details are as follows: The mean turbulence spectrum width is calculated by the following formula and variance Profiles varying with height h: ； ； in, Indicates the preset time window At different moments in the time window, N represents the number of detection data in the time window; the preset time window The length of the time period depends on the temporal resolution of the data and the spatial variation scale of the atmospheric turbulent motion.

Images