Frequency hopping multi-channel laser radar atmospheric absolute temperature detection system and method

Abstract

The application discloses a frequency hopping multi-channel laser radar atmospheric absolute temperature detection system, which comprises a frequency hopping laser emission module, a telescope receiving module, a light splitting module and a data acquisition and inversion module. The frequency hopping laser emission module generates an atmospheric backscattering light signal which is received by the telescope receiving module. The telescope receiving module receives the atmospheric backscattering signal, is coupled to a multimode optical fiber, and outputs the signal into the light splitting module through the optical fiber. The light splitting module is used for separating and extracting Rayleigh scattering signals. The data acquisition and inversion module acquires the Rayleigh scattering echo signals separated by the light splitting module, and realizes atmospheric absolute temperature detection by utilizing the relationship between Rayleigh spectrum width and atmospheric absolute temperature. The system realizes atmospheric absolute temperature detection. The application further discloses a frequency hopping multi-channel laser radar atmospheric absolute temperature detection method.

Description

Technical Field

[0001] This invention belongs to the field of atmospheric temperature detection technology, specifically relating to a frequency-hopping multi-channel lidar atmospheric absolute temperature detection system and a frequency-hopping multi-channel lidar atmospheric absolute temperature detection method. Background Technology

[0002] Atmospheric temperature is a crucial meteorological element. Vertically, the atmosphere can be divided into the troposphere, stratosphere, mesosphere, thermosphere, and exosphere. The troposphere, as the lower layer of the atmosphere, experiences intense vertical convection, leading to complex atmospheric phenomena and weather changes. Tropospheric atmospheric parameters are closely related to agricultural production, aircraft safety, and major disaster prevention; therefore, achieving high-precision measurement of these parameters has significant social value and importance. Atmospheric temperature profiles describe the vertical distribution of atmospheric temperature with altitude. These profiles play a vital role in predicting weather conditions and obtaining other atmospheric parameters.

[0003] Since the spectral linewidth of Rayleigh scattering by atmospheric molecules is related to the laser emission wavelength and atmospheric temperature, high-spectral-resolution lidar extracts the Rayleigh scattering signal from atmospheric molecules using a high-spectral-resolution spectrometer. Atmospheric temperature can then be obtained by utilizing the relationship between the Rayleigh scattering spectrum's full width at half maximum (FWHM) and temperature, or by fitting it to a theoretical model. Compared to the rotating Raman method, Rayleigh scattering has a larger scattering cross-section and a Rayleigh scattering intensity 3-4 orders of magnitude higher. It achieves a better signal-to-noise ratio without requiring a long accumulation time, and the narrowband filter suppresses solar background light, making it a promising all-day temperature detection solution. Since both Mie scattering by atmospheric aerosols and Rayleigh scattering by atmospheric molecules are elastic scattering, their scattering spectra have the same center wavelength (excitation laser wavelength). To detect changes in the Rayleigh scattering spectral width, the rejection of Mie scattering must be considered. Therefore, current high-spectral-resolution lidar temperature detection methods mainly use fixed narrowband discriminators, such as FPEs, Michelson interferometers, atomic or molecular absorption cells, to separate Mie and Rayleigh scattering signals. By using the intensity changes of different Rayleigh spectra, the Rayleigh spectral width change is obtained through differential analysis, and the relative temperature value is obtained by inversion. This is a relative temperature detection method. To obtain the absolute value, a reference temperature needs to be used.

[0004] Calibration and correction of lidar temperature profiles have always been a challenge in this field. Lidar detection suffers from blind spots, and the signal-to-noise ratio of the detection signal decreases with increasing altitude, making system calibration and correction impossible using conventional metrological standards. Currently, profiles obtained from radiosondes are generally used as a benchmark for lidar detection profile calibration and correction. However, the actual detection path of the radiosonde is significantly affected by atmospheric turbulence, horizontal wind speed and direction, leading to inconsistencies between its detection path and time and the lidar's. This easily introduces systematic errors in lidar correction, reducing the relative accuracy of high-spectral-resolution lidar temperature detection. Furthermore, since atmospheric conditions are a randomly changing physical process, generally considered irreproducible, this increases the difficulty of achieving high-precision lidar system correction. Therefore, researching atmospheric temperature profile (absolute value) detection techniques to eliminate the technical bottlenecks in system calibration and correction is currently a cutting-edge and hot topic in lidar temperature profile detection technology. Summary of the Invention

[0005] The first objective of this invention is to provide a frequency-hopping multi-channel lidar atmospheric absolute temperature detection system to achieve atmospheric absolute temperature detection.

[0006] The second objective of this invention is to provide a method for detecting atmospheric absolute temperature using a frequency-hopping multi-channel lidar.

[0007] The first technical solution adopted in this invention is a frequency-hopping multi-channel lidar atmospheric absolute temperature detection system, which includes a frequency-hopping laser emission module, a telescope receiving module, a beam splitting module, and a data acquisition and inversion module.

[0008] The frequency-hopping laser emission module generates atmospheric backscattered light signals, which are received by the telescope receiving module.

[0009] The telescope receiving module receives atmospheric backscattered signals, couples them to a multimode fiber, and outputs them through the fiber to the beam splitter module.

[0010] The beam splitter module is used to separate and extract the Rayleigh scattering signal;

[0011] The data acquisition and inversion module acquires Rayleigh scattering echo signals separated by the spectrometer module, and uses the relationship between Rayleigh spectral width and atmospheric absolute temperature to invert and realize atmospheric absolute temperature detection.

[0012] The invention is further characterized in that,

[0013] The frequency-hopping laser emission module includes a laser seed laser, an Nd:YAG pulse laser, a second-harmonic crystal, a third-harmonic crystal, a laser beam expander, and a first 45° total reflection mirror, arranged sequentially according to the laser transmission path. The frequency-hopping laser emitted by the seed laser is injected into the Nd:YAG pulse laser to generate a frequency-hopping pulse laser. After passing through the second-harmonic crystal and the third-harmonic crystal, a frequency-hopping pulse laser with a wavelength of 354.72nm and a frequency hopping frequency of 1.5GHz is generated. The laser beam is then expanded by the laser beam expander, reflected by the first 45° total reflection mirror, and perpendicularly incident into the atmosphere to generate an atmospheric backscattered light signal, which is received by the telescope receiving module.

[0014] The telescope receiving module includes a receiving telescope, a high-efficiency coupler, and a multimode fiber; atmospheric backscattered light signals are received by the receiving telescope, coupled through the high-efficiency coupler, and transmitted to the beam splitter through the multimode fiber.

[0015] The beam splitting module includes a collimating convex lens, beam splitters BS0, BS1, and BS2, a second 45° total reflection mirror, and a three-channel FPI. Atmospheric backscattered light signals from the multimode fiber are collimated by the collimating convex lens and incident on beam splitter BS0. A portion of the light is reflected into the data acquisition and inversion module as a reference signal. Another portion of the light signal is transmitted through beam splitter BS0 to beam splitter BS1. Beam splitter BS1 transmits a portion of the signal through one channel of the three-channel FPI and reflects it into the data acquisition and inversion module. The remaining signal is reflected by beam splitter BS1 to beam splitter BS2. Beam splitter BS2 reflects a portion of this signal through the second channel of the three-channel FPI and reflects it into the data acquisition and inversion module. The remaining signal is transmitted through beam splitter BS2 and reflected by the second 45° total reflection mirror into the third channel of the three-channel FPI and then into the data acquisition and inversion module.

[0016] The reflection-to-transmission ratio of beam splitter BS0 is 1:99; the reflection-to-transmission ratio of beam splitter BS1 is 70:30; and the reflection-to-transmission ratio of beam splitter BS2 is 50:50.

[0017] The data acquisition and inversion module includes photodetectors PMT0, PMT1, PMT2, and PMT3, as well as a digital oscilloscope. Photodetector PMT0 is responsible for detecting the reference signal, while photodetectors PMT1, PMT2, and PMT3 respectively detect the optical signals passing through the three channels of the three-channel FPI. All four signals are acquired and stored by the digital oscilloscope for inverting atmospheric temperature.

[0018] The second technical solution adopted in this invention is a frequency-hopping multi-channel lidar method for detecting atmospheric absolute temperature, which is implemented according to the following steps:

[0019] Step 1: Initialize the frequency-hopping multi-channel lidar atmospheric absolute temperature detection system described above;

[0020] Step 2: Use the frequency hopping laser emission module to output frequency hopping laser pulses, and use the telescope receiving module to receive atmospheric backscattered signals. After being split by the beam splitting module, the signals are collected by the corresponding photodetectors of each channel.

[0021] Step 3: Perform splicing, normalization, denoising, and deconvolution on the collected atmospheric backscattering signals to obtain the atmospheric Rayleigh-Brillouin scattering spectrum.

[0022] Step 4: Fit the atmospheric Rayleigh-Brillouin scattering spectrum obtained in Step 3 with the Rayleigh-Brillouin theoretical model;

[0023] Step 5: By fitting the scattering spectrum at each altitude, the atmospheric temperature at each altitude is obtained by inversion, and finally the atmospheric temperature profile is obtained.

[0024] The invention is further characterized in that,

[0025] Step 3 is implemented in the following steps:

[0026] The seed laser frequency was set to the initial frequency. Using the excitation signal of the Nd:YAG pulsed laser as a reference, when the pulse signal was greater than 0, the optical signals PMT1, PMT2, and PMT3 of the three measurement channels of the three-channel FPI, as well as the reference channel optical signal PMT0 between the photodetector PMT0 and the beam splitter BS0, were acquired. The PMT1, PMT2, and PMT3 signals were normalized using the PMT0 signal to eliminate errors caused by laser energy fluctuations. The acquired data was recorded using a digital oscilloscope. For the raw data acquired by the digital oscilloscope, data corresponding to the same height at different laser output frequencies were superimposed, and wavelet denoising was performed on the superimposed signal. Wiener filtering and deconvolution were then applied to the denoised data.

[0027] Step 4 is implemented in the following steps:

[0028] Step 4.1: First, determine that the frequency discriminator used in the radar system is a three-channel FPI. The ideal transmittance function T(v) of the FPI is:

[0029]

[0030] Where I0 is the maximum transmitted intensity, v FSR For the free spectral range, v FWHM The full width at half maximum (FWHM) of the transmittance curve is given by ν, where ν is the frequency of the emitted light; the actual transmittance T is the full width at half maximum (FWHM). p (v) is the convolution of the Airy function and the surface defect function D(v), which is:

[0031]

[0032] The defect function is represented as follows:

[0033]

[0034] Where, σ g The global defect parameter represents the mirror defect; the Airy function, written as a Fourier series, is:

[0035]

[0036] The FPI intensity transmittance function after convolution is:

[0037]

[0038] The signal detected by the detector is the convolution of the scattered light Rayleigh-Brillouin spectrum and the FPI transmittance, that is:

[0039]

[0040] The molecular scattering spectrum S(v) is obtained by Wiener filtering and deconvolution;

[0041] Step 4.2: When the input signal is an FP transfer function, and the desired output is an impulse function δ, then:

[0042] t(v)*G≈δ

[0043] Similarly, using the properties of convolution operations, we obtain the following formula:

[0044] I*G=S*t*G=S*δ=S

[0045] That is, the input is the measured spectrum signal, and the output approximates the true scattering spectrum signal; the error between the filter output and the desired output δ is denoted as E, and expressed as:

[0046] E=Σ(G*t-δ) 2

[0047] This is an equation about G. When we take the first partial derivative with respect to G and set it to 0, the root mean square error is minimized, yielding the Tobleitz equation:

[0048] Σr yy (mn)G(n)=r yx (m)

[0049] In the formula: r yy The autocorrelation sequence of the input function; r yx The cross-correlation sequence between the input function and the impulse function δ;

[0050] Write it in matrix form.

[0051] R yy G=R yx

[0052] That is, solve for the transfer function G, restore the original signal by Wiener filtering, process the data corresponding to each height to obtain the Rayleigh-Brillouin scattering spectrum corresponding to each height, and fit it to the Rayleigh-Brillouin line shape of the theoretical model using the least squares fitting method.

[0053] Step 4.2 uses the least squares fitting method to fit the Rayleigh-Brillouin line shape of the theoretical model. The specific method is as follows:

[0054] The Tenti-S6 model is employed. The Tenti-S6 model obtains the parameter values ​​by solving matrix equations. The scattering spectrum is obtained using the relationship between the gas density perturbation v(ξ, y) and the scattered light power spectrum S(ξ, y). The matrix elements in the solution matrix are combined to obtain the shear viscosity η and the bulk viscosity η. b The gas transport coefficient, along with the thermal conductivity ζ, is used to represent the spectral coefficient. Therefore, the spectral calculations are entirely defined by four dimensionless parameters x, y, z, and f, which are expressed as follows:

[0055]

[0056] In the formula: ω is the angular frequency of the incident light; p is the pressure; c int The internal specific heat of each molecule divided by the Boltzmann constant k B ;

[0057] By fitting the optimal matching line shape, the atmospheric temperature can be obtained by inversion.

[0058] The beneficial effects of this invention are:

[0059] The system of this invention is designed to detect absolute atmospheric temperature. The frequency-hopping multi-channel hyperspectral resolution lidar for absolute atmospheric temperature detection proposed in this invention can detect absolute atmospheric temperature and obtain atmospheric temperature profiles without the need for a reference temperature. Furthermore, since Rayleigh scattering has a larger scattering cross-section than Raman scattering, it can achieve all-weather detection. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the structure of the frequency-hopping multi-channel lidar atmospheric absolute temperature detection system of the present invention;

[0061] Figure 2 This is a schematic diagram of the receiving field of view of the telescope of the present invention;

[0062] Figure 3 This is a schematic diagram of the Mi-Rayleigh scattering spectral distribution and the three-channel FP transmittance distribution at different laser frequencies.

[0063] Figure 4This is a schematic diagram of the Rayleigh-Mille scattering spectrum of the atmosphere at different temperatures.

[0064] In the diagram, 1. Seed laser, 2. Nd:YAG pulsed laser, 3. Frequency-second harmonic crystal, 4. Frequency-third harmonic crystal, 5. Laser beam expander, 6. First 45° total reflection mirror, 7. Receiving telescope, 8. High-efficiency coupler, 9. Multimode fiber, 10. Collimating convex lens, 11. Beam splitter BS0, 12. Beam splitter BS1, 13. Beam splitter BS2, 14. Second 45° total reflection mirror, 15. Three-channel FPI, 16. Photodetector PMT0, 17. Photodetector PMT1, 18. Photodetector PMT2, 19. Photodetector PMT3, 20. Digital oscilloscope. Detailed Implementation

[0065] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0066] This invention provides a frequency-hopping multi-channel lidar system for detecting atmospheric absolute temperature, such as... Figure 1 As shown, it includes a frequency-hopping laser emission module, a telescope receiving module, a beam splitting module, and a data acquisition and inversion module;

[0067] The frequency-hopping laser emission module employs a seed-injected high-energy Nd:YAG pulsed laser, achieving frequency-hopping laser output by adjusting the seed input frequency. The frequency adjustment range is 1.5 GHz, and the full width of the atmospheric backscattered Rayleigh-Mie signal spectrum is approximately 10 GHz. Figure 4 As shown, with the help of a three-channel FPI, Rayleigh scattering spectrum sampling can be achieved, generating atmospheric backscattered light signals that are received by the telescope receiving module.

[0068] The telescope receiving module receives atmospheric backscattered signals, couples them to a multimode fiber, and outputs them through the fiber to the beam splitter module.

[0069] The spectrometer module uses a three-channel FPI to separate and extract Rayleigh-Brillouin scattering signals of atmospheric molecules;

[0070] The data acquisition and inversion module acquires Rayleigh scattering echo signals separated by the spectrometer module, and uses the relationship between Rayleigh spectral width and atmospheric absolute temperature to invert and realize atmospheric absolute temperature detection.

[0071] The frequency-hopping laser emission module includes a laser seed laser 1, an Nd:YAG pulsed laser 2, a second-harmonic crystal 3, a third-harmonic crystal 4, a laser beam expander 5, and a first 45° total reflection mirror 6, arranged sequentially along the laser transmission path. The seed laser 1 provides frequency-tunable continuous wave radiation with high frequency stability and a frequency adjustment range of 1.5 GHz, thereby achieving frequency-hopping pulsed laser output. Specifically, the frequency-hopping laser emitted by the seed laser 1 is injected into the Nd:YAG pulsed laser 2 to generate a frequency-hopping pulsed laser. After passing through the second-harmonic crystal 3 and the third-harmonic crystal 4, a frequency-hopping pulsed laser with a wavelength near 355 nm (frequency hopping 1.5 GHz, center 354.72 nm) is generated. The laser beam expander 5 then expands the beam, and after being reflected by the first 45° total reflection mirror 6, it is perpendicularly incident into the atmosphere, generating atmospheric backscattered light signals that are received by the telescope receiving module.

[0072] The telescope receiving module includes a receiving telescope 7, a high-efficiency coupler 8, and a multimode fiber 9. Atmospheric backscattered light signals are received by the receiving telescope 7, coupled through the high-efficiency coupler 8, and transmitted to the beam splitter module via the multimode fiber 9. The telescope receiving module is a large-aperture telescope responsible for receiving atmospheric backscattered signals, coupling them to the multimode fiber 9, and outputting them to the beam splitter module. Assuming timing starts from the moment the pulse front is emitted, and a return signal from the laser pulse front is detected at time t, then the pulse front is scattered at z1 = ct / 2, where c is the speed of light. Simultaneously, the received pulse trailing edge signal is scattered back at z2 = c(t-τ) / 2, where τ is the pulse width. z = z1 - z2 is called the scattering volume length, and the light signal within this volume is scattered back to the receiving system. It is worth noting that because the receiving system has a certain size and field of view (assuming the effective aperture of the telescope is D0), only scattered light within a specific solid angle can be returned and received. This solid angle can be represented by D0 / z2, as shown below. Figure 2 As shown.

[0073] The beam splitting module includes a collimating convex lens 10, beam splitters BS011, BS112, BS213, a second 45° total reflection mirror 14, and a three-channel FPI 15. The atmospheric backscattered light signal output from the multimode fiber 9 is collimated by the collimating convex lens 10 and incident on beam splitter BS011. A small portion of this signal is reflected into the photodetector PMT016 of the data acquisition and inversion module as a reference signal. The majority of the remaining light signal is transmitted through beam splitter BS011 to beam splitter BS112. Beam splitter BS112 then transmits a portion of the signal through the three-channel FPI 15. One channel of I15 is reflected into the photodetector PMT117 of the data acquisition and inversion module for acquisition; the remaining signal is reflected by beam splitter BS112 to beam splitter BS213; beam splitter BS213 reflects a portion of this signal through the second channel of the three-channel FPI15 and into the photodetector PMT218 of the data acquisition and inversion module for acquisition; the remaining signal is transmitted through beam splitter BS213 and reflected by the second 45° total reflection mirror 14 into the third channel of the three-channel FPI15 and into the photodetector PMT319 of the data acquisition and inversion module for acquisition.

[0074] The core of the beam splitter module is a three-channel FPI (Fiber Optic Injector). The light output via fiber optic cable enters beam splitter BS011, where a small portion of the reflected signal light (1% energy) is directly received by detector PMT016 as an energy monitoring channel. The remaining 99% of the signal light passes through beam splitters BS112 and BS213, and a second 1445° total internal reflection mirror, entering the three channels of the FPI. All three channels use the same coating material and thickness. Three photomultiplier tubes (PMT1, PMT2, and PMT3) are used to detect changes in the output spectral energy of the atmospheric Rayleigh scattering spectrum on the three-channel FPI15. Figure 3 As shown. Due to differences in atmospheric temperature and pressure at different altitudes, the corresponding Rayleigh-Brillouin spectral lines also differ. Therefore, a single pulse emission, after being filtered by a three-channel FP15, yields three points of the Rayleigh-Brillouin spectral line corresponding to a specific altitude at a given sampling time. Points sampled at different times correspond to Rayleigh-Brillouin scattering spectral lines at different altitudes (some may also include Mie scattering, depending on the position of the FP center relative to the laser frequency center). By accumulating multiple pulse signals to improve the signal-to-noise ratio, and by using a seed laser to change the frequency of the output pulse laser, the position of the FP center relative to the laser frequency center can be altered, thus obtaining points on the Rayleigh-Brillouin spectral lines at each altitude that differ from those sampled at the previous laser frequency. Considering that the full width of the atmospheric Rayleigh-Brillouin scattering spectrum is approximately 10 GHz, if the frequency hopping step is 100 MHz, 15 frequency values ​​are required, and 45 energy spectral points can be obtained within the half-spectrum. Compared to the commonly used single-channel frequency discriminator, the time to obtain the same energy spectral points can be reduced by a factor of three. The short detection time is highly advantageous for acquiring atmospheric parameters with continuously changing states.

[0075] The reflection-to-transmission ratio of beam splitter BS011 is 1:99; that of beam splitter BS112 is 70:30; and that of beam splitter BS213 is 50:50.

[0076] The data acquisition and inversion module includes photodetectors PMT016, PMT117, PMT218, and PMT319, as well as a digital oscilloscope 20. Photodetector PMT016 is responsible for detecting the reference signal, while photodetectors PMT117, PMT218, and PMT319 respectively detect the optical signals passing through the three channels of the three-channel FPI15. All four signals are acquired and stored by the digital oscilloscope 20 for inverting atmospheric temperature.

[0077] This invention also provides a method for detecting atmospheric absolute temperature using frequency-hopping multi-channel lidar, which is implemented according to the following steps:

[0078] Step 1: Initialize the frequency-hopping multi-channel lidar atmospheric absolute temperature detection system described above;

[0079] Step 2: Use the frequency hopping laser emission module to output frequency hopping laser pulses, and use the telescope receiving module to receive atmospheric backscattered signals. After being split by the beam splitting module, the signals are collected by the corresponding photodetectors PMT117, PMT218, and PMT319 of each channel.

[0080] Step 3: Perform splicing, normalization, denoising, and deconvolution on the collected atmospheric backscattering signals to obtain the atmospheric Rayleigh-Brillouin scattering spectrum.

[0081] Step 3 is implemented in the following steps:

[0082] The seed laser 1 frequency is set to the initial frequency. Using the excitation signal of the Nd:YAG pulsed laser 2 as a reference, when the pulse signal is greater than 0, the three measurement channel optical signals PMT1, PMT2, and PMT3 of the three-channel FP15, as well as the reference channel optical signal PMT0 between the photodetector PMT016 and the beam splitter BS011, are acquired. The PMT1, PMT2, and PMT3 signals are normalized using the PMT0 signal to eliminate errors caused by laser energy fluctuations. The acquired data is recorded using a digital oscilloscope 20. The signal-to-noise ratio is improved by accumulating multiple pulse signals. The output pulse laser frequency is changed using the seed laser frequency converter, altering the position of the FP center relative to the laser frequency center. This allows sampling of points on the Rayleigh-Brillouin spectrum at each height that differ from previous laser frequency sampling. For the raw data acquired by the digital oscilloscope 20, data corresponding to the same height at different laser output frequencies are superimposed, and wavelet denoising is performed on the superimposed signal. To eliminate the influence of the instrument's transfer function, Wiener filtering and deconvolution are applied to the denoised data.

[0083] Step 4: Fit the atmospheric Rayleigh-Brillouin scattering spectrum obtained in Step 3 with the Rayleigh-Brillouin theoretical model;

[0084] Step 4 is implemented in the following steps:

[0085] Step 4.1: First, determine that the frequency discriminator used in the radar system is a three-channel FPI. The ideal transmittance function T(v) of the FPI is:

[0086]

[0087] Where I0 is the maximum transmitted intensity, v FSR For the free spectral range (FSR), v FWHM Let T be the full width at half maximum (FWHM) of the transmittance curve, and v be the frequency of the emitted light. In reality, FPI usually has surface defects, so the actual transmittance T will vary. p (v) is the convolution of the Airy function and the surface defect function D(v), which is:

[0088]

[0089] The defect function can be considered to approximately follow a Gaussian distribution, as follows:

[0090]

[0091] Where, σ g The global defect parameter represents the mirror defect; the Airy function, written as a Fourier series, is:

[0092]

[0093] The FPI intensity transmittance function after convolution is:

[0094]

[0095] The signal detected by the detector is the convolution of the scattered light Rayleigh-Brillouin spectrum and the FPI transmittance, that is:

[0096]

[0097] The molecular scattering spectrum S(v) is obtained by Wiener filtering deconvolution. The most important step in Wiener filtering deconvolution is to obtain the transfer function G.

[0098] Step 4.2, when the noise signal N s In cases where the effect is negligible, an ideal Wiener filter can eliminate the influence of the FP transfer function. Therefore, when the input signal has an FP transfer function, and the desired output is an impulse function δ, we have:

[0099] t(v)*G≈δ

[0100] Similarly, using the properties of convolution operations, we obtain the following formula:

[0101] I*G=S*t*G=S*δ=S

[0102] That is, the input is the measured spectrum signal, and the output approximates the true scattering spectrum signal; the error between the filter output and the desired output δ is denoted as E, and expressed as:

[0103] E=Σ(G*t-δ) 2

[0104] This is an equation about G. When we take the first partial derivative with respect to G and set it to 0, the root mean square error is minimized, yielding the Tobleitz equation:

[0105] Σr yy (mn)G(n)=r yx (m)

[0106] In the formula: r yy The autocorrelation sequence of the input function; r yx The cross-correlation sequence between the input function and the impulse function δ;

[0107] Write it in matrix form.

[0108] R yy G=R yx

[0109] That is, solve for the transfer function G, restore the original signal by Wiener filtering, process the data corresponding to each height to obtain the Rayleigh-Brillouin scattering spectrum corresponding to each height, and fit it to the Rayleigh-Brillouin line shape of the theoretical model using the least squares fitting method, generally using the S6 model.

[0110] Step 4.2 uses the least squares fitting method to fit the Rayleigh-Brillouin line shape of the theoretical model. The specific method is as follows:

[0111] The Tenti-S6 model is employed. The Tenti-S6 model obtains the parameter values ​​by solving matrix equations. The scattering spectrum is obtained using the relationship between the gas density perturbation v(ξ, y) and the scattered light power spectrum S(ξ, y). The matrix elements in the solution matrix are combined to obtain the shear viscosity η and the bulk viscosity η. b The gas transport coefficient, along with the thermal conductivity ζ, is used to represent the spectral coefficient. Therefore, the spectral calculations are entirely defined by four dimensionless parameters x, y, z, and f, which are expressed as follows:

[0112]

[0113] In the formula: ω is the angular frequency of the incident light; p is the pressure; c int The internal specific heat of each molecule divided by the Boltzmann constant k B ;

[0114] Because the Tenti-S6 model was developed in the context of solid-state physics and is in best agreement with experimental data obtained in molecular gases, it is considered the best model for describing Rayleigh-Brillouin scattering spectra.

[0115] By fitting the optimal matching line shape, the atmospheric temperature can be obtained by inversion.

[0116] Step 5: By fitting the scattering spectrum at each altitude, the atmospheric temperature at each altitude is obtained by inversion, and finally the atmospheric temperature profile is obtained.

[0117] Compared to relative detection methods, the method of this invention utilizes laser scanning to obtain the entire scattering spectrum distribution. Sampling points at different times correspond to scattering spectral lines at different altitudes. Data points at different positions on the scattering spectral lines at various altitudes are acquired through laser frequency hopping output, obtaining the scattering spectral lines. Atmospheric temperature is then derived by fitting a scattering model. No response function is required, and temperature measurements at different altitudes are independent of each other, not dependent on a reference temperature. This method is an absolute temperature detection method, overcoming the technical bottleneck of difficult calibration and correction in lidar relative detection systems.

Claims

1. A frequency-hopping multi-channel lidar method for detecting atmospheric absolute temperature, characterized in that, The specific steps are as follows: Step 1: Initialize the frequency-hopping multi-channel lidar atmospheric absolute temperature detection system; In step 1, the frequency-hopping multi-channel lidar atmospheric absolute temperature detection system includes a frequency-hopping laser emission module, a telescope receiving module, a beam splitting module, and a data acquisition and inversion module; The frequency-hopping laser emission module generates atmospheric backscattered light signals, which are received by the telescope receiving module. The telescope receiving module receives atmospheric backscattered signals, couples them to a multimode fiber, and outputs them through the fiber to the beam splitter module. The beam splitter module is used to separate and extract the Rayleigh scattering signal; The data acquisition and inversion module acquires the Rayleigh scattering echo signal separated by the spectrometer module, and uses the relationship between Rayleigh spectral width and atmospheric absolute temperature to invert and realize the detection of atmospheric absolute temperature. The frequency-hopping laser emission module includes a laser seed laser (1), an Nd:YAG pulse laser (2), a second-harmonic crystal (3), a third-harmonic crystal (4), a laser beam expander (5), and a first 45° total reflection mirror (6) arranged sequentially according to the laser transmission path. The frequency-hopping laser emitted by the seed laser (1) is injected into the Nd:YAG pulse laser (2) to generate a frequency-hopping pulse laser. After passing through the second-harmonic crystal (3) and the third-harmonic crystal (4), a frequency-hopping pulse laser with a wavelength of 354.72nm and a frequency hopping frequency of 1.5GHz is generated. Then, the laser beam expander (5) expands the beam, and after being reflected by the first 45° total reflection mirror (6), it is perpendicularly injected into the atmosphere to generate atmospheric backscattered light signals, which are received by the telescope receiving module. The beam splitting module includes a collimating convex lens (10), beam splitter BS0 (11), beam splitter BS1 (12), beam splitter BS2 (13), a second 45° total reflection mirror (14), and a three-channel FPI (15). The atmospheric backscattered light signal output from the multimode fiber (9) is collimated by the collimating convex lens (10) and incident on beam splitter BS0 (11). A portion of the light signal is reflected into the data acquisition and inversion module as a reference signal; the other portion of the light signal is transmitted through beam splitter BS0 (11) to beam splitter BS1 (12). Beam splitter BS1 (12) A portion of the signal is transmitted through one channel of the three-channel FPI (15) and reflected into the data acquisition and inversion module; the remaining signal is reflected by beam splitter BS1 (12) to beam splitter BS2 (13); beam splitter BS2 (13) reflects a portion of this signal through the second channel of the three-channel FPI (15) and is reflected into the data acquisition and inversion module; the remaining portion of the signal is transmitted through beam splitter BS2 (13) and reflected by the second 45° total reflection mirror (14) into the third channel of the three-channel FPI (15) and is reflected into the data acquisition and inversion module. The data acquisition and inversion module includes photodetector PMT0 (16), photodetector PMT1 (17), photodetector PMT2 (18), photodetector PMT3 (19) and digital oscilloscope (20); photodetector PMT0 (16) is responsible for detecting the reference signal, and photodetector PMT1 (17), photodetector PMT2 (18), and photodetector PMT3 (19) respectively detect the light signals through the three channels of the three-channel FPI (15). All four signals are acquired and stored by the digital oscilloscope (20) for inverting the atmospheric temperature; Step 2: Use the frequency hopping laser emission module to output frequency hopping laser pulses, and use the telescope receiving module to receive atmospheric backscattered signals. After being split by the beam splitting module, the signals are collected by the corresponding photodetectors of each channel. Step 3: Perform splicing, normalization, denoising, and deconvolution on the collected atmospheric backscattering signals to obtain the atmospheric Rayleigh-Brillouin scattering spectrum. Step 3 is implemented in the following steps: Set the seed laser (1) frequency to the initial frequency, and take the excitation signal of the Nd:YAG pulsed laser (2) as a reference. When the pulse signal is greater than 0, start collecting the optical signals PMT1, PMT2, PMT3 of the three measurement channels of the three-channel FPI (15) and the reference channel optical signal PMT0 between the photodetector PMT0 (16) and the beam splitter BS0 (11). The PMT0 signal is used to normalize the PMT1, PMT2, and PMT3 signals to eliminate errors caused by laser energy fluctuations; the acquired data is recorded using a digital oscilloscope (20); for the raw data acquired by the digital oscilloscope (20), the data of different laser output frequencies corresponding to the same height are superimposed, and wavelet denoising is performed on the superimposed signal; Wiener filtering and deconvolution are performed on the denoised data. Step 4: Fit the atmospheric Rayleigh-Brillouin scattering spectrum obtained in Step 3 with the Rayleigh-Brillouin theoretical model; Step 5: By fitting the scattering spectrum at each altitude, the atmospheric temperature at each altitude is obtained by inversion, and finally the atmospheric temperature profile is obtained.

2. The method for detecting atmospheric absolute temperature using frequency-hopping multi-channel lidar according to claim 1, characterized in that, Step 4 is implemented in the following steps: Step 4.1: First, determine that the frequency discriminator used in the radar system is a three-channel FPI, and the ideal transmittance function of the FPI. for: in, I 0 represents the maximum transmittance. v FSR For the free spectral range, v FWHM The full width at half maximum (FWHM) of the transmittance curve. v The frequency of the emitted light; actual transmittance. Airy function and surface defect function D ( v The convolution of ) is: The defect function is represented as follows: in, σ g The global defect parameter represents the mirror defect; the Airy function, written as a Fourier series, is: The FPI intensity transmittance function after convolution is: The signal detected by the detector is the convolution of the scattered light Rayleigh-Brillouin spectrum and the FPI transmittance, that is: Molecular scattering spectroscopy S ( v It is obtained by Wiener filtering and deconvolution; Step 4.2: When the input signal is an FP transfer function, the desired output is an impulse function. δ Then there are: Similarly, using the properties of convolution operations, we obtain the following formula: That is, the input is the measured spectrum signal, and the output approximates the true scattering spectrum signal; the error between the filter output and the desired output δ is denoted as... E , is represented as: This is about G The equation, when for G When the first-order partial derivative is taken and set to zero, the root mean square error is minimized, yielding the Tobleitz equation: In the formula: r yy The autocorrelation sequence of the input function; r yx Input function and impulse function δ Cross-correlation sequences; Write it in matrix form. That is, solving the transfer function G The original signal was restored by Wiener filtering. By processing the data corresponding to each height, the Rayleigh-Brillouin scattering spectra corresponding to each height were obtained. The least squares fitting method was then used to fit the Rayleigh-Brillouin line shape of the theoretical model.

3. The method for detecting atmospheric absolute temperature using frequency-hopping multi-channel lidar according to claim 1, characterized in that, Step 4.2 uses the least squares fitting method to fit the Rayleigh-Brillouin line shape of the theoretical model. The specific method is as follows: The Tenti-S6 model is used. The Tenti-S6 model obtains the parameter values ​​by solving matrix equations and utilizing gas density perturbations. v ( ξ , y ) and scattered light power spectrum S ( ξ , y The scattering spectrum is obtained from the relationship between the matrix elements and the shear viscosity. η Volume viscosity η b and thermal conductivity ζ The gas transport coefficient is represented by the , therefore the spectral calculation is entirely based on four dimensionless parameters. x , y , z , f The definitions are as follows: In the formula: ω ω is the angular frequency of the incident light; p Pressure; c int The internal specific heat of each molecule divided by the Boltzmann constant k B .

Images