A shipborne marine atmospheric boundary layer refractive index profile detection system and method

Abstract

The application discloses a shipborne marine atmospheric boundary layer refractive index profile detection system and method, relates to the technical field of marine atmospheric environment monitoring and electromagnetic wave propagation guarantee, and comprises the following steps: acquiring sea-air downward infrared radiation spectrum data and sea surface air temperature, relative humidity and air pressure data; constructing a background atmospheric profile by using reanalysis data, and correcting the initial profile bottom layer by using the sea surface observation air temperature, relative humidity and air pressure data to form a prior state vector; constructing an inversion cost function containing an observation residual term, a background prior term and a sea surface boundary constraint term, and inverting to obtain an atmospheric temperature profile and a relative humidity profile; and calculating an atmospheric refractive index profile according to the temperature profile, the relative humidity profile and the air pressure profile. The application can continuously acquire the marine atmospheric refractive index profile under the shipborne dynamic observation condition, and provides reliable data support for marine electromagnetic propagation environment analysis and waveguide diagnosis.

Description

Technical Field

[0001] This invention relates to the field of marine atmospheric environment monitoring and electromagnetic wave propagation protection technology, and in particular to a shipborne marine atmospheric boundary layer refractive index profile detection system and method. Background Technology

[0002] The refractive index structure of the marine atmospheric boundary layer is a key environmental factor affecting radar detection performance, wireless communication quality, maritime navigation accuracy, and electronic countermeasures effectiveness. The atmospheric refractive index profile determines the propagation path of electromagnetic waves in the marine atmosphere. Particularly under anomalous refraction conditions such as evaporation waveguides, surface waveguides, and raised waveguides, the propagation path of electromagnetic waves changes significantly, thus affecting radar detection range, communication coverage, and target identification effectiveness. Therefore, obtaining high-precision, continuous marine atmospheric boundary layer refractive index profiles is of great significance for marine environmental monitoring and maritime electromagnetic support.

[0003] Existing methods for detecting atmospheric refractive index at sea mainly include direct radiosonde measurement, meteorological gradient tower observation, radar inversion, and microwave radiometer remote sensing inversion. Radiosondes can provide high-precision temperature, humidity, and pressure profiles, but they suffer from high single-use costs, poor continuous observation capabilities, and significant susceptibility to sea conditions and weather. While meteorological gradient towers can acquire near-surface parameters over long periods, their height limits their coverage of the entire boundary layer. Radar inversion methods often rely on specific propagation conditions and models, limiting their applicability. Although microwave radiometers can achieve continuous observation, their vertical resolution and inversion accuracy in the lower atmosphere near the sea surface are typically insufficient.

[0004] Atmospheric Emitted Radiance Interferometer (AERI) can acquire atmospheric thermodynamic information through hyperspectral infrared radiation observations, providing an effective means for inverting atmospheric temperature and humidity profiles. However, due to the weak weighting function of infrared radiation on the near-surface layer near the sea surface, directly using AERI for low-altitude temperature and humidity inversion can easily lead to problems such as insufficient bottom-layer constraints and unstable near-sea-surface gradients, thus affecting the accuracy of refractive index profile calculations.

[0005] Therefore, how to effectively incorporate measured micrometeorological parameters from the sea surface into the infrared hyperspectral inversion process under shipborne dynamic observation conditions, so as to improve the accuracy of atmospheric parameter inversion near the sea surface and further obtain high-precision atmospheric refractive index profiles, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] The purpose of this invention is to provide a shipborne marine atmospheric boundary layer refractive index profile detection system and method to solve the problem of insufficient stability of temperature and humidity inversion in the low-altitude layer near the sea surface in the prior art, which in turn affects the accuracy of refractive index profile calculation.

[0007] The technical solution of this invention is as follows: a shipborne marine atmospheric boundary layer refractive index profile detection system, including an atmospheric emitted radiation interferometer, used to acquire sea and air downlink infrared radiation spectrum data;

[0008] Shipborne micrometeorological observation module is used to measure temperature, relative humidity and air pressure at the sea surface in real time;

[0009] The data processing module, connected to the atmospheric emitted radiation interferometer and the shipborne micrometeorological observation module, is used to construct observation vectors based on infrared radiation spectral data; construct background atmospheric profiles based on reanalysis data; and correct the underlying data using measured sea surface air temperature, relative humidity, and air pressure data to construct a priori state vectors. An inversion cost function is established, which includes observation residuals, background priors, and sea surface boundary constraints. Atmospheric temperature profiles and relative humidity profiles are obtained through inversion. Atmospheric refractive index profiles are calculated based on the temperature profiles, relative humidity profiles, and pressure profiles.

[0010] The aforementioned shipborne marine atmospheric boundary layer refractive index profile detection system further includes an attitude stabilization module and a navigation information acquisition module. The navigation information acquisition module is used to acquire hull roll, pitch, and yaw information, and the attitude stabilization module is used to perform attitude compensation based on the observation field of view of the atmospheric emission radiation interferometer.

[0011] The aforementioned shipborne marine atmospheric boundary layer refractive index profile detection system further includes a cloud and sky identification and quality control module, which is used to determine the brightness temperature value and its time series fluctuation standard deviation of the characteristic channel in the window region of the infrared radiation spectrum; when the brightness temperature value exceeds a preset brightness temperature threshold, or the time series fluctuation standard deviation exceeds a preset fluctuation threshold, it determines whether the current observation data is valid, and stops the inversion or uses the valid result from the previous moment when the observation data is invalid.

[0012] The aforementioned shipborne marine atmospheric boundary layer refractive index profile detection system also includes an anti-salt fog thermal curtain device, a thermal curtain air source and temperature control module. The anti-salt fog thermal curtain device is installed outside the optical observation window of the atmospheric emission radiation interferometer to prevent contamination and condensation on the optical observation window. The thermal curtain air source and temperature control module is used to provide heated compressed air for the anti-salt fog thermal curtain device.

[0013] The aforementioned shipborne marine atmospheric boundary layer refractive index profile detection system, specifically the anti-salt spray thermal air curtain device, is an annular jet nozzle located outside the optical observation window of the atmospheric emission radiation interferometer. The thermal air curtain gas source and temperature control module heats the compressed air to 45℃-50℃ and then ejects it from the annular jet nozzle.

[0014] A shipborne method for detecting the refractive index profile of the marine atmospheric boundary layer based on the above system includes the following steps:

[0015] Step 1: Obtain sea-air downlink infrared radiation spectrum data using an atmospheric emitted radiation interferometer, and simultaneously acquire sea surface air temperature, relative humidity, and air pressure data measured by the shipborne micro-meteorological observation module;

[0016] Step 2: Construct a background atmospheric profile using the temperature, relative humidity, and air pressure profiles from the reanalysis data. Replace the lowest layer background parameters with measured sea surface temperature, relative humidity, and air pressure. Smooth the background profile using linear interpolation within the range from the sea surface to a preset transition height to form a priori state vector. The prior state vector serves as the background prior vector for the optimal estimation inversion in step 3, and also as the initial state vector for the iteration in the inversion solution.

[0017] Step 3: Construct an observation vector y based on the infrared radiation spectral data, and establish an inversion cost function that includes observation residuals, background priors, and sea surface boundary constraints.

[0018] ;

[0019] in, Let be the atmospheric state vector to be solved. For the forward radiative transfer model, The observation error covariance matrix, The prior background covariance matrix, The parameters for the lowest layer of the inverted profile are... These are measured parameters from the sea surface. The weights are the constraints for the sea surface boundary.

[0020] Step 4: Minimize the cost function using the optimal estimation method to obtain the atmospheric temperature profile and relative humidity profile; stop iterating when the maximum number of iterations is reached.

[0021] Step 5: Calculate the water vapor partial pressure based on the temperature profile, relative humidity profile, and pressure profile obtained in Step 4, and further calculate the atmospheric refractive index profile; wherein, the pressure profile is obtained from the background pressure profile in the reanalysis data, and the bottom layer is corrected using the measured sea surface pressure as the bottom boundary condition.

[0022] In the aforementioned shipborne marine atmospheric boundary layer refractive index profile detection method, step 3 involves constructing a low-dimensional embedding space from the infrared radiation spectral data using the t-SNE method before performing optimal estimation inversion. The real-time acquired infrared radiation spectral data is then projected onto this low-dimensional embedding space according to a pre-established mapping relationship to form an observation vector. .

[0023] In the above-mentioned shipborne marine atmospheric boundary layer refractive index profile detection method, the sea surface boundary constraint term in step 3 is used to constrain the lowest layer temperature or relative humidity parameter in the state vector to be optimized by using the measured sea surface temperature or relative humidity parameter as the lowest layer boundary condition.

[0024] The aforementioned shipborne marine atmospheric boundary layer refractive index profile detection method further includes a low-altitude vertical smoothing constraint term in step 3. The low-altitude vertical smoothing constraint term applies to the inter-layer differences in temperature or relative humidity parameters between adjacent altitude layers within the range from the sea surface to a preset altitude, in order to suppress non-physical oscillations in the low-altitude region inversion results and maintain profile continuity. The specific cost function is expressed as follows:

[0025] .

[0026] Beneficial effects

[0027] Compared with the prior art, the present invention has the following beneficial effects: (1) By introducing the measured micro-meteorological parameters of the sea surface into the inversion process and using them as the lowest layer boundary constraints, the stability of the inversion of temperature and relative humidity in the low-altitude layer near the sea surface can be effectively improved; (2) By using the reanalysis background profile and the measured sea surface data to jointly construct the prior state vector, the initial state of the inversion can be made closer to the actual marine boundary layer environment, thereby improving the efficiency of iterative solution; (3) By setting the low-altitude vertical smoothing constraint term, the non-physical oscillations in the inversion results of the low-altitude region near the sea surface can be suppressed, so that the inversion profile maintains good physical continuity in the key low-altitude layer; (4) The present invention can continuously acquire the refractive index profile of the marine atmosphere under shipborne dynamic observation conditions, providing reliable data support for marine electromagnetic propagation environment analysis and waveguide diagnosis. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the process of this invention;

[0029] Figure 2 This is a schematic diagram of the principle of the anti-salt spray thermal curtain device of the present invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] This invention provides a shipborne marine atmospheric boundary layer refractive index profile detection system, including an atmospheric emission radiation interferometer, a shipborne micro-meteorological observation module, a data processing module, and optionally an attitude stabilization module, a cloud and sky identification and quality control module, and an optical window protection device.

[0032] The atmospheric emitted radiation interferometer is installed in an open location such as the ship's deck or mast platform to acquire downlink infrared radiation spectral data over the sea and air. Preferably, the atmospheric emitted radiation interferometer is a high-spectral-resolution infrared interferometer with an operating frequency band covering 3~19μm and a spectral resolution better than 1cm. -1 .

[0033] The shipborne micrometeorological observation module is installed near the ship's deck to measure air temperature, relative humidity, and air pressure at the sea surface in real time. Preferably, the shipborne micrometeorological observation module can also measure sea surface temperature to help construct a priori conditions near the sea surface.

[0034] The data processing module is connected to the atmospheric emitted radiation interferometer and the shipborne micro-meteorological observation module. It is used to receive infrared radiation spectrum data and sea surface micro-meteorological observation data, construct observation vectors and prior state vectors, establish an inversion cost function, invert the atmospheric temperature profile and relative humidity profile, and further calculate the atmospheric refractive index profile.

[0035] Preferably, the system further includes a navigation information acquisition module and an attitude stabilization module. The navigation information acquisition module is used to acquire hull roll, pitch, and yaw information. The attitude stabilization module includes a three-axis stabilization gimbal and an inertial measurement unit (IMU), and is used to perform attitude compensation on the observation field of the atmospheric emission radiation interferometer based on the hull roll, pitch, and yaw information provided by the navigation information acquisition module. The IMU is used to acquire the motion state information of the gimbal itself to assist in achieving stable control of the observation field of view.

[0036] Preferably, the system further includes a cloud and sky identification and quality control module, which is used to determine whether the current observation data is valid based on the brightness temperature value of the characteristic channel in the window region of the infrared radiation spectrum and its time series fluctuation standard deviation, and to stop the inversion or use the valid result from the previous moment when the observation data is invalid.

[0037] Specifically, the cloud and sky recognition and quality control module selects one or more feature channels in the infrared atmospheric window region and converts their radiance into equivalent brightness temperature. The standard deviation of the fluctuation was calculated for the brightness temperature sequence over several consecutive time periods. .when or If the condition is met, the current observation data is deemed invalid; otherwise, the current observation data is deemed valid. The equivalent brightness temperature of the characteristic channel in the window area. The standard deviation of the brightness temperature sequence over a continuous period of time is given. To preset the brightness temperature threshold, The preset brightness temperature threshold and the preset fluctuation threshold can be preset based on the instrument noise level, marine environmental background, and historical observation statistics.

[0038] Preferably, the system further includes an anti-salt spray thermal curtain device, a thermal curtain air source and temperature control module. The anti-salt spray thermal curtain device is installed outside the optical observation window of the atmospheric emission radiation interferometer to prevent contamination and condensation on the optical observation window. The thermal curtain air source and temperature control module provides heated compressed air for the anti-salt spray thermal curtain.

[0039] Specifically, the anti-salt spray thermal air curtain device is located at the annular high-pressure jet nozzle outside the optical observation window of the atmospheric emission radiation interferometer. The thermal air curtain air source and temperature control module heat the compressed air to 45℃-50℃ and then eject it from the annular high-pressure jet nozzle. Figure 2 As shown.

[0040] Based on the above system, this embodiment also discloses a shipborne method for detecting the refractive index profile of the marine atmospheric boundary layer, the specific process of which is as follows: Figure 1 As shown, it includes the following steps:

[0041] Step 1, Adaptive Data Acquisition: Obtain down-the-sea infrared radiation spectrum data through an atmospheric emitted radiation interferometer, and simultaneously acquire sea surface air temperature, relative humidity and air pressure data measured by the shipborne micro-meteorological observation module.

[0042] The system monitors the field-of-view deviation after attitude compensation in real time, triggering infrared radiation spectrum acquisition only when the field-of-view deviation is less than a preset threshold. Simultaneously, a cloud and sky identification and quality control module is used to determine the validity of the observation data. This module extracts atmospheric window region feature channel information from the infrared radiation spectrum, converts it into equivalent brightness temperature, and combines this with the brightness temperature threshold and its time-series fluctuation characteristics to determine whether the current observation is affected by clouds or precipitation. If the observation is determined to be invalid, the current inversion is stopped or the valid result from the previous moment is used.

[0043] Step 2, Constructing the Prior State Vector: Using the temperature, relative humidity, and air pressure profiles from the reanalysis data as the background atmospheric profile, and replacing the lowest layer background parameters with measured sea surface temperature, relative humidity, and air pressure, a linear interpolation method is used to smoothly transition the background atmospheric profile within the range from the sea surface to the preset transition height, in order to form the prior state vector. In this invention, the background atmospheric profile may be derived from numerical weather prediction reanalysis data, such as ERA5, NCEP, or other reanalysis products. The background atmospheric profile includes at least a temperature profile and relative humidity and pressure profiles. The resulting prior state vector... Participate in the inversion solution in step 3.

[0044] Step 3: Construct the cost function and perform inversion: Construct the observation vector based on the infrared radiation spectrum data, establish an inversion cost function that includes the observation residual term, background prior term and sea surface boundary constraint term, and solve for the atmospheric temperature profile and relative humidity profile based on the inversion cost function.

[0045] The cost function can be expressed as:

[0046] ;

[0047] in, Let be the atmospheric state vector to be solved. For the forward radiative transfer model, The observation error covariance matrix, The prior background covariance matrix, For the lowest layer parameters of the inversion profile These are measured parameters from the sea surface. The weights are the sea surface boundary constraints.

[0048] Let the observation error covariance matrix be the observation error covariance matrix. The uncertainty in infrared radiation spectroscopy observations is characterized by errors originating from instrument measurement noise and forward radiative transfer model errors. Preferably, the instrument measurement noise is determined based on the noise equivalent temperature difference or radiation calibration residual of the atmospheric emitted radiation interferometer; the forward radiative transfer model error is obtained statistically from the residuals between historical observed spectra and forward simulated spectra. The prior background covariance matrix; the prior background covariance matrix This is used to characterize the uncertainty of temperature and relative humidity parameters in the prior state vector. Preferably, the prior background covariance matrix is ​​obtained statistically from multi-year historical radiosonde data or reanalysis data of the target sea area, wherein the diagonal elements represent the variance of temperature or relative humidity parameters at each altitude layer, and the off-diagonal elements represent the covariance between different altitude layers.

[0049] Preferably, in constructing the observation vector Previously, a low-dimensional embedding space was constructed based on historical infrared radiation spectrum samples using the t-SNE method. Real-time acquired infrared radiation spectrum data was then mapped to this low-dimensional embedding space according to a pre-established mapping relationship to form observation vectors. .

[0050] Furthermore, to suppress non-physical oscillations in the inversion results of the low-altitude region near the sea surface, a low-altitude vertical smoothing constraint term can be added to the cost function:

[0051] ;

[0052] in, and These represent atmospheric parameters at adjacent altitudes within a preset altitude range from the sea surface. To smooth the constraint weights, The number of layers within the preset low-altitude range is defined. Among them, the sea surface boundary constraint term is used to anchor the true value of the lowest layer observation, and the low-altitude smoothing constraint term is used to constrain the temperature or relative humidity difference between adjacent height layers within the preset height range from the sea surface, so as to suppress non-physical oscillations in the low-altitude region inversion results, thereby maintaining the physical continuity of the inversion profile in the low-altitude layers.

[0053] At this point, the total cost function can be expressed as:

[0054] .

[0055] Step 4: Iteratively solve for the temperature profile and relative humidity profile.

[0056] The inversion cost function is minimized using an iterative optimization method to obtain atmospheric temperature profiles and relative humidity profiles. Iteration stops when the maximum number of iterations is reached.

[0057] In this embodiment, an optimal estimation method is preferably used for iterative solution. During the iteration process, the prior state vector obtained in step 2 is used. As an initial reference state, the state vector is updated incrementally to gradually reduce the inversion cost function. Each iteration calculates the simulation results of the forward radiative transfer model based on the current state vector, and corrects the state vector according to the changes in the cost function.

[0058] Step 5: Calculate the partial pressure of water vapor and the refractive index.

[0059] First, the saturated vapor pressure and partial vapor pressure at each altitude layer are calculated, and then the atmospheric refractive index profile is calculated. The pressure profile is obtained from the background pressure profile in the reanalysis data and corrected for the bottom layer using the measured sea surface pressure.

[0060] First, the absolute temperature Convert to Celsius :

[0061] ;

[0062] Then, the saturated vapor pressure is calculated using an empirical formula:

[0063] ;

[0064] in, This is the saturated vapor pressure, expressed in hPa.

[0065] Furthermore, the partial pressure of water vapor is calculated using the following formula:

[0066] ;

[0067] in, This is the partial pressure of water vapor, expressed in hPa. Relative humidity, in %; This is absolute temperature, measured in Kelvin (K).

[0068] Finally, the atmospheric refractive index is calculated using the following formula:

[0069] ;

[0070] in, This is atmospheric pressure, expressed in hPa. This is absolute temperature, measured in Kelvin (K). This is the partial pressure of water vapor, expressed in hPa. The refractive index is the atmospheric refractive index.

[0071] Through the above steps, the refractive index profile of the marine atmospheric boundary layer can be further obtained from the temperature profile and relative humidity profile obtained by inversion.

[0072] In this invention, by introducing a sea surface boundary constraint term during the optimal estimation inversion process and using sea surface observation data to correct the underlying prior state vector, the stability of the inversion of the temperature profile and relative humidity profile in the lower atmosphere near the sea surface can be effectively enhanced. Furthermore, by using a low-altitude vertical constraint term, non-physical oscillations in the near-surface inversion results can be reduced, thereby improving the accuracy of the refractive index profile in the key sea surface region.

[0073] In this embodiment, the sea surface air temperature, relative humidity, and air pressure measured by the shipborne micrometeorological observation module are first used to correct the background profile bottom layer parameters to form a priori state vector. Subsequently, the high-dimensional infrared radiation spectral data acquired by the atmospheric emitted radiation interferometer are subjected to t-SNE nonlinear dimensionality reduction processing to obtain the observation vector y required for inversion. Specifically, based on the historical infrared radiation spectral sample library, the t-SNE method is used to construct a low-dimensional embedding space corresponding to the high-dimensional spectral samples to maintain the local neighborhood structure between the high-dimensional spectral samples; for the real-time acquired infrared radiation spectral data, according to the pre-established mapping relationship or the reference sample neighborhood interpolation relationship, it is mapped to the low-dimensional embedding space in real time to form the observation vector. This is used for subsequent optimal estimation inversion. Next, an inversion cost function is constructed, comprising observation residuals, background priors, sea surface boundary constraints, and low-altitude gradient correction terms, and solved nonlinearly using an optimal estimation algorithm. During iteration, the sea surface boundary constraints are used to anchor the true values ​​of the lowest layer observations, and the low-altitude gradient correction term is used to suppress non-physical oscillations of parameters in adjacent layers within the range from the sea surface to a preset height. After iteration convergence, the temperature and relative humidity inversion results for each altitude layer are obtained, and the corresponding refractive index profiles are then calculated based on the pressure profiles.

[0074] In a shipborne maritime observation experiment, an atmospheric emitted radiation interferometer was installed on the observation platform on the upper deck of the ship. An attitude stabilization module compensated for the observation field of view, ensuring it was stably pointed towards the zenith. A shipborne micro-meteorological observation module was installed near the deck to simultaneously measure sea surface temperature, relative humidity, and air pressure. A cloud and sky identification and quality control module was used to determine the validity of the current infrared radiation spectrum observation data. If the observation was invalid, the current inversion was stopped, or the valid results from the previous moment were reused.

[0075] At a certain observation moment, the shipborne micro-meteorological observation module measured the following parameters near the sea surface:

[0076] Sea surface temperature ;

[0077] relative humidity ;

[0078] air pressure ;

[0079] The atmospheric emission radiation interferometer simultaneously acquires the downlink infrared radiation spectrum data at that moment.

[0080] In this embodiment, the inversion altitude range is set to 0~1000m, with an altitude layer interval of 100m, resulting in a total of 11 altitude layers. The temperature, relative humidity, and pressure profiles of the corresponding region and time in the ERA5 reanalysis data are used as the background atmospheric profile; an example of its lower-level portion is shown below:

[0081]

[0082] 1. Construct the prior state vector

[0083] The background atmospheric profile at the bottom layer is corrected using measured sea surface data. Specifically, the lowest layer background parameters are directly replaced with measured sea surface air temperature, relative humidity, and air pressure, i.e.:

[0084] ;

[0085] Within the range from the sea surface to the preset transition height, a linear interpolation method is used to achieve a smooth transition from measured sea surface parameters to the background atmospheric profile. Above the preset transition height, the background atmospheric profile remains unchanged, thus forming a priori state vector. .

[0086] In this embodiment, the prior state vector can be represented as:

[0087] ;

[0088] The prior state vector It serves as the background prior vector in the inversion cost function and as the initial state vector for subsequent iterations.

[0089] 2. Construct the observation vector and inversion cost function

[0090] Based on historical infrared radiation spectrum samples, a low-dimensional embedding space is constructed using the t-SNE method. Real-time acquired infrared radiation spectrum data is mapped to this low-dimensional embedding space according to a pre-established mapping relationship, forming observation vectors. Then construct the inversion cost function: ;

[0091] in, Let be the atmospheric state vector to be solved. For the forward radiative transfer model, The observation error covariance matrix, The prior background covariance matrix, For the lowest layer parameters of the inversion profile These are measured parameters from the sea surface. The weights are the constraints for the sea surface boundary.

[0092] In this embodiment, to suppress non-physical oscillations in the inversion results of the low-altitude region near the sea surface, a low-altitude vertical smoothing constraint term is also introduced:

[0093] ;

[0094] Therefore, the total cost function can be written as:

[0095] ;

[0096] Among them, the sea surface boundary constraint term is used to constrain the deviation between the lowest layer temperature or relative humidity parameter and the measured value of the sea surface; the low-altitude vertical smoothing constraint term is used to constrain the difference of temperature or relative humidity parameters of adjacent height layers within the range from the sea surface to the preset height, so as to suppress non-physical oscillations and maintain the continuity of the profile.

[0097] Subsequently, the high-dimensional infrared radiation spectrum data acquired by the atmospheric emitted radiation interferometer were subjected to nonlinear dimensionality reduction based on t-SNE to obtain the low-dimensional observation feature vectors required for inversion. This is used for subsequent optimal estimation and inversion. The t-SNE dimensionality reduction process is used to preserve the local neighborhood structure between high-dimensional infrared spectral samples, thereby extracting nonlinear low-dimensional features that are more sensitive to temperature and relative humidity inversion.

[0098] 3. Iteratively solve for the temperature profile and relative humidity profile.

[0099] The total cost function is minimized using the optimal estimation method, and the prior state vector is used as the solution. This serves as the initial reference state for the inversion. In each iteration, the output of the forward radiative transfer model is calculated based on the current state vector, and the state vector is updated according to the cost function. Iteration stops when the maximum number of iterations is reached.

[0100] In this embodiment, the atmospheric temperature profile and relative humidity profile at that moment are obtained after 5 iterations. An example of the results is as follows:

[0101]

[0102] As can be seen from the above results, by constraining the sea surface boundary and the low-altitude vertical smoothing constraint, the inversion results of temperature and relative humidity in the low-altitude layer near the sea surface remain continuous and smooth.

[0103] 4. Calculate the partial pressure and refractive index profile of water vapor.

[0104] Based on the temperature profile, relative humidity profile, and pressure profile obtained from the inversion, the saturated vapor pressure and partial pressure of water vapor at each altitude layer are first calculated. The pressure profile is obtained from the background pressure profile in the ERA5 reanalysis data and is then corrected for the bottom layer using the measured sea surface pressure.

[0105] First, the absolute temperature Convert to Celsius :

[0106] ;

[0107] Then, the saturated vapor pressure is calculated using an empirical formula:

[0108] ;

[0109] in, This is the saturated vapor pressure, expressed in hPa.

[0110] Furthermore, the partial pressure of water vapor is calculated using the following formula:

[0111] ;

[0112] in, This is the partial pressure of water vapor, expressed in hPa. Relative humidity, in %; This is absolute temperature, measured in Kelvin (K).

[0113] Finally, the atmospheric refractive index is calculated using the following formula:

[0114] ;

[0115] in, This is atmospheric pressure, expressed in hPa. This is absolute temperature, measured in Kelvin (K). This is the partial pressure of water vapor, expressed in hPa. The refractive index is the atmospheric refractive index.

[0116] The calculated refractive index profile is shown in the following example:

[0117] .

[0118] As the results above show, by constructing a priori state vectors using reanalysis background profiles and measured sea surface parameters, and by introducing sea surface boundary constraints and low-altitude vertical smoothing constraints during the inversion process, this invention can improve the stability of temperature and relative humidity inversion in the lower atmosphere near the sea surface, and further obtain a continuous, smooth, and physically consistent atmospheric refractive index profile. Compared with conventional inversion without the above constraints, this invention can reduce non-physical oscillations in the inversion profile in the low-altitude region near the sea surface, thereby improving the continuity and reliability of the refractive index profile in the key lower atmosphere.

Claims

1. A shipborne marine atmospheric boundary layer refractive index profile detection system, characterized in that, This includes an atmospheric emitted radiation interferometer, used to acquire downlink infrared radiation spectral data for sea and air; Shipborne micrometeorological observation module is used to measure temperature, relative humidity and air pressure at the sea surface in real time; The data processing module, connected to the atmospheric emitted radiation interferometer and the shipborne micrometeorological observation module, is used to construct observation vectors based on infrared radiation spectral data; construct background atmospheric profiles based on reanalysis data; and correct the underlying data using measured sea surface air temperature, relative humidity, and air pressure data to construct a priori state vectors. ,in, Let RH(z) be the absolute temperature and RH(z) be the relative humidity. An inversion cost function is established, which includes observation residuals, background priors, and sea surface boundary constraints. The atmospheric temperature profile and relative humidity profile are obtained by inversion. The atmospheric refractive index profile is calculated based on the temperature profile, relative humidity profile, and air pressure profile. The shipborne marine atmospheric boundary layer refractive index profile detection method based on the above system includes the following steps: Step 1: Obtain sea-air downlink infrared radiation spectrum data using an atmospheric emitted radiation interferometer, and simultaneously acquire sea surface air temperature, relative humidity, and air pressure data measured by the shipborne micro-meteorological observation module; Step 2: Construct a background atmospheric profile using the temperature, relative humidity, and air pressure profiles from the reanalysis data. Replace the lowest layer background parameters with measured sea surface temperature, relative humidity, and air pressure. Smooth the background profile using linear interpolation within the range from the sea surface to a preset transition height to form a priori state vector. The prior state vector serves as the background prior vector for the optimal estimation inversion in step 3, and also as the initial state vector for the iteration in the inversion solution. Step 3: Construct an observation vector y based on the infrared radiation spectral data, and establish an inversion cost function that includes observation residuals, background priors, and sea surface boundary constraints. in, Let be the atmospheric state vector to be solved. For the forward radiative transfer model, The observation error covariance matrix, The prior background covariance matrix, The parameters for the lowest layer of the inverted profile are... These are measured parameters from the sea surface. The weights are the constraints for the sea surface boundary. Step 4: Minimize the cost function using the optimal estimation method to obtain the atmospheric temperature profile and relative humidity profile; stop iterating when the maximum number of iterations is reached. Step 5: Calculate the water vapor partial pressure based on the temperature profile, relative humidity profile, and pressure profile obtained in Step 4, and further calculate the atmospheric refractive index profile; wherein, the pressure profile is obtained from the background pressure profile in the reanalysis data, and the bottom layer is corrected using the measured sea surface pressure as the bottom boundary condition.

2. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that: The system also includes an attitude stabilization module and a navigation information acquisition module. The navigation information acquisition module is used to acquire hull roll, pitch and yaw information, and the attitude stabilization module is used to perform attitude compensation based on the observation field of the atmospheric emission radiation interferometer.

3. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that: The system also includes a cloud and sky recognition and quality control module, which is used to determine the brightness temperature value of the characteristic channel of the window region in the infrared radiation spectrum and its time series fluctuation standard deviation. When the brightness temperature value exceeds the preset brightness temperature threshold, or the time series fluctuation standard deviation exceeds the preset fluctuation threshold, it determines whether the current observation data is valid, and stops the inversion or uses the valid result of the previous moment when the observation data is invalid.

4. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that, It also includes an anti-salt spray thermal curtain device, a thermal curtain air source and temperature control module. The anti-salt spray thermal curtain device is installed outside the optical observation window of the atmospheric emission radiation interferometer to prevent contamination and condensation on the optical observation window. The thermal curtain air source and temperature control module is used to provide heated compressed air for the anti-salt spray thermal curtain device.

5. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 4, characterized in that, The anti-salt spray thermal air curtain device is specifically a ring-shaped high-pressure jet nozzle located outside the optical observation window of the atmospheric emission radiation interferometer. The thermal air curtain air source and temperature control module heat the compressed air to 45℃-50℃ and then spray it out from the ring-shaped high-pressure jet nozzle.

6. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that: In step 3, before performing the optimal estimation inversion, a low-dimensional embedding space is constructed from the infrared radiation spectral data using the t-SNE method. The real-time acquired infrared radiation spectral data is then projected into the low-dimensional embedding space according to a pre-established mapping relationship to form an observation vector. .

7. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that: The sea surface boundary constraint term in step 3 is used to constrain the lowest layer temperature or relative humidity parameter in the state vector to be optimized, using the measured sea surface temperature or relative humidity parameter as the lowest layer boundary condition.

8. The shipborne marine atmospheric boundary layer refractive index profile detection system according to claim 1, characterized in that: Step 3 also includes a low-altitude vertical smoothing constraint term. ,in To smooth the constraint weights, and These represent atmospheric parameters at adjacent altitudes within the range from the sea surface to a preset altitude. The low-altitude vertical smoothing constraint term applies to the inter-layer differences in temperature or relative humidity parameters between adjacent altitudes within the range from the sea surface to the preset altitude, in order to suppress non-physical oscillations in the low-altitude region inversion results and maintain profile continuity. The specific cost function is expressed as follows: 。

Images