A method for realizing gas gradient observation based on ground-based remote sensing detection technology

Abstract

The application discloses a kind of based on ground-based remote sensing detection technology and realizes the method for gas gradient observation, including building observation system;Gas concentration when each layer in observation system is on mirror at position 1 and position 2 and height difference between position 1 and position 2 is obtained to the gas concentration of the height where corresponding layer is on mirror;Based on gas concentration and real-time meteorological data, state space model is established, and model state is updated using Kalman filtering algorithm to correct the concentration change caused by the time difference between position 1 and position 2;By fusing historical concentration data and real-time monitored concentration data and corresponding real-time meteorological data, the parameters of state space model and Kalman filtering algorithm are updated.The advantage is: through the optimization of hardware system and algorithm, not only the high-precision gradient observation of gas concentration is realized, but also the intelligentization and self-adaptive ability of observation system are improved to adapt to complex environmental changes, and more reliable data support is provided for atmospheric environment research.

Description

Technical Field

[0001] This invention relates to the field of gas concentration monitoring technology, and in particular to a method for gas gradient observation based on ground-based remote sensing technology. Background Technology

[0002] Observing the vertical gradients of gas concentrations such as carbon dioxide, methane, and water vapor is crucial for studying the atmospheric environment and the carbon cycle in ecosystems. Traditional gas concentration measurement methods primarily involve installing inlets at different heights along a gas path, pumping gas into a chamber, and then measuring the concentration using an analyzer within the chamber. Due to the long pipeline layout, there is a significant time delay during measurement. Furthermore, switching between pipelines at different heights via gas path valves to ensure gas enters the measurement chamber for the corresponding channel makes it difficult to accurately obtain the corresponding concentration distribution, failing to meet the needs of in-depth research on the vertical variations of gases such as carbon dioxide and water vapor within the atmospheric boundary layer. In addition, the overall setup and maintenance of such multi-layered extraction systems is labor-intensive, requiring regular path calibration and filter replacement to ensure measurement accuracy, thus significantly increasing system operating costs.

[0003] Ground-based remote sensing technologies such as TDLAS (Tunable Diode Laser Absorption Spectroscopy) and DOAS (Differential Absorption Spectroscopy) are primarily based on the Lambert-Beer law. When a laser beam with frequency v passes through a gaseous medium of length L, the intensity attenuation satisfies:

[0004] I(v)=I0(v)exp[-σ(v)nL]

[0005] Where I0(v) is the incident light intensity; I(v) is the emitted light intensity; σ(v) is the absorption cross section of the gas molecules at frequency v; and n is the concentration of the gas molecules.

[0006] By measuring the attenuation of light intensity, information such as the concentration of gas molecules can be obtained. The key lies in using TDLAS / DOAS technology to detect the absorption of light within a corresponding wavelength range, thereby enabling the observation of the absorption characteristics of specific gases. However, existing gradient observation methods based on such technologies still have room for improvement in optical path stability, multi-parameter collaborative measurement, and data processing accuracy. Summary of the Invention

[0007] The purpose of this invention is to provide a method for observing gas gradients based on ground-based remote sensing technology, thereby solving the aforementioned problems existing in the prior art.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A method for observing gas gradients based on ground-based remote sensing technology includes the following steps:

[0010] S1. Observation System Setup: The observation system includes an intelligent gas analyzer, a dynamic optical path adjustment reflector mechanism, and a data processing terminal. The dynamic optical path adjustment reflector mechanism includes a reflector support, an upper reflector, and a lower reflector. On the vertically mounted reflector support, one lower reflector and n evenly spaced upper reflectors are installed from bottom to top along its height direction. The lower reflector is tilted at 45°, and the upper reflectors are horizontally positioned and capable of moving up and down and flipping up and down. The laser emitting end and laser receiving end of the intelligent gas analyzer face horizontally towards the lower reflector. The data processing terminal is connected to the intelligent gas analyzer.

[0011] S2. Basic Concentration Calculation: First, measure the gas concentration when the first-layer upper reflector is horizontally at position 1, starting from bottom to top. Then, move the first-layer upper reflector vertically downward to position 2 and measure the gas concentration when the first-layer upper reflector is horizontally at position 2. After that, rotate the first-layer upper reflector downward by 90° to position 3. Repeat the above process to measure the gas concentration of the remaining upper reflectors at positions 1 and 2. Use the gas concentration of each upper reflector at positions 1 and 2, as well as the height difference between positions 1 and 2, to obtain the gas concentration at the corresponding height of the upper reflector.

[0012] A state-space model is established based on gas concentration and real-time meteorological data. The Kalman filter algorithm is used to update the model state to correct the concentration change caused by the time difference between position 1 and position 2. The parameters of the state-space model and the Kalman filter algorithm are adaptively updated by fusing historical concentration data, real-time monitored concentration data and corresponding real-time meteorological data.

[0013] Preferably, the intelligent gas analyzer adopts TDLAS and DOAS technologies, and includes a beam emitter, a receiver, and a signal conversion unit, which can directly output the corresponding gas concentration value; the intelligent gas analyzer adopts a dual-optical-path parallel structure, the main optical path uses TDLAS technology to achieve high-precision single-point measurement, and the auxiliary optical path uses DOAS technology to cover a wide spectral range. The signals from different optical paths are integrated into the intelligent gas analyzer through a beam splitter and an optical fiber coupler.

[0014] The base of the intelligent gas analyzer is equipped with a 360-degree gimbal and a four-quadrant adjuster to achieve automatic signal focusing; the intelligent gas analyzer has a built-in edge computing unit to support real-time data processing and equipment status monitoring.

[0015] Preferably, step S2 specifically involves the following steps: For the gas concentration at a corresponding height for each layer, since the measurement times at positions 1 and 2 are close, it is approximately assumed that the gas concentration along the path remains constant for a short period. When calculating the gas concentration at each layer, the total gas concentration from the two measurements is calculated based on the optical path length and the measured gas concentration values. The difference between the two measurements is then used to calculate the gas concentration value at the corresponding spatial height based on the volume difference. The calculation formula is as follows:

[0016] ρ=(ρ1*V1-ρ2*V2) / (V1-V2)

[0017] Where ρ is the gas concentration value at the corresponding spatial height; ρ1 and ρ2 are the gas concentrations corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively; V1 and V2 are the total gas volume corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively.

[0018] Since the cross-sectional area of ​​the optical column measured by the TDLAS laser analyzer is constant, the volume ratio is equal to the optical path length ratio. Therefore, the formula for calculating the gas concentration value at the corresponding spatial height is:

[0019] ρ=(ρ1*L1-ρ2*L2) / (L1-L2)

[0020] Where L1 and L2 are the heights between the upper and lower reflectors of the corresponding layers when the upper reflector is in position 1 and position 2, respectively.

[0021] Preferably, step S3 specifically includes the following:

[0022] S31. Establish state transition equations and observation equations based on state variables and observation variables respectively to construct a state-space model;

[0023] S32. Use the Kalman filter algorithm to predict and update the state space model;

[0024] S33. Update the corresponding parameters in the state-space model and Kalman filter algorithm using historical concentration data and real-time meteorological data;

[0025] S34. Update the historical dataset using real-time monitored concentration data and real-time meteorological data, and continuously update the corresponding parameters in the state-space model and Kalman filter algorithm using the updated dataset.

[0026] Preferably, step S31 specifically includes the following:

[0027] S311. Select variables related to gas concentration as state variables, and based on the physical laws of gas diffusion and the analysis of historical data, establish the following state transition equation:

[0028] Xk =FX k-1 +W k-1

[0029] Among them, X k Let X be the state variable at time k; k-1 Let F be the state variable at time k-1; F is the state transition matrix, describing the change relationship of the state variables from time k-1 to time k; W k-1 This represents the process noise at time k-1;

[0030] S312. By observing the variables related to the state variables, the observation equations are established based on the state variables as follows.

[0031] Z k =HX k +V k

[0032] Among them, Z k Let V be the observed variable at time k; H is the observation matrix, used to map the state variables to the observation space; V k Let be the observation noise at time k.

[0033] Preferably, step S32 specifically includes the following:

[0034] S321. Based on the state transition equation, predict the state at the next moment and calculate the covariance matrix of the predicted state.

[0035]

[0036] P k∣k-1 =FP k-1∣k-1 F T +Q

[0037] in, It is the optimal state estimate at time k-1; P is the optimal state estimate at time k; k-1∣k-1 P is the state estimation covariance matrix at time k-1; k∣k-1 Let be the state estimation covariance matrix at time k; Q is the process noise covariance matrix.

[0038] S322. Calculate the Kalman gain based on the covariance matrix of the predicted state; update the state estimate and the state estimate covariance matrix based on the Kalman gain and the observed variables.

[0039] K k =P k∣k-1 H T HP k∣k-1 H T +R) -1

[0040]

[0041] P k∣k =(IK k H)P k∣k-1

[0042] Among them, K k R is the Kalman gain; R is the covariance matrix of the observation noise. Updated optimal state estimate at time k; P k∣k Estimate the covariance matrix for the updated time k; I is the identity matrix.

[0043] Preferably, step S33 specifically includes the following:

[0044] S331. Organize and analyze historical concentration data to extract valuable information, and convert it into a form that matches the state variables and observed variables;

[0045] S332. Incorporate real-time meteorological data into the observation matrix using an appropriate function form;

[0046] Based on historical concentration data and corresponding meteorological conditions, S333 adjusts and optimizes the state transition matrix F, observation matrix H, process noise covariance matrix Q, and observation noise covariance matrix R to determine the state space model parameters that best describe the relationship between gas concentration changes and meteorological data, thereby enabling the state space model to better adapt to actual conditions.

[0047] Preferably, step S34 specifically includes the following:

[0048] S341. During the operation of the observation system, continuously monitor the real-time data of gas concentration and changes in meteorological data. When a deviation is found between the actual data and the prediction results of the state-space model, analyze the reasons in a timely manner.

[0049] S342. Based on the real-time monitoring results, the parameters of the Kalman filter algorithm are dynamically adjusted adaptively.

[0050] S343. Over time, new concentration and meteorological data are continuously accumulated. New data are regularly incorporated into the historical dataset, and the state-space model is retrained and optimized.

[0051] Preferably, the observation system includes a displacement device and a rotation device mounted on a reflector support, wherein the upper reflector is moved up and down and rotated up and down by the displacement device and the rotation device, respectively.

[0052] The observation system also includes a miniature meteorological sensor mounted on the reflector bracket and an automatic cleaning device. During the measurement process, the miniature meteorological sensor monitors meteorological data in real time. When the meteorological data changes, the displacement device and rotation device automatically adjust the spacing and angle of the upper reflector to compensate for the optical path offset. If the reflector is detected to be contaminated during the measurement process, the automatic cleaning device is activated to blow the reflector to keep it clean.

[0053] The observation system includes a visible light alignment device to assist in the rapid and accurate calibration of the optical path by humans; and a chopper is introduced to enable flexible switching of the optical path to adapt to different detection scenarios and needs.

[0054] Preferably, the observation system integrates an intelligent diagnostic module, which can monitor light intensity fluctuations in real time, automatically identify abnormal conditions in the optical path, issue alarm information in a timely manner, and automatically execute corresponding maintenance measures according to the abnormal conditions.

[0055] The beneficial effects of this invention are: 1. High-precision measurement: The combination of a dual-optical-path parallel structure and a Kalman filter algorithm improves the accuracy and data integrity of various gas concentration gradient measurements. 2. Environmental adaptability: The dynamic optical path adjustment system and intelligent operation and maintenance technology effectively cope with environmental interference such as atmospheric turbulence and specular contamination, ensuring long-term stable operation of the system. 3. Intelligence and efficiency: Multi-source data fusion and adaptive algorithms achieve intelligent data processing optimization; modular design facilitates rapid deployment and flexible expansion, reducing operation and maintenance costs. 4. Multifunctional monitoring: It can simultaneously acquire multiple gas concentration gradients, providing richer data dimensions for atmospheric environment research. Attached Figure Description

[0056] Figure 1 This is a structural diagram of the observation system in an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0058] This embodiment provides a method for gas gradient observation based on ground-based remote sensing technology. Utilizing gas analyzers employing technologies such as TDLAS (Tunable Diode Laser Absorption Spectroscopy) and DOAS (Differential Absorption Spectroscopy), and through an innovative mirror system and intelligent algorithms, it achieves high-precision gradient observation of gases such as carbon dioxide and water vapor. This observation method is the first to apply TDLAS and DOAS remote sensing technologies to the observation of gas gradients such as CO2 and H2O, effectively solving problems such as low measurement accuracy and poor data synchronization compared to traditional methods. The innovative remote sensing approach avoids the need for multi-layered and complex pipeline layouts, significantly reducing gradient observation maintenance costs. The optimized hardware system and advanced algorithms significantly improve observation accuracy and regional monitoring capabilities, providing strong technical support for regional gas gradient observation. The method specifically includes the following parts:

[0059] I. Observation System Setup

[0060] like Figure 1 As shown, the observation system consists of an intelligent gas analyzer module, a dynamic optical path adjustment reflector system, and a data processing terminal.

[0061] 1.1 Intelligent Gas Analyzer Module: Utilizing TDLAS and DOAS technologies, this module includes a beam emitter, receiver, and signal conversion unit, directly outputting the concentration values ​​of the corresponding gases. The analyzer base is equipped with a 360-degree pan-tilt unit and a four-quadrant adjuster for automatic signal focusing. Furthermore, it features a built-in edge computing unit to support real-time data processing and equipment status monitoring.

[0062] 1.2 Dynamic Optical Path Adjustment Reflector System: Composed of a lower reflector and multiple reconfigurable upper reflectors. The lower reflector is tilted at 45° to perform 90° directional conversion of the optical signal. The multiple upper reflectors are installed at different heights, and the reflector support integrates a miniature meteorological sensor array (temperature, humidity, wind speed, and wind direction sensors) as well as displacement and rotation devices driven by servo motors. Each upper reflector can be moved to three positions (e.g., ...) via the displacement and rotation devices. Figure 1 The mirror can be moved between positions 1, 2, and 3. The vertical height difference between positions 1 and 2 is 20cm, while their horizontal orientation is the same. The upper reflector adopts a magnetic modular design with a built-in ID chip for easy and quick replacement and automatic system recognition and calibration.

[0063] Figure 1In the diagram, A represents the TDLAS and DOAS intelligent gas analyzer, which includes a beam emitter, receiver, and signal conversion unit, and can directly output the concentration value of the corresponding gas. The analyzer base is equipped with a 360-degree pan-tilt head and a four-quadrant adjuster to achieve automatic signal focusing. B is the lower reflector, which performs 90° directional conversion on the light signal; C consists of n upper reflectors at different heights, with a displacement and rotation device on the reflector support to switch the reflector positions from 1 to 3; D represents the laser emission and laser reflection receiving path.

[0064] The placement of the intelligent gas analyzer and the reflector requires manual adjustment to ensure that the light source can reach the reflector and be reflected back to the analyzer's receiver. Simultaneously, all reflectors employ a diffuse reflection design. This design effectively addresses issues such as abnormal light reflection and focusing caused by weather changes or dirty mirror surfaces. Diffuse reflection generates multiple light paths in different directions, ensuring that the reflected light inevitably reaches the receiver.

[0065] The upper reflectors at positions 1 and 2 in each layer differ only in vertical height by 20cm, while maintaining the same horizontal orientation. During the first layer gas concentration measurement, the measurement at position 1 is performed first. After this measurement, the upper reflector is moved down to position 2 via a displacement device, and a second concentration measurement is performed. After this measurement, the upper reflector rotates horizontally to position 3. At this point, the upper reflector is no longer in the light measurement path, allowing for the measurement of the gas concentration in the second layer. The measurement process for the second and subsequent layers is the same as for the first layer. During the measurement process, a miniature weather sensor monitors meteorological data in real time. When the meteorological data changes (e.g., wind speed exceeds the limit), the servo motor automatically adjusts the reflector spacing and angle to compensate for the light path offset. If reflector contamination is detected, the system triggers an automatic cleaning mechanism (miniature air pump purging). Furthermore, the system innovatively features a dual-optical-path parallel structure: the main optical path employs TDLAS technology for high-precision single-point measurement, while the auxiliary optical path utilizes DOAS technology to cover a wide spectral range. By using a beam splitter and fiber optic coupler, signals from different optical paths are integrated into the same analyzer, enabling the simultaneous acquisition of concentration gradients for multiple gases (such as CO2, CH4, and O3). The auxiliary optical path serves as a reference signal, primarily correcting for measurement path errors caused by other factors. In other words, DOAS acts as the reference signal, while TDLAS serves as the main measurement signal.

[0066] 1.3 Data processing terminal: Receives electrical signals transmitted by the analyzer, integrates multi-source data for concentration gradient calculation and analysis, and realizes remote data transmission and remote equipment control through the 5G network.

[0067] After the observation system is operating normally, the integrated intelligent diagnostic module monitors light intensity fluctuations in real time, automatically identifies abnormalities in the optical path, such as dirt, obstruction, or misalignment, and promptly issues alarm messages. The system then automatically executes corresponding maintenance measures. The gas analyzer, with the aid of a pan-tilt unit and a four-quadrant adjuster, achieves automatic focusing even with minute displacements based on the signal position shift of the reflected light. Simultaneously, it automatically triggers cleaning mechanisms such as a miniature dust removal device based on signal strength. Furthermore, the system includes a visible light alignment device to assist manual, rapid, and accurate optical path calibration; and a chopper is introduced to enable flexible switching of optical paths to adapt to different detection scenarios and requirements. In subsequent operation, apart from periodic manual system maintenance, all other maintenance tasks can be completed automatically, greatly improving the instrument's intelligence and operational stability.

[0068] II. Basic Concentration Calculation

[0069] First, measure the gas concentration when the first-layer upper reflector is horizontally at position 1, starting from the bottom up. Then, move the first-layer upper reflector vertically downward to position 2 and measure the gas concentration when the first-layer upper reflector is horizontally at position 2. After that, rotate the first-layer upper reflector downward by 90° to position 3. Repeat the above process to measure the gas concentration of the remaining upper reflectors at positions 1 and 2. Use the gas concentration of each upper reflector at positions 1 and 2, as well as the height difference between positions 1 and 2, to obtain the gas concentration at the corresponding height of the upper reflector.

[0070] For the gas concentration at each corresponding height of each layer, since the measurement times at positions 1 and 2 are close, it is approximately assumed that the gas concentration along the path remains constant for a short period of time. When calculating the gas concentration at each layer, the total gas concentration from the two measurements is calculated based on the optical path length and the magnitude of the measured gas concentration values. The difference between the two measurements is then used to calculate the gas concentration value at the corresponding spatial height based on the volume difference. The calculation formula is as follows:

[0071] ρ=(ρ1*V1-ρ2*V2) / (V1-V2)

[0072] Where ρ is the gas concentration value at the corresponding spatial height; ρ1 and ρ2 are the gas concentrations corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively; V1 and V2 are the total gas volume corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively.

[0073] Since the cross-sectional area of ​​the optical column measured by the TDLAS laser analyzer is constant, the volume ratio is equal to the optical path length ratio. Therefore, the formula for calculating the gas concentration value at the corresponding spatial height is:

[0074] ρ=(ρ1*L1-ρ2*L2) / (L1-L2)

[0075] Where L1 and L2 are the heights between the upper and lower reflectors of the corresponding layers when the upper reflector is in position 1 and position 2, respectively.

[0076] For each layer of the upper reflector at the installation height, L2-L1 = 20cm (height difference between position 1 and position 2). The gas concentration value in the corresponding height area is related to the installation position of the upper reflector. Since the installation height of the upper reflector is known, all parameters in the formula are known quantities, and the concentration value at the corresponding height can be directly calculated.

[0077] III. Concentration Correction

[0078] The Kalman filter algorithm, combined with meteorological data (wind speed, temperature, humidity, etc.), corrects for the impact of short-term gas diffusion on measurement results (i.e., corrects for the concentration change caused by the time difference between location 1 and location 2). By employing the Kalman filter algorithm and fusing historical concentration data with real-time meteorological data, dynamic compensation for concentration calculations is achieved, further improving data accuracy.

[0079] 3.1 Establishing a state-space model

[0080] (1) Define state variables and establish state transition equations: Select variables related to gas concentration as state variables, such as gas concentration values ​​at different altitudes. Assume the state variable is X. k =[C k,1 C k,2 ,…,C k,i ,…C k,n ] T , where C k,i Let represent the gas concentration at the i-th altitude at time k.

[0081] Based on the physical laws of gas diffusion and the analysis of historical data, a state transition equation is established.

[0082] X k =FX k-1 +W k-1

[0083] Among them, X k Let X be the state variable at time k; k-1 Let F be the state variable at time k-1; F is the state transition matrix, describing the change relationship of the state variables from time k-1 to time k; W k-1 The process noise at time k-1 is used to represent the uncertainty and external disturbances of the model, and is usually assumed to follow a Gaussian distribution with a mean of zero.

[0084] (2) Define the observation vector and establish the observation equation:

[0085] Information related to state variables is obtained through observation. Observations can include concentration values ​​measured by a gas analyzer and real-time meteorological data. Let the observation vector be Z. k =[Z k,1 Z k,2 ,…,Z k,j ,…Z k,m ] T Z k,j This represents the j-th observation at time k.

[0086] Establish the observation equation based on the state transition equation and the observation vector.

[0087] Z k =HX k +V k

[0088] Among them, Z k Let V be the observed variable at time k; H is the observation matrix, used to map the state variables to the observation space; V k The observation noise at time k also follows a Gaussian distribution with a mean of zero.

[0089] 3.2 Kalman Filtering

[0090] (1) Prediction

[0091] State prediction: Predict the state at the next moment based on the state transition equation;

[0092]

[0093] in, It is the optimal state estimate at time k-1; This is the optimal state estimate at time k. Covariance prediction: Calculate the covariance matrix of the predicted state;

[0094] P k∣k-1 =FP k-1∣k-1 F T +Q

[0095] Among them, P k-1∣k-1 P is the state estimation covariance matrix at time k-1; k∣k-1 Let be the state estimation covariance matrix at time k; Q is the process noise covariance matrix.

[0096] (2) Update

[0097] Calculate the Kalman gain:

[0098] K k =P k∣k-1 H T HP k∣k-1 H T +R)-1

[0099] Among them, K k Let R be the Kalman gain, and R be the covariance matrix of the observation noise.

[0100] Status Update:

[0101]

[0102] in, The updated optimal state estimate at time k.

[0103] Covariance update:

[0104] P k∣k =(IK k H)P k∣k-1

[0105] Among them, P k∣k Estimate the covariance matrix for the updated time k; I is the identity matrix.

[0106] 3.3 Integrating historical concentration data with real-time meteorological data

[0107] (1) Data preprocessing: Historical concentration data are organized and analyzed to extract valuable information, such as concentration change trends and seasonal patterns over different time periods. At the same time, meteorological data, such as wind speed, wind direction, temperature, and humidity, are acquired in real time and transformed into a form that matches the state variables and observed variables.

[0108] (2) Incorporating into the observation equation: Real-time meteorological data is incorporated as part of the observation equation. For example, the relationship between parameters such as wind speed and wind direction and gas concentration can be incorporated into the observation matrix H through an appropriate functional form. In this way, the observation vector Z k It includes not only the concentration values ​​measured by the gas analyzer, but also real-time meteorological data, enabling Kalman filtering to estimate state variables more accurately based on this information.

[0109] (3) Update model parameters using historical data: Adjust and optimize the state transition matrix F, observation matrix H, and noise covariance matrices Q and R based on historical concentration data and corresponding meteorological conditions. For example, historical data can be fitted using methods such as least squares to find the model parameters that best describe the relationship between gas concentration changes and meteorological data, thereby enabling the model to better adapt to actual conditions.

[0110] 3.4 Dynamic optimization and real-time adjustment

[0111] (1) Real-time monitoring and feedback: During system operation, the changes in real-time gas concentration data and meteorological data are continuously monitored. When a large deviation is found between the actual data and the model prediction results, the cause is analyzed in a timely manner. It may be that the model parameters need to be further adjusted, or that new interference factors have appeared.

[0112] (2) Adaptive parameter adjustment: Based on the real-time monitoring results, the parameters of the Kalman filter are dynamically adjusted adaptively. For example, when meteorological conditions change significantly, the observation matrix H and the noise covariance matrix R are adjusted accordingly to reflect the new observation relationships and uncertainties. At the same time, the state transition matrix F is fine-tuned according to the changing trend of gas concentration, so that the model can better track the dynamic changes in gas concentration.

[0113] (3) Regularly update historical data: Over time, new concentration and meteorological data are continuously accumulated. New data are regularly incorporated into the historical dataset, and the model is retrained and optimized to ensure that the model can reflect the latest patterns and trends in gas concentration changes in a timely manner, thereby continuously improving data accuracy.

[0114] By adopting the above-disclosed technical solution of this invention, the following beneficial effects are obtained:

[0115] This invention provides a method for gas gradient observation based on ground-based remote sensing technology. High-precision measurement: The combination of a dual-optical-path parallel structure and a Kalman filter algorithm improves the accuracy and data integrity of various gas concentration gradient measurements. Environmental adaptability: A dynamic optical path adjustment system and intelligent operation and maintenance technology effectively cope with environmental interference such as atmospheric turbulence and specular contamination, ensuring long-term stable operation of the system. Intelligence and efficiency: Multi-source data fusion and adaptive algorithms achieve intelligent data processing optimization; modular design facilitates rapid deployment and flexible expansion, reducing operation and maintenance costs. Multifunctional monitoring: Multiple gas concentration gradients can be acquired simultaneously, providing richer data dimensions for atmospheric environmental research.

[0116] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for observing gas gradients based on ground-based remote sensing technology, characterized in that: Includes the following steps, S1. Observation System Setup: The observation system includes an intelligent gas analyzer, a dynamic optical path adjustment reflector mechanism, and a data processing terminal. The dynamic optical path adjustment reflector mechanism includes a reflector support, an upper reflector, and a lower reflector. On the vertically mounted reflector support, one lower reflector and n evenly spaced upper reflectors are installed from bottom to top along its height direction. The lower reflector is tilted at 45°, and the upper reflectors are horizontally positioned and capable of moving up and down and flipping up and down. The laser emitting end and laser receiving end of the intelligent gas analyzer face horizontally towards the lower reflector. The data processing terminal is connected to the intelligent gas analyzer. S2. Basic Concentration Calculation: First, measure the gas concentration when the first-layer upper reflector is horizontally at position 1, starting from bottom to top. Then, move the first-layer upper reflector vertically downward to position 2 and measure the gas concentration when the first-layer upper reflector is horizontally at position 2. Next, rotate the first-layer upper reflector downward by 90° to position 3. Measure the gas concentration of the remaining upper reflectors at positions 1 and 2 using the same process as measuring the gas concentration of the first layer. Obtain the gas concentration at the corresponding height of the upper reflector using the gas concentration of each upper reflector at positions 1 and 2 and the height difference between positions 1 and 2. S3. Concentration Correction: A state-space model is established based on gas concentration and real-time meteorological data. The Kalman filter algorithm is used to update the model state to correct the concentration change caused by the time difference between position 1 and position 2. The parameters of the state-space model and the Kalman filter algorithm are adaptively updated by fusing historical concentration data, real-time monitored concentration data and corresponding real-time meteorological data. Step S3 specifically includes the following: S31. Establish state transition equations and observation equations based on state variables and observation variables respectively to construct a state-space model; S32. Use the Kalman filter algorithm to predict and update the state space model; S33. Update the corresponding parameters in the state-space model and Kalman filter algorithm using historical concentration data and real-time meteorological data; S34. Update the historical dataset using real-time monitored concentration data and real-time meteorological data, and continuously update the corresponding parameters in the state-space model and Kalman filter algorithm using the updated dataset.

2. The method for gas gradient observation based on ground-based remote sensing technology according to claim 1, characterized in that: The intelligent gas analyzer uses TDLAS and DOAS technologies and includes a beam emitter, receiver and signal conversion unit, which can directly output the corresponding gas concentration value. The intelligent gas analyzer adopts a dual-optical-path parallel structure. The main optical path uses TDLAS technology to achieve high-precision single-point measurement, and the auxiliary optical path uses DOAS technology to cover a wide spectral range. The signals from different optical paths are integrated into the intelligent gas analyzer through a beam splitter and fiber optic coupler. The base of the intelligent gas analyzer is equipped with a 360-degree gimbal and a four-quadrant adjuster to achieve automatic signal focusing; the intelligent gas analyzer has a built-in edge computing unit to support real-time data processing and equipment status monitoring.

3. The method for gas gradient observation based on ground-based remote sensing technology according to claim 2, characterized in that: Step S2 specifically involves the following steps: For the gas concentration at each corresponding height of each layer, since the measurement times at positions 1 and 2 are close, it is approximated that the gas concentration along the path remains constant for a short period. When calculating the gas concentration at each layer, the total gas concentration from the two measurements is calculated based on the optical path length and the measured gas concentration values. The difference between the two measurements is then used to calculate the gas concentration value at the corresponding spatial height based on the volume difference. The calculation formula is as follows: ρ=(ρ1*V1-ρ2*V2) / (V1-V2) Where ρ is the gas concentration value at the corresponding spatial height; ρ1 and ρ2 are the gas concentrations corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively; V1 and V2 are the total gas volume corresponding to the upper reflector of the corresponding layer when it is in position 1 and position 2, respectively. Since the cross-sectional area of ​​the optical column measured by the TDLAS laser analyzer is constant, the volume ratio is equal to the optical path length ratio. Therefore, the formula for calculating the gas concentration value at the corresponding spatial height is: ρ=(ρ1*L1-ρ2*L2) / (L1-L2) Where L1 and L2 are the heights between the upper and lower reflectors of the corresponding layers when the upper reflector is in position 1 and position 2, respectively.

4. The method for gas gradient observation based on ground-based remote sensing technology according to claim 3, characterized in that: Step S31 specifically includes the following: S311. Select variables related to gas concentration as state variables, and based on the physical laws of gas diffusion and the analysis of historical data, establish the following state transition equation: X k = FX k-1 + W k-1 Among them, X k Let X be the state variable at time k; k-1 Let F be the state variable at time k-1; F is the state transition matrix, describing the change relationship of the state variables from time k-1 to time k; W k-1 This represents the process noise at time k-1; S312. By observing the variables related to the state variables, the observation equations are established based on the state variables as follows. Z k =HX k +V k Among them, Z k Let V be the observed variable at time k; H is the observation matrix, used to map the state variables to the observation space; V k Let be the observation noise at time k.

5. The method for gas gradient observation based on ground-based remote sensing technology according to claim 4, characterized in that: Step S32 specifically includes the following: S321. Based on the state transition equation, predict the state at the next moment and calculate the covariance matrix of the predicted state. P k∣k-1 =FP k-1∣k-1 F T +Q in, It is the optimal state estimate at time k-1; P is the optimal state estimate at time k; k-1∣k-1 P is the state estimation covariance matrix at time k-1; k∣k-1 Let be the state estimation covariance matrix at time k; Q is the process noise covariance matrix. S322. Calculate the Kalman gain based on the covariance matrix of the predicted state; update the state estimate and the state estimate covariance matrix based on the Kalman gain and the observed variables. K k =P k∣k-1 H T (HP k∣k-1 H T +R) -1 P k∣k =(I-K k H)P k∣k-1 Among them, K k R is the Kalman gain; R is the covariance matrix of the observation noise. Updated optimal state estimate at time k; P k∣k Estimate the covariance matrix for the updated time k; I is the identity matrix.

6. The method for gas gradient observation based on ground-based remote sensing technology according to claim 5, characterized in that: Step S33 specifically includes the following: S331. Organize and analyze historical concentration data to extract valuable information, and convert it into a form that matches the state variables and observed variables; S332. Incorporate real-time meteorological data into the observation matrix using an appropriate function form; Based on historical concentration data and corresponding meteorological conditions, S333 adjusts and optimizes the state transition matrix F, observation matrix H, process noise covariance matrix Q, and observation noise covariance matrix R to determine the state space model parameters that best describe the relationship between gas concentration changes and meteorological data, thereby enabling the state space model to better adapt to actual conditions.

7. The method for gas gradient observation based on ground-based remote sensing technology according to claim 6, characterized in that: Step S34 specifically includes the following: S341. During the operation of the observation system, continuously monitor the real-time data of gas concentration and changes in meteorological data. When a deviation is found between the actual data and the prediction results of the state-space model, analyze the reasons in a timely manner. S342. Based on the real-time monitoring results, the parameters of the Kalman filter algorithm are dynamically adjusted adaptively. S343. Over time, new concentration and meteorological data are continuously accumulated. New data are regularly incorporated into the historical dataset, and the state-space model is retrained and optimized.

8. The method for gas gradient observation based on ground-based remote sensing technology according to any one of claims 1 to 7, characterized in that: The observation system includes a displacement device and a rotation device mounted on a reflector support. The upper reflector can move up and down and rotate up and down respectively through the displacement device and the rotation device. The observation system also includes a miniature meteorological sensor mounted on the reflector bracket and an automatic cleaning device. During the measurement process, the miniature meteorological sensor monitors meteorological data in real time. When the meteorological data changes, the displacement device and rotation device automatically adjust the spacing and angle of the upper reflector to compensate for the optical path offset. If the reflector is detected to be contaminated during the measurement process, the automatic cleaning device is activated to blow the reflector to keep it clean. The observation system includes a visible light alignment device to assist in the rapid and accurate calibration of the optical path by humans; and a chopper is introduced to enable flexible switching of the optical path to adapt to different detection scenarios and needs.

9. The method for gas gradient observation based on ground-based remote sensing technology according to claim 8, characterized in that: The observation system integrates an intelligent diagnostic module, which can monitor light intensity fluctuations in real time, automatically identify abnormalities in the optical path, issue alarm information in a timely manner, and automatically execute corresponding maintenance measures based on the abnormalities.

Images