Atmospheric detection laser radar calibration method based on slope iron tower

Abstract

The invention relates to an atmospheric detection laser radar calibration method based on a slope iron tower, and aims to solve the problem that an existing laser radar lacks a real reference section on an observation path. Iron tower observation units are arranged in a mountain slope area with a continuous gradient, and the height or distance of an iron tower is dynamically adjusted in combination with a terrain matching algorithm, so that an observation path of a laser radar is parallel to a slope and forms an intersection path with consistent space with a sensor sampling point on the tower. According to the invention, point-to-point correspondence between a radar detection path and actually measured reference data is realized through a multi-point meteorological observation iron tower with reasonable height distribution, so that spatial continuity calibration and calibration are carried out on radar inversion data; the method significantly improves the measurement precision and stability of the laser radar, is suitable for various meteorological conditions and terrain environments, and has wide practical application values and popularization prospects.

Description

Technical Field

[0001] This invention relates to the field of remote sensing technology, and in particular to a calibration method for atmospheric sounding lidar based on a sloping iron tower. Background Technology

[0002] Ground-based lidar is a high-precision observation device based on optical remote sensing principles, widely used in atmospheric boundary layer structure detection, vertical temperature and humidity profile analysis, and inversion studies of trace gases such as CO2 and H2O. In atmospheric science, carbon cycle research, air quality monitoring, and numerical weather prediction, lidar has become an important ground-based observation tool due to its advantages of non-contact, long-distance, and continuous vertical profile observation. However, the accuracy of atmospheric parameters acquired by lidar is not only limited by the optical performance of the lidar itself, but also affected by multiple factors such as aerosol distribution, humidity gradient, changes in optical thickness, optical path offset, and sensor aging, causing systematic biases and even distortions in the observation results. Especially under complex meteorological conditions such as strong boundary layer disturbances, cloud penetration, and multi-layer temperature inversions, lidar inversion errors increase significantly, making it difficult to guarantee the comparability and reliability of long-term continuous monitoring data.

[0003] Existing lidar calibration methods mainly include self-calibration methods based on inversion data and comparative calibration methods based on external observation data. Self-calibration methods based on inversion data use theoretical data derived from internal inversion channels or physical models as a reference to estimate the calibration coefficients of the system. However, the accuracy of these methods is highly dependent on the physical model and parameter settings used, and systematic biases can easily arise when atmospheric conditions change or model simplification is inaccurate. Comparative calibration methods based on external observation data rely on field-collected external observation data as a reference standard, such as measured data on temperature, humidity, and gas concentration obtained from platforms like radiosondes, UAVs, or weather towers. However, these methods have several practical limitations: radiosondes are limited by the number of launches and the coverage in time and space, making continuous calibration difficult; UAVs, while flexible, are limited by flight altitude, endurance, and stability, making it impossible to continuously acquire high-resolution profile data; and fixed weather towers typically have limited coverage height, making it difficult to match the entire observation path of the lidar. These limitations lead to certain errors in the calibration accuracy of existing methods. Summary of the Invention

[0004] The purpose of this invention is to provide a calibration method for atmospheric sounding lidar based on a sloping iron tower, thereby solving the aforementioned problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: a method for calibrating an atmospheric sounding lidar based on a sloping iron tower, comprising the following steps:

[0006] S100. In mountainous slope areas with continuous slopes, multiple standard observation points are set up along the main slope direction, and iron tower observation units are constructed. The height of the iron tower observation units is optimized according to the characteristics of atmospheric vertical stratification to better reflect the characteristics of atmospheric vertical stratification.

[0007] S200: The terrain matching algorithm is used to dynamically calculate the optimal elevation angle of the radar and adjust the height of the tower, so that the observation path of the lidar is basically parallel to the slope and forms a spatially consistent intersection path with the sampling points of the sensors on the tower. The terrain matching algorithm is based on the projection distance of the tower slope and the terrain undulation data, and achieves point fitting optimization by minimizing the vertical deviation between the observation path and the line connecting the sampling points. The input of this algorithm includes digital elevation model (DEM) data and a custom horizontal position of the tower, and the output is the optimal observation elevation angle θ of the radar and the height of each tower.

[0008] S300: Install a data logger, an air temperature and humidity sensor, and a standard material module corresponding to the calibrated radar detection parameters on each standard point observation tower;

[0009] S400: A central station is built at the foot of the mountain. The radar to be calibrated is installed in the central station. The radar elevation angle is adjusted so that its transmission path and the sensor measurement point on the tower form a spatially nearly consistent intersection path. A high-precision GPS clock module is used to synchronize the time of all nodes. The central station has time drift monitoring and correction functions to ensure that the collected data has strict timestamp consistency.

[0010] S500 and the central station trigger the standard material modules of each standard point to release the target material at a set concentration within a fixed time window through commands. The sensors of each tower unit perform real-time measurements synchronously. After the measurement values ​​stabilize, the central station starts the sampling action of the radar to be calibrated. All sampling data are accompanied by a unified timestamp and are transmitted back to the central station in real time.

[0011] The S600 system uses measured data transmitted from standard points along the lidar path to form a spatially continuous and temporally synchronized profile chain. The system performs point-by-point comparative analysis to perform calibration, constructs an error linear correction model, fits the model correction coefficients a and b, and constructs a model in the form L_ calibrated =a·L_ measured The +b correction model is applied to the entire radar path profile data, and the corrected model is applied to the entire radar inversion profile to complete the full path calibration of the radar. After calibration, the system further checks the error between the corrected profile and the measured data. If the residual of the observation point is found to exceed the preset threshold, a new round of calibration task is automatically triggered to improve the overall calibration accuracy.

[0012] Preferably, when deploying tower units, a higher density of observation points is deployed in the near-surface layer, while a sparser distribution of points is deployed in the upper boundary layer region above 200m.

[0013] Preferably, the deployment of the tower units is planned based on a terrain matching algorithm. This algorithm comprehensively considers the installation elevation angle of the lidar and the effective sampling height difference between each tower to determine the optimal elevation angle of the lidar and the tower height, so that the measurement positions of the sensors on multiple towers form an oblique profile that is approximately parallel to the radar observation path. During the deployment phase, the algorithm adjusts the height and spacing of each tower based on the elevation changes of the mountain slope to avoid observation obstruction or sampling offset caused by terrain undulations.

[0014] After processing, the DEM data of the slope can be represented as a continuous function H(x) along the main slope direction, where x is the horizontal distance and H(x) is the ground elevation at that horizontal distance. The slope profile function H(x) is linearly fitted using the least squares method to calculate the optimal radar elevation angle, ensuring the radar path is parallel to the slope trend. The radar elevation angle algorithm is as follows:

[0015]

[0016] Where θ is the optimal elevation angle of the lidar, x i H(x) represents discrete sampling points along the x-axis. i ) represents the slope height of the corresponding sampling point.

[0017] The central station is located at the foot of the mountain, at a distance x0 from the zero point of the slope. Based on the elevation angle θ and x0, the radar's observation path Z can be obtained. Tower observation points can be customized along the radar observation path according to certain principles to obtain the horizontal distance x along the x-axis for each tower observation point. k and the corresponding observation height Z(x); finally, based on x k H(x) position k ) and Z(x k The height of each tower is obtained by calculation; the tower height algorithm is as follows:

[0018] h k =Z(x) k )-H(x k )

[0019] Among them, h k Let be the height of the k-th tower (k = 1, 2, ..., k).

[0020] Preferably, the standard substance module includes two configuration options:

[0021] (1) The standard version includes: a water vapor generator and its concentration measuring device, a CO2 release device and its concentration measuring device, a methane release device and its concentration measuring device, a wind speed regulating fan and a wind speed measuring device, and an ozone generator and its concentration measuring device.

[0022] (2) The simplified version is suitable for calibration sites with relatively stable meteorological conditions. The standard material module is replaced by one or more of the following analyzers: CO2 / H2O gas analyzer, methane analyzer, three-dimensional ultrasonic anemometer, and ozone analyzer.

[0023] Preferably, before the calibration task is started, the central station first triggers the release of standard material modules of each tower and the sampling device of the corresponding sensor. The sampling data is transmitted back to the central station in real time for concentration change trend monitoring and stability judgment. When the parameters measured by the sensor reach a stable state at the tower node, the central station controls the lidar to start sampling to ensure the stability of the gas field environment at the radar observation path. If the stability condition is not met, the start is delayed and the judgment is repeated until the trigger condition is met.

[0024] Preferably, after calibration, the system performs error analysis on the lidar inversion profile and the measured profile data of each standard material module. When the root mean square error (RMSE) of an observation point exceeds a preset threshold, the system automatically triggers a new round of calibration process and adjusts the release concentration parameters and sampling time window of the standard material accordingly to improve the accuracy and consistency of subsequent calibration results.

[0025] Preferably, the deployment of the tower observation units in step S100 further includes:

[0026] In sloping terrain, a lidar echo intensity pre-scanning mechanism is introduced to obtain the aerosol backscattering coefficient profile along the slope through lidar pre-scanning, and to dynamically identify signal blind spots caused by terrain obstruction or atmospheric turbulence.

[0027] Based on the blind spot identification results, the tower spacing and height are recalculated, and the deployment scheme is optimized by minimizing the blind spot coverage rate, with the blind spot boundary as a constraint.

[0029] The beneficial effects of this invention are:

[0030] The atmospheric sounding lidar calibration method based on a sloping iron tower proposed in this invention has the following significant advantages compared with the prior art:

[0031] 1. Significantly improves lidar calibration accuracy

[0032] By deploying multiple meteorological observation towers at reasonable heights in mountainous slope areas, and dynamically adjusting the tower height or spacing using a terrain matching algorithm, the lidar's observation path is made parallel to the slope and forms a spatially consistent intersection path with the sensor sampling points on the towers. This unique layout and method provides a true reference profile that highly matches the lidar's observation path, thereby calibrating and standardizing the spatial continuity of the lidar inversion data. Compared to existing calibration methods, this invention significantly improves lidar measurement accuracy, reduces systematic errors, and ensures the comparability and reliability of long-term continuous monitoring data.

[0033] 2. Extend the coverage of standard data

[0034] This invention constructs a spatially continuous and temporally synchronized profile chain by deploying multiple standard observation points on a slope and installing standard material modules and sensors at each point. This design significantly extends the coverage of standard data along the entire detection path, providing more comprehensive and continuous reference data compared to existing technologies (such as radiosondes, UAVs, or fixed weather towers), effectively addressing the shortcomings of existing calibration methods in terms of spatiotemporal resolution.

[0035] 3. Enhance data time consistency and reliability

[0036] By installing a high-precision GPS clock module on each observation tower, this invention enables unified time synchronization across all nodes, ensuring data time consistency. This time synchronization mechanism effectively reduces time deviations during data acquisition, improving the accuracy and reliability of calibration results. Attached Figure Description

[0037] Figure 1 This is a flowchart of an atmospheric sounding lidar calibration method based on a sloping iron tower according to the present invention;

[0038] Figure 2 This is a schematic diagram of the calibration method of the present invention;

[0039] Figure 3 This is a schematic diagram of the terrain matching algorithm of the present invention. Detailed Implementation

[0040] 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.

[0041] Reference Figure 1 , Figure 2 and Figure 3 The method for calibrating an atmospheric sounding lidar based on a sloping iron tower, as shown, includes the following steps:

[0042] S100. In mountainous slope areas with continuous slopes, multiple standard observation points are set up along the main slope direction, and iron tower observation units are constructed. The height of the iron tower observation units is optimized according to the vertical stratification characteristics of the atmosphere to better reflect the vertical stratification characteristics of the atmosphere.

[0043] This step is specifically described as follows: A calibration site is selected in a mountainous slope area with continuous gradients. Multiple standard observation points are established along the main slope direction, and a tower observation unit is constructed at each observation point. The height of the towers is gradient-optimized based on the characteristics of atmospheric vertical stratification to better reflect these characteristics. The specific layout is as follows:

[0044] Near-surface layer (0-100m): Due to the drastic changes in the vertical atmospheric gradient, a higher density of observation points are deployed, with the towers spaced approximately 20-30 meters apart, to ensure that rapid changes in atmospheric parameters can be captured.

[0045] Upper boundary layer (above 200m): The atmospheric vertical gradient changes relatively little, so observation points can be set up sparsely, with the tower spacing being about 50-80 meters, to reduce energy consumption and equipment redundancy.

[0046] S200: The terrain matching algorithm is applied to dynamically calculate the radar elevation angle and adjust the tower height so that the observation path of the lidar is basically parallel to the slope and forms a spatially consistent intersection path with the sampling points of the sensors on the tower. The terrain matching algorithm is based on the slope projection distance of the tower and the terrain undulation data. It achieves point fitting optimization by minimizing the vertical deviation between the observation path and the line connecting the sampling points. The input of the algorithm includes digital elevation model (DEM) data and the custom horizontal position of the tower. The output is the optimal radar observation elevation angle θ and the height of each tower.

[0047] In this step, specifically described as follows: The terrain matching algorithm is used to calculate the optimal radar elevation angle and optimize the tower height so that the radar transmission path and the sensor sampling points form a spatially consistent intersection path, and remain parallel to the terrain slope overall. Its goal is to minimize the vertical deviation between the laser path and the sensor connection line;

[0048] During the deployment phase, the terrain matching algorithm adjusts the radar elevation angle and the height of each tower based on the elevation changes of the mountain slope to avoid observation obstruction or sampling offset caused by terrain undulation.

[0049] After processing, the DEM data of the slope can be represented as a continuous function H(x) along the main slope direction, where x is the horizontal distance and H(x) is the ground elevation at that horizontal distance. The slope profile function H(x) is linearly fitted using the least squares method to calculate the optimal radar elevation angle, ensuring the radar path is parallel to the slope trend. The radar elevation angle algorithm is as follows:

[0050]

[0051] Where θ is the optimal elevation angle of the lidar, x i H(x) represents discrete sampling points along the x-axis. i ) represents the slope height of the corresponding sampling point.

[0052] The central station is located at the foot of the mountain, at a distance x0 from the zero point of the slope. Based on the elevation angle θ and x0, the radar observation path Z can be obtained. Along the radar observation path, tower observation points can be customized according to principle 2, and the horizontal distance x along the x-axis of each tower observation point can be obtained. k and the corresponding observation height Z(x); finally, based on x k H(x) position k ) and Z(x k The height of each tower is obtained by calculating its height. The algorithm for calculating the tower height is as follows:

[0053] h k =Z(x) k )-H(x k )

[0054] Among them, h k Let be the height of the k-th tower (k = 1, 2, ..., k). This formula ensures that the lidar's observation path is parallel to the slope by calculating the optimal elevation angle of the radar and the height of the tower, and that the lidar beam and the sensor sampling points on the tower form a spatially consistent intersection path.

[0055] Through the detailed description above, the principle and objective of the terrain matching algorithm have been clearly explained. This algorithm is one of the core innovations of this invention. By dynamically adjusting the tower layout, it ensures the parallelism and spatial consistency between the lidar observation path and the slope, significantly improving the positioning accuracy and reliability of the lidar beacon.

[0056] S300: Install a data logger, an air temperature and humidity sensor, and a standard material module corresponding to the calibrated radar detection parameters on each standard observation tower.

[0057] In this step, it is specifically described as follows:

[0058] 1. Data Logger Configuration: Each tower unit is equipped with a data logger as the core control and data acquisition device. It features the following functions: multi-channel synchronous sampling, high-precision timestamp annotation, remote control and parameter setting capabilities, real-time operational status feedback, and self-diagnostic functions.

[0059] The data logger maintains communication with the central station via wireless or wired means to ensure the real-time performance and stability of command execution and data transmission.

[0060] 2. Environmental Sensor Deployment: Air temperature and humidity sensors are used to collect local environmental parameters in real time, providing background reference and correction basis for LiDAR data. The sensors are installed on the tower structure in an unobstructed location, avoiding direct sunlight and strong winds, to ensure the representativeness and validity of the measurement data.

[0061] 3. Standard Material Module Composition: Depending on different radar calibration requirements and site environmental conditions, the standard material module can be selected from either the standard version or the simplified version.

[0062] The standard version is suitable for high-precision calibration scenarios, and may include: a water vapor generator and its concentration measurement device, a CO2 release device and its concentration measurement device, a methane release device and its concentration measurement device, a wind speed control fan and its wind speed measurement device, and an ozone generator and its concentration measurement device. All modules support remote control and flow regulation, enabling active gas release and synchronous measurement to construct a high-fidelity reference profile.

[0063] The simplified solution is suitable for calibration scenarios with stable meteorological conditions. It primarily removes the active release device, retaining only the passive observation modules, including: a CO2 / H2O gas analyzer, a methane analyzer, a 3D ultrasonic anemometer, and an ozone analyzer. The simplified solution has lower deployment costs and is easier to operate, making it suitable for calibration needs under low-disturbance background conditions.

[0064] 4. System integration and scalability:

[0065] All equipment supports modular deployment, and its operation status is monitored uniformly by the central station. Components can be flexibly added or removed according to the specific site and radar type to achieve high adaptability and long-term stable operation.

[0066] S400: A central station is built at the foot of the mountain. The radar to be calibrated is installed in the central station. The radar elevation angle is adjusted so that its transmission path and the sensor measurement point on the tower form a spatially nearly consistent intersection path. A high-precision GPS clock module is used to synchronize the time of all nodes. The central station has time drift monitoring and correction functions to ensure that the collected data has strict timestamp consistency.

[0067] In this step, the specific description is as follows: The central station should be located in an area directly opposite the slope to ensure a good viewing angle and communication channel. After the lidar is installed, the elevation angle is initially set based on the terrain matching algorithm, and then fine-tuned based on the on-site echo conditions to ensure that its observation path passes through the measurement points of each sensor. This elevation angle matching process is based on geometric inverse calculations of the radar height, the tower sensor height, and the horizontal distance.

[0068] To achieve system-wide time synchronization, high-precision GPS timing modules are configured at the central station and each tower. Each node receives a unified time signal to align the time of its data recording devices, ensuring that the collected data has consistent timestamps and supporting subsequent joint inversion and fitting.

[0069] After the system was built, functional testing, link testing and data consistency verification were conducted to confirm that the radar transmission path, the deployment path of each node and the time synchronization mechanism were working properly, ensuring the spatial matching and time consistency of the calibration process.

[0070] The S500 and central station trigger the standard substance modules at each standard point to release the target substance at a set concentration within a fixed time window. The sensors of each tower unit perform real-time measurements synchronously. After the measured values ​​stabilize, the central station initiates the sampling action of the radar to be calibrated. All sampling data are accompanied by a unified timestamp and are transmitted back to the central station in real time.

[0071] In this step, the central station pre-configures parameters such as the type of substance to be released, set concentration, release duration, and stability threshold for each standard point. It then sends task commands to each tower unit via wireless communication to control the standard substance module to begin release and simultaneously activate its corresponding sensor sampling device. Real-time sampling data from each node is timestamped and transmitted back to the central station. The central station dynamically determines whether the gas field has reached a stable state based on the concentration change curve. If the concentration change rate measured at all key observation points is within the preset stable range, the central station immediately initiates the lidar sampling task, ensuring the radar detection path covers the stable gas profile area. If some nodes do not meet the stability criteria, the radar activation is delayed, and the system continues to assess the situation at fixed intervals until the conditions are met.

[0072] During the formal sampling phase, all devices are time-aligned using a unified GPS timing signal. Collected data is uniformly timestamped and transmitted back to the central station in real time via wired or wireless links. The central station performs unified data storage, preliminary formatting, and integrity verification. The system possesses capabilities for task instruction feedback monitoring, data synchronization error monitoring, and communication quality assessment. It can issue real-time alarms for abnormal states during task execution, ensuring the entire calibration process is conducted with dual guarantees of spatial path consistency and temporal sampling consistency.

[0073] The S600 system uses measured data transmitted from standard points along the lidar path to form a spatially continuous and temporally synchronized profile chain. The system performs point-by-point comparative analysis to perform calibration, constructs an error linear correction model, fits the model correction coefficients a and b, and constructs a model in the form L_ calibrated =a·L_ measuredThe +b correction model is applied to the radar path profile data to complete the radar's full path calibration. After calibration, the system further checks the error between the corrected profile and the measured data. If the residual of the observation point is found to exceed the preset threshold, a new round of calibration task is automatically triggered to improve the overall calibration accuracy.

[0074] In this step, specifically described as follows: The system pairs and compares radar inversion data with measured data from each observation point through timestamp matching and spatial path alignment. A linear error correction model is then constructed using the least squares fitting method.

[0075] L calibrated =a×L measured +b

[0076] Among them, L calibrated L represents the standard measurement value for each tower node. measured This represents the radar measurement value to be calibrated, and 'a' and 'b' are correction coefficients obtained by fitting the target to minimize the error. This model can be applied in batches to all observation point data along the radar detection path, completing the accuracy correction for the entire path. This process can significantly improve the profile consistency and data reliability of lidar during long-term operation.

[0077] After comparing the initial lidar profile data with the measured data at each standard observation point, the system automatically initiates an accuracy verification program to perform error analysis on all sampling points. The system extracts the measured concentration values ​​of the radar inversion profile and the corresponding spatiotemporal points of each tower's standard material module, and calculates the root mean square error (RMSE) for each point using the following formula:

[0078]

[0079] Among them, L calibrated,i For the data of the i-th point after radar calibration, L standard,i is the measured concentration at the corresponding point in the standard substance module, and n is the total number of points.

[0080] After calculating the RMSE for all points, if the error at any single point exceeds a preset threshold (e.g., 10% relative error), the system considers the calibration results of this round to lack sufficient consistency. The system then automatically triggers the next calibration process, including the following steps:

[0081] (1) Adjust the release concentration of standard substance: Based on the error amplitude and spatial distribution characteristics of the previous round, automatically increase or decrease the release concentration to enhance the radar response contrast.

[0082] (2) Optimize sampling time window: Based on the previous air mass stabilization time estimation results, automatically extend or shorten the sampling duration to allow the air mass to expand more fully and the sensor sampling phase to be concentrated in the stable section.

[0083] Preferably, the tower unit deployment density is set according to the atmospheric vertical stratification characteristics, with high-density sampling points set in the near-ground layer and low-density sampling points set in the area above 200m.

[0084] Preferably, the deployment of the tower units is planned based on a terrain matching algorithm. This algorithm comprehensively considers the installation elevation angle of the lidar and the effective sampling height difference between each tower to determine the optimal elevation angle of the lidar and the tower height, so that the measurement positions of the sensors on multiple towers form an oblique profile that is approximately parallel to the radar observation path. During the deployment phase, the algorithm adjusts the height and spacing of each tower based on the elevation changes of the mountain slope to avoid observation obstruction or sampling offset caused by terrain undulations.

[0085] After processing, the DEM data of the slope can be represented as a continuous function H(x) along the main slope direction, where x is the horizontal distance and H(x) is the ground elevation at that horizontal distance. The slope profile function H(x) is linearly fitted using the least squares method to calculate the optimal radar elevation angle, ensuring the radar path is parallel to the slope trend. The radar elevation angle algorithm is as follows:

[0086]

[0087] Where θ is the optimal elevation angle of the lidar, x i H(x) represents discrete sampling points along the x-axis. i ) represents the slope height of the corresponding sampling point;

[0088] The central station is located at the foot of the mountain, at a distance x0 from the zero point of the slope. Based on the elevation angle θ and x0, the radar's observation path Z can be obtained. Tower observation points can be customized along the radar observation path according to certain principles to obtain the horizontal distance x along the x-axis for each tower observation point. k and the corresponding observation height Z(x); finally, based on x k H(x) position k ) and Z(x k The height of each tower is obtained by calculation; the tower height algorithm is as follows:

[0089] h k =Z(x) k )-H(x k )

[0090] Among them, h k Let be the height of the k-th tower (k = 1, 2, ..., k).

[0091] Preferably, the standard substance module includes two configuration options:

[0092] (1) The standard version includes: a water vapor generator and its concentration measuring device, a CO2 release device and its concentration measuring device, a methane release device and its concentration measuring device, a wind speed regulating fan and a wind speed measuring device, and an ozone generator and its concentration measuring device.

[0093] (2) The simplified version is suitable for calibration sites with relatively stable meteorological conditions. The standard material module is replaced by one or more of the following analyzers: CO2 / H2O gas analyzer, methane analyzer, three-dimensional ultrasonic anemometer, and ozone analyzer.

[0094] Preferably, before the calibration task is started, the central station first triggers the release of standard material modules of each tower and the sampling device of the corresponding sensor. The sampling data is transmitted back to the central station in real time for concentration change trend monitoring and stability judgment. When the parameters measured by the sensor reach a stable state at the tower node, the central station controls the lidar to start sampling to ensure the stability of the gas field environment at the radar observation path. If the stability condition is not met, the start is delayed and the judgment is repeated until the trigger condition is met.

[0095] Preferably, after calibration, the system performs error analysis on the lidar inversion profile and the measured profile data of each standard material module. When the root mean square error (RMSE) of an observation point exceeds a preset threshold, the system automatically triggers a new round of calibration process and adjusts the release concentration parameters and sampling time window of the standard material accordingly to improve the accuracy and consistency of subsequent calibration results.

[0096] Preferably, the deployment of the tower observation units in step S100 further includes:

[0097] In sloping terrain, a lidar echo intensity pre-scanning mechanism is introduced to obtain the aerosol backscattering coefficient profile along the slope through lidar pre-scanning, and to dynamically identify signal blind spots caused by terrain obstruction or atmospheric turbulence.

[0098] Based on the blind spot identification results, the tower spacing and height are recalculated, and the deployment scheme is optimized by minimizing the blind spot coverage rate, with the blind spot boundary as a constraint.

[0100] By adopting the above-disclosed technical solution of this invention, the following beneficial effects are obtained:

[0101] 1. Significantly improves lidar calibration accuracy

[0102] By deploying multiple meteorological observation towers at reasonable heights in mountainous slope areas, and dynamically adjusting the tower height or spacing using a terrain matching algorithm, the lidar's observation path is made parallel to the slope and forms a spatially consistent intersection path with the sensor sampling points on the towers. This unique layout and method provides a true reference profile that highly matches the lidar's observation path, thereby calibrating and standardizing the spatial continuity of the lidar inversion data. Compared to existing calibration methods, this invention significantly improves lidar measurement accuracy, reduces systematic errors, and ensures the comparability and reliability of long-term continuous monitoring data.

[0103] 2. Extend the coverage of standard data

[0104] This invention constructs a spatially continuous and temporally synchronized profile chain by deploying multiple standard observation points on a slope and installing standard material modules and sensors at each point. This design significantly extends the coverage of standard data along the entire detection path, providing more comprehensive and continuous reference data compared to existing technologies (such as radiosondes, UAVs, or fixed weather towers), effectively addressing the shortcomings of existing calibration methods in terms of spatiotemporal resolution.

[0105] 3. Enhance data time consistency and reliability

[0106] By installing a high-precision GPS clock module on each observation tower, this invention enables unified time synchronization across all nodes, ensuring data time consistency. This time synchronization mechanism effectively reduces time deviations during data acquisition, improving the accuracy and reliability of calibration results.

[0107] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles 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 calibrating an atmospheric sounding lidar based on a sloping iron tower, characterized in that, Includes the following steps: S100. In mountainous slope areas with continuous slopes, multiple standard observation points are set up along the main slope direction, and iron tower observation units are constructed. The height of the iron tower observation units is optimized according to the vertical stratification characteristics of the atmosphere to better reflect the vertical stratification characteristics of the atmosphere. S200: The optimal elevation angle of the radar is dynamically calculated using a terrain matching algorithm, and the height of the tower is adjusted so that the observation path of the lidar is basically parallel to the slope and forms a spatially consistent intersection path with the sampling points of the sensors on the tower. The terrain matching algorithm is based on the slope projection distance of the tower and the terrain undulation data, and achieves point fitting optimization by minimizing the vertical deviation between the observation path and the line connecting the sampling points. The input of this algorithm includes digital elevation model data and a custom horizontal position of the tower, and the output is the optimal observation elevation angle θ of the radar and the height of each tower. S300: Install a data logger, an air temperature and humidity sensor, and a standard material module corresponding to the calibrated radar detection parameters on each standard point observation tower; S400: A central station is built at the foot of the mountain. The radar to be calibrated is installed in the central station. The radar elevation angle is adjusted so that its transmission path and the sensor measurement point on the tower form a spatially nearly consistent intersection path. A high-precision GPS clock module is used to synchronize the time of all nodes. The central station has time drift monitoring and correction functions to ensure that the collected data has strict timestamp consistency. S500 and the central station trigger the standard material modules of each standard point to release the target material at a set concentration within a fixed time window through commands. The sensors of each tower unit perform real-time measurements synchronously. After the measurement values ​​stabilize, the central station starts the sampling action of the radar to be calibrated. All sampling data are accompanied by a unified timestamp and are transmitted back to the central station in real time. The S600 system uses measured data transmitted from standard points along the lidar path to form a spatially continuous and temporally synchronized profile chain. The system performs point-by-point comparative analysis to perform calibration, constructs an error linear correction model, fits the model correction coefficients a and b, and constructs a model in the form L_ calibrated =a·L_ measured The +b correction model is applied to the entire radar path profile data, and the correction model is applied to the entire radar inversion profile to complete the full path calibration of the radar. After calibration, the system further checks the error between the corrected profile and the measured data. If the residual of the observation point is found to exceed the preset threshold, a new round of calibration task is automatically triggered to improve the overall calibration accuracy.

2. The atmospheric sounding lidar calibration method based on an inclined iron tower according to claim 1, characterized in that, The tower unit deployment density is set according to the vertical stratification characteristics of the atmosphere, with high-density sampling points set in the near-ground layer and low-density sampling points set in the area above 200m.

3. The atmospheric sounding lidar calibration method based on a sloping iron tower according to claim 2, characterized in that, The deployment of the tower units is planned based on a terrain matching algorithm. This algorithm comprehensively considers the installation elevation angle of the lidar and the effective sampling height difference between each tower to determine the optimal elevation angle of the lidar and the tower height, so that the measurement positions of the sensors on multiple towers form an oblique profile that is approximately parallel to the radar observation path. During the deployment phase, the algorithm adjusts the height and spacing of each tower based on the elevation changes of the mountain slope to avoid observation obstruction or sampling offset caused by terrain undulations. After processing, the DEM data of the slope can be represented as a continuous function H(x) along the main slope direction, where x is the horizontal distance and H(x) is the ground elevation at that horizontal distance. The slope profile function H(x) is linearly fitted using the least squares method to calculate the optimal radar elevation angle, ensuring the radar path is parallel to the slope trend. The radar elevation angle algorithm is as follows: Where θ is the optimal elevation angle of the lidar, x i H(x) represents discrete sampling points along the x-axis. i ) represents the slope height of the corresponding sampling point; The central station is located at the foot of the mountain, at a distance x0 from the zero point of the slope. Based on the elevation angle θ and x0, the radar's observation path Z can be obtained. Tower observation points can be customized along the radar observation path according to certain principles to obtain the horizontal distance x along the x-axis for each tower observation point. k and the corresponding observation height Z(x); finally, based on x k H(x) position k ) and Z(x k The height of each tower is obtained by calculation; the algorithm for calculating the tower height is as follows: h k =Z(x k )-H(x k ) Among them, h k Let be the height of the k-th tower (k = 1, 2, ..., k).

4. The atmospheric sounding lidar calibration method based on a sloping iron tower according to claim 3, characterized in that, The standard substance module includes two configuration options: (1) The standard version includes: a water vapor generator and its concentration measuring device, a CO2 release device and its concentration measuring device, a methane release device and its concentration measuring device, a wind speed regulating fan and a wind speed measuring device, and an ozone generator and its concentration measuring device. (2) The simplified version is suitable for calibration sites with relatively stable meteorological conditions. The standard material module is replaced by one or more of the following analyzers: CO2 / H2O gas analyzer, methane analyzer, three-dimensional ultrasonic anemometer, and ozone analyzer.

5. The atmospheric sounding lidar calibration method based on a sloping iron tower according to claim 4, characterized in that, Before the calibration task is started, the central station prioritizes the release of standard material modules of each tower and the sampling devices of the corresponding sensors. The sampling data is transmitted back to the central station in real time for concentration change trend monitoring and stability judgment. When the parameters measured by the sensors reach a stable state at the tower nodes, the central station controls the lidar to start sampling to ensure the stability of the atmospheric environment at the radar observation path. If the stability condition is not met, the start is delayed and the judgment is repeated until the triggering condition is met.

6. The atmospheric sounding lidar calibration method based on a sloping iron tower according to claim 5, characterized in that, After calibration, the system performs error analysis on the lidar inversion profile and the measured profile data of each standard material module. When the root mean square error of an observation point exceeds the preset threshold, the system automatically triggers a new round of calibration process and adjusts the release concentration parameters and sampling time window of the standard material accordingly to improve the accuracy and consistency of subsequent calibration results.

7. The atmospheric sounding lidar calibration method based on a sloping iron tower according to claim 6, characterized in that, The deployment of the tower observation unit in step S100 further includes: In sloping terrain, a lidar echo intensity pre-scanning mechanism is introduced to obtain the aerosol backscattering coefficient profile along the slope through lidar pre-scanning, and to dynamically identify signal blind spots caused by terrain obstruction or atmospheric turbulence. Based on the blind spot identification results, the tower spacing and height are recalculated, and the deployment scheme is optimized by minimizing the blind spot coverage rate, with the blind spot boundary as a constraint.

Images