Night residual layer height range detection method and system

Abstract

The invention discloses a night residual layer height range detection method and system, and relates to the technical field of meteorological data analysis, and the method comprises the steps: obtaining a vertical profile of a particle extinction coefficient, the vertical profile being profile data of atmospheric particulate extinction coefficients distributed along with the height; preprocessing the vertical profile to suppress noise and retain gradient change characteristics; performing continuous wavelet covariance transformation on the preprocessed vertical profile according to a wavelet function to generate a two-dimensional wavelet coefficient matrix, and calculating to obtain a wavelet energy spectrum; extracting an energy spectrum ridge line representing the boundary of the night residual layer according to the wavelet energy spectrum diagram, and performing inversion according to the energy spectrum ridge line to obtain a lower boundary height and an upper boundary height; the height range of the night residual layer is determined according to the lower boundary height and the upper boundary height, and a visual detection result is output according to the height range, and the accuracy and reliability of the night residual layer height range detection process can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of meteorological data analysis, and in particular to a residual layer height range detection method and system. BACKGROUND

[0002] The residual layer refers to the atmospheric stratification above the stable nocturnal boundary layer, which is evolved from the previous day's convective boundary layer and still retains the uniform mixing characteristics. This layer stores most of the atmospheric pollutants that have been fully mixed in the convective boundary layer during the day, forming a "pollution storage" in the vertical direction. As the convective boundary layer develops the next day, the residual layer gradually disintegrates, and the stored pollutants are re-mixed and affect the air quality at the ground. Therefore, accurately determining the height range of the residual layer is the key prerequisite for quantifying its pollutant storage capacity and predicting the next day's air quality evolution. However, accurately determining the height range of the residual layer still faces difficulties: since the residual layer itself retains the uniform mixing characteristics, the vertical variation of particulate matter concentration within it is gentle, resulting in a gradual transition between the residual layer and the free atmosphere above and the stable boundary layer below, with weak gradient changes and unclear inflection points. This poses a fundamental difficulty for boundary identification.

[0003] In related technologies, gradient methods or threshold methods are mainly used to automatically identify the residual layer boundary. However, traditional gradient methods rely on finding obvious gradient extreme points, which are difficult to effectively capture the gentle gradient changes at the residual layer boundary, often leading to underestimation of the residual layer thickness and pollutant storage capacity. Both gradient thresholds and absolute thresholds of extinction coefficients are highly dependent on human pre-setting. These parameters lack physical universality and have poor adaptability in different regions, seasons, or pollution conditions, resulting in low comparability of the determination results. In addition, in the presence of low-altitude thin clouds, data noise, or complex multi-layer structures, gradient methods or threshold methods based on fixed rules are prone to misjudgment or omission, and the algorithm stability is insufficient.

[0004] Therefore, there is an urgent need to develop a residual layer height range detection method that can adaptively extract weak gradient mutation characteristics in vertical profiles, has strong anti-interference ability, and has objective and stable results. SUMMARY

[0005] The purpose of the present application is to provide a residual layer height range detection method and system that can improve the accuracy and reliability of the residual layer height range detection process.

[0006] To achieve the above-mentioned purpose, the present application provides the following solutions: In a first aspect, the application provides a method for detecting the height range of the residual layer at night, the method comprising: obtaining a vertical profile of the extinction coefficient of particles, wherein the vertical profile is profile data of the extinction coefficient of atmospheric particulate matter with respect to height; preprocessing the vertical profile to suppress noise and retain gradient change characteristics; selecting a wavelet function, performing continuous wavelet covariance transformation on the preprocessed vertical profile according to the wavelet function to generate a two-dimensional wavelet coefficient matrix, and calculating a wavelet energy spectrum diagram from the two-dimensional wavelet coefficient matrix; extracting an energy spectrum ridge line representing the boundary of the residual layer at night from the wavelet energy spectrum diagram, and inversely calculating a lower boundary height and an upper boundary height from the energy spectrum ridge line; determining the height range of the residual layer at night according to the lower boundary height and the upper boundary height, and outputting a visual detection result according to the height range.

[0007] Exemplarily, when preprocessing the vertical profile, a sliding average filtering algorithm is used to calculate the neighborhood average of the data points of the vertical profile with a specified window width, thereby suppressing random noise and impulse interference while retaining the gradient change characteristics of the vertical profile at the boundary of the residual layer at night. The gradient change characteristics are physical signals representing changes in the rate of change of the extinction coefficient of particles with respect to height.

[0008] Exemplarily, the calculation method of the specified window width in the sliding average filtering algorithm is as follows: wherein, is a physical smoothing scale, and the value range is 30 meters to 150 meters, represents the vertical resolution of the vertical profile, represents the specified window width.

[0009] Exemplarily, the selected wavelet function is a Mexican hat wavelet function, and the expression is as follows: wherein, is an independent variable, representing the position of the relative center within the wavelet function window, and the wavelet function has a sensitive response to the inflection point of the input signal, which is used to capture the gradient mutation characteristics representing the layer junction change in the vertical profile.

[0010] Exemplarily, the continuous wavelet covariance transformation, the generation of the two-dimensional wavelet coefficient matrix, and the calculation of the wavelet energy spectrum diagram specifically include: presetting the value range of the scale parameter according to the typical physical thickness of the residual layer at night, and generating a scale sequence containing a plurality of discrete scale values within the value range; calculating the wavelet coefficient based on each scale parameter in the scale sequence and each translation parameter covered by the vertical profile, and the calculation method is as follows: wherein, is a scale parameter, representing the thickness of the analyzed feature, with a value range of 300 meters to 1500 meters, is a translation parameter, representing the height position of the window center in the wavelet function, represents the pre-processed vertical profile, is an integral variable, representing the height, represents the wavelet function; and represent the lowest and highest effective height values of the profile data; by traversing all scale parameters and translation parameters, the corresponding wavelet coefficients are calculated and arranged to obtain a two-dimensional wavelet coefficient matrix; each element in the two-dimensional wavelet coefficient matrix is squared to obtain the wavelet energy spectrum map.

[0011] Exemplarily, according to the wavelet energy spectrum map, the energy spectrum ridge line representing the boundary of the residual layer at night is extracted, specifically including: for each scale parameter, finding the local maximum value point of the wavelet energy spectrum along the height direction as the candidate boundary point under the corresponding scale; according to the spatial proximity principle, connecting the candidate boundary points between different scales to form a continuously changing curve in the scale-height space as a candidate energy spectrum ridge line; from all candidate energy spectrum ridge lines, selecting two most continuous energy spectrum ridge lines extending along a specific height as the characteristic ridge lines representing the gradient mutation of the lower boundary and the upper boundary of the residual layer at night, respectively.

[0012] Exemplarily, the lower boundary height and the upper boundary height are inversely calculated according to the energy spectrum ridge line, specifically including: traversing each point on the energy spectrum ridge line, performing scale weighted average according to the position of the height coordinate in the energy spectrum ridge line to calculate the weighted average height of the energy spectrum ridge line; taking the weighted average heights of the two energy spectrum ridge lines as the lower boundary height and the upper boundary height of the residual layer at night, respectively, and the height range of the residual layer at night is the difference between the upper boundary height and the lower boundary height.

[0013] Exemplarily, the visualized detection result includes the wavelet energy spectrum map and the distribution synthesis map labeled with the lower boundary height and the upper boundary height.

[0014] Exemplarily, the vertical profile of the particle extinction coefficient is obtained by the following ways: real-time observation by laser radar, query from an atmospheric environment monitoring database, and download from a public scientific research data platform; the time resolution of the vertical profile of the particle extinction coefficient is not higher than 15 minutes, the vertical resolution is not higher than 30 meters, and the effective detection height is not less than 3 kilometers.

[0015] In a second aspect, the application provides a residual layer height range detection system at night, comprising: a data acquisition module for acquiring a vertical profile of particle extinction coefficient; a preprocessing module for preprocessing the vertical profile to suppress noise and retain gradient change characteristics; a wavelet transform module for performing continuous wavelet covariance transformation on the preprocessed vertical profile according to a selected wavelet function, generating a two-dimensional wavelet coefficient matrix, and calculating a wavelet energy spectrum diagram from the two-dimensional wavelet coefficient matrix; a ridge line extraction and inversion module for extracting an energy spectrum ridge line representing the boundary of the residual layer at night from the wavelet energy spectrum diagram, and inversely calculating a lower boundary height and an upper boundary height from the energy spectrum ridge line; and a result calculation and output module for confirming the height range of the residual layer at night according to the lower boundary height and the upper boundary height, and outputting a visual detection result according to the height range According to the specific embodiments provided by the application, the following technical effects are disclosed: The application provides a residual layer height range detection method and system at night. Through a data preprocessing step, observation noise and transient peak interference can be effectively suppressed. Wavelet covariance transformation is introduced. Through multi-scale analysis, weak and gentle gradient change information in a one-dimensional vertical profile can be converted and enhanced into obvious extreme value characteristics (ridge line) in a two-dimensional energy spectrum space. The defects of insensitivity to weak signals of the residual layer boundary are overcome, and the accuracy and reliability of boundary identification are improved. Core boundary determination depends on the global and continuous characteristics of the energy spectrum ridge line in the wavelet energy spectrum diagram, rather than simple comparison of local thresholds. The dependence on artificial preset parameters is reduced, and the objectivity, stability and repeatability of the results are significantly enhanced. The dynamic change characteristics of the residual layer under different meteorological conditions can be adaptively captured, and the spatiotemporal universality is better. The final output is a composite graph labeled with upper and lower boundary heights, which can intuitively show the position and range of the residual layer at night. It is not only convenient for automatic algorithm processing, but also provides clear and intuitive basis for manual verification and in-depth analysis. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, brief descriptions will be given to the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0017] Figure 1 A flowchart of a residual layer height range detection method at night in an embodiment of the application.

[0018] Figure 2 A schematic diagram of a vertical profile of particle extinction coefficient based on laser radar observation in an embodiment of the application.

[0019] Figure 3 This is a wavelet energy spectrum diagram from an embodiment of this application.

[0020] Figure 4 This is the final output distribution composite map in the embodiments of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] like Figure 1 As shown in the figure, this application provides a method for detecting the height range of residual layers at night, including the following steps: S110. Obtain the vertical profile of the extinction coefficient of particles, where the vertical profile is the profile data of the distribution of the extinction coefficient of atmospheric particulate matter with height.

[0024] S120. Preprocess the vertical profile to suppress noise and preserve gradient change characteristics.

[0025] S130. Select a wavelet function, perform continuous wavelet covariance transformation on the preprocessed vertical profile according to the wavelet function, generate a two-dimensional wavelet coefficient matrix, and calculate the wavelet energy spectrum through the two-dimensional wavelet coefficient matrix.

[0026] S140. Based on the wavelet energy spectrum, extract the energy spectrum ridge line that represents the boundary of the nighttime residual layer, and obtain the lower boundary height and upper boundary height based on the energy spectrum ridge line inversion.

[0027] S150. Determine the height range of the residual layer at night based on the lower and upper boundary heights, and output the visual detection results based on the height range.

[0028] The nighttime residual layer height range detection method provided in this application effectively suppresses observation noise and instantaneous spike interference through data preprocessing steps. It introduces wavelet covariance transform and, through multi-scale analysis, converts and enhances the weak and gentle gradient change information in the one-dimensional vertical profile into obvious extreme value features (ridges) in the two-dimensional energy spectrum space. This overcomes the deficiency of insensitivity to weak signals at the residual layer boundary, improving the accuracy and reliability of boundary identification. Core boundary determination relies on the global and continuous characteristics of the energy spectrum ridges in the wavelet energy spectrum map, rather than a simple comparison of local thresholds, reducing reliance on manually preset parameters and significantly enhancing the objectivity, stability, and repeatability of the results. It can adaptively capture the dynamic change characteristics of the residual layer under different meteorological conditions, exhibiting better spatiotemporal universality. The final output composite map, labeled with upper and lower boundary heights, intuitively displays the location and range of the nighttime residual layer, facilitating automated algorithm processing and providing clear and intuitive evidence for manual verification and in-depth analysis.

[0029] For example, the vertical profile represents the cross-sectional data of the atmospheric particulate extinction coefficient distribution with height. The vertical profile of the particulate extinction coefficient is obtained through the following methods: real-time observation using lidar, querying from an atmospheric environmental monitoring database, or downloading from a publicly available scientific research data platform. The temporal resolution of the vertical profile of the particulate extinction coefficient is no higher than 15 minutes, the vertical resolution is no higher than 30 meters, and the effective detection height is no less than 3 kilometers. Figure 2 The figure shown is a schematic diagram of the vertical profile of particle extinction coefficient based on lidar observations. The horizontal axis represents time, and the vertical axis represents altitude. The horizontal axis in the figure is local time (e.g., UTC+8 time zone), and the vertical axis is altitude (unit: meters). Figure 2 As can be seen, the effective detection altitude ranges from the ground to approximately 3 kilometers. From night to dawn, there exists a stratification with a relatively uniform extinction coefficient and a weak vertical gradient, which is the candidate region for the nighttime residual layer to be detected, demonstrating the spatiotemporal distribution characteristics of the original data.

[0030] To ensure the effectiveness and stability of wavelet covariance transform analysis during the preprocessing of the vertical profile, this embodiment employs a moving average filtering algorithm: neighborhood averaging is performed on the data points of the vertical profile with a specified window width. This suppresses random noise and impulse interference while preserving the gradient change characteristics of the vertical profile at the boundary of the nighttime residual layer. The gradient change characteristics are the physical signal of how the particle extinction coefficient changes with the rate of change of height. The calculation method for the specified window width in the moving average filtering algorithm is as follows: In the formula, The physical smoothness scale ranges from 30 meters to 150 meters. This indicates the vertical resolution of the vertical profile. represents the designated window width. In some embodiments, the vertical resolution of the vertical profile is 7.5 meters, The physical smoothness is 30 meters, and the corresponding designated window width is 4 data points.

[0031] The wavelet function selected in the embodiments of the present application is the Mexican hat wavelet function, the expression of which is as follows: wherein, is the independent variable, representing the position of the relative center within the wavelet function window, and the second derivative form of the wavelet function in the embodiments has a sensitive response to the inflection point of the input signal, and is used to capture the gradient mutation characteristics representing the layer junction change in the vertical profile.

[0032] After the vertical profile is preprocessed, a continuous wavelet covariance transform is performed by using the wavelet function to generate a two-dimensional wavelet coefficient matrix and calculate a wavelet energy spectrum. The specific process is as follows: According to the typical physical thickness of the night residual layer, the value range of the scale parameter is preset, and a scale sequence containing a plurality of discrete scale values is generated in the value range. Based on each scale parameter in the scale sequence and each translation parameter covered by the vertical profile, a wavelet coefficient is calculated, and the calculation method is as follows: wherein, is the scale parameter, representing the thickness of the analyzed feature, and the value range is 300 meters to 1500 meters, is the translation parameter, representing the height position of the center of the window in the wavelet function, represents the vertical profile after preprocessing, is the integral variable, representing the height, represents the wavelet function; and represent the lowest and highest effective height values of the profile data.

[0033] All scale parameters and translation parameters are traversed to calculate the corresponding wavelet coefficients and arrange them to obtain a two-dimensional wavelet coefficient matrix W(a, b). Each element in the two-dimensional wavelet coefficient matrix is squared to obtain the wavelet energy spectrum, that is: E(a, b) = [W(a, b)] 2 , wherein E(a, b) is the wavelet energy spectrum. As Figure 3The wavelet energy spectrum is shown in the embodiment of the present application. The horizontal coordinate is height b (unit: meter), and the vertical coordinate is scale (unit: meter). As can be seen from the figure, near the height of about 400 meters and 1400 meters, there are two bright yellow bands that are nearly vertically extended and have significantly enhanced energy, which are energy spectrum ridges. The scales (vertical coordinates) corresponding to the two energy spectrum ridges indicate that the gradient mutation characteristics have a multi-scale form. The energy value in the energy spectrum ridge region is significantly higher than the background noise, clearly separating and enhancing the layer boundary signal from the data.

[0034] After obtaining the wavelet energy spectrum, the energy spectrum ridges representing the residual layer boundary at night are extracted from the wavelet energy spectrum, and the specific process is as follows: For each scale parameter, the local maximum points of the wavelet energy spectrum are found along the height direction as the candidate boundary points under the corresponding scale. According to the spatial proximity principle, the candidate boundary points are connected between different scales to form a curve that continuously changes in the scale-height space, which is a candidate energy spectrum ridge. From all the candidate energy spectrum ridges, the two most continuous energy spectrum ridges extending along a specific height are selected as the characteristic ridges representing the gradient mutation of the lower boundary and the upper boundary of the residual layer at night, respectively.

[0035] Through the wavelet energy spectrum, local maxima are found at each scale, and the maximum points are tracked across scales to obtain candidate ridges. The candidate ridges are screened to obtain the strongest ridge. The strongest ridge represents the candidate ridge with the strongest energy and the best continuity. The positions of the two energy spectrum ridges on the height coordinate are inversely calculated to the actual altitude in the form of scale-weighted average, which are the lower boundary height (i.e., the height of the nocturnal boundary layer, NBLH) and the upper boundary height (i.e., the height of the residual layer, RLH) of the residual layer, respectively. The specific process is as follows: Each point on the energy spectrum ridge is traversed, and the scale-weighted average is calculated according to the position of the height coordinate on the energy spectrum ridge. Exemplarily, the height coordinate of each point is weighted and averaged with the energy value E(a i ,b i ) of the corresponding point on the wavelet energy spectrum as the weight, and the calculation method is as follows: wherein, is a translation parameter, indicating the height position of the window center in the wavelet function, i is the index of all points on the energy spectrum ridge, and are the scale value and height value of the i th point, respectively, The weighted average height is represented. The weighted average height of the two energy spectrum ridge lines is taken as the lower boundary height and the upper boundary height of the residual layer at night, respectively. The height range of the residual layer at night is the difference between the upper boundary height and the lower boundary height, that is, RLH - NBLH.

[0036] After the height range of the residual layer at night is calculated, the visual detection result is output based on the height range. The visual detection result includes a wavelet energy spectrum diagram and a distribution synthesis diagram marked with the lower boundary height and the upper boundary height. As shown in Figure 4 the distribution synthesis diagram finally output in the embodiment of the present application, the vertical profile of the particle extinction coefficient is usually taken as the background, and the above boundary lines are superimposed and marked on the corresponding height for intuitive display and manual auxiliary verification. The abscissa and the ordinate in the figure represent time and height, respectively. The identified NBLH and RLH are 437±87m and 1418±210m, respectively. The distribution synthesis diagram integrates the original profile shape and the transformed domain features, realizes the visualization verification of the detection process and result. In addition, the mixed layer height (MLH) is also marked in the figure. This height is obtained by applying the same wavelet covariance transform method to analyze the observation data during the day. The method provided by the present application is not only suitable for the residual layer at night, but also can be used to identify the atmospheric stratification boundary such as the mixed layer top during the day, which embodies the universality of the method.

[0037] The embodiment of the present application also provides a residual layer height range detection system at night. The system comprises a data acquisition module, a preprocessing module, a wavelet transform module, a ridge line extraction and inversion module, and a result calculation and output module.

[0038] The data acquisition module is used to acquire the vertical profile of the particle extinction coefficient. The preprocessing module is used to pre-process the vertical profile to suppress noise and retain gradient change features. The wavelet transform module is used to perform continuous wavelet covariance transform on the pre-processed vertical profile according to the selected wavelet function, generate a two-dimensional wavelet coefficient matrix, and calculate the wavelet energy spectrum diagram through the two-dimensional wavelet coefficient matrix. The ridge line extraction and inversion module is used to extract the energy spectrum ridge line representing the boundary of the residual layer at night according to the wavelet energy spectrum diagram, and to inversely calculate the lower boundary height and the upper boundary height according to the energy spectrum ridge line. The result calculation and output module is used to confirm the height range of the residual layer at night according to the lower boundary height and the upper boundary height, and to output the visual detection result according to the height range.

[0039] The method and system for detecting the height range of the residual layer at night provided in the embodiments of the present application can effectively suppress observation noise and instantaneous peak interference through a data preprocessing step, and a wavelet covariance transform is introduced. Through multi-scale analysis, weak and gentle gradient change information in one-dimensional vertical profiles can be converted and enhanced into obvious extreme features (ridge lines) in two-dimensional energy spectrum space, overcoming the defect of insensitivity to weak signals of the residual layer boundary, and improving the accuracy and reliability of boundary identification. The core boundary determination depends on the global and continuous features of the energy spectrum ridge lines in the wavelet energy spectrum diagram, rather than the simple comparison of local thresholds, reducing the dependence on artificial preset parameters and significantly enhancing the objectivity, stability and repeatability of the results. The method can adaptively capture the dynamic change characteristics of the residual layer under different meteorological conditions, and has better spatiotemporal universality. The final output of the synthesized image with upper and lower boundary heights can intuitively show the position and range of the residual layer at night, which is not only convenient for automatic algorithm processing, but also provides clear and intuitive basis for manual verification and in-depth analysis.

[0040] In an exemplary embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above method embodiments when executing the computer program.

[0041] In an exemplary embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program implements the steps in the above method embodiments when executed by a processor.

[0042] In an exemplary embodiment, a computer program product is provided, including a computer program, and the computer program implements the steps in the above method embodiments when executed by a processor.

[0043] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0044] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to a memory, a database or other medium used in the embodiments provided in the present application can include at least one of a non-volatile and a volatile memory. The non-volatile memory can include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, an optical storage, a high-density embedded non-volatile memory, a resistive random access memory (ReRAM), a magnetoresistive random access memory (MRAM), a ferroelectric random access memory (FRAM), a phase change memory (PCM), a graphene memory, etc. The volatile memory can include a random access memory (RAM) or an external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc.

[0045] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0046] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0047] The principles and implementation modes of the present application are described by applying specific examples in the present application. The above-mentioned embodiments are only used to help understand the method and its core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range can be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A method of detecting a range of night residual layer height, characterized by, The method comprises the following steps: obtaining a vertical profile of the extinction coefficient of atmospheric particulate matter, wherein the vertical profile is profile data of the distribution of the extinction coefficient of atmospheric particulate matter with height; preprocessing the vertical profile to suppress noise and retain gradient change characteristics; selecting a wavelet function, and performing continuous wavelet covariance transformation on the preprocessed vertical profile according to the wavelet function to generate a two-dimensional wavelet coefficient matrix, and calculating a wavelet energy spectrum map from the two-dimensional wavelet coefficient matrix; extracting an energy spectrum ridge line representing the boundary of the residual layer at night from the wavelet energy spectrum map, and inversely deriving a lower boundary height and an upper boundary height from the energy spectrum ridge line; confirming the height range of the residual layer at night according to the lower boundary height and the upper boundary height, and outputting a visual detection result according to the height range.

2. The night residual layer height range detection method according to claim 1, characterized by, When preprocessing the vertical profile, a sliding average filtering algorithm is used to calculate the neighborhood average of the data points of the vertical profile with a specified window width, thereby suppressing random noise and impulse interference while retaining the gradient change characteristics of the vertical profile at the boundary of the residual layer at night. The gradient change characteristics are physical signals representing changes in the rate of change of the extinction coefficient of atmospheric particulate matter with height.

3. The night residual layer height range detection method according to claim 2, characterized by, The calculation method of the specified window width in the sliding average filtering algorithm is as follows: wherein is a physical smoothing scale, having a value in the range 30 to 150 meters, denotes the vertical resolution of the vertical profile, denotes a specified window width.

4. The method of claim 1, wherein The selected wavelet function is a Mexican hat wavelet function, and the expression is as follows: wherein is an independent variable representing a position relative to the center of a wavelet function window having a sensitive response to a kink in an input signal for capturing a gradient jump feature characteristic of a layer junction change in the vertical profile.

5. The method of claim 4, wherein The continuous wavelet covariance transformation, the generation of the two-dimensional wavelet coefficient matrix, and the calculation of the wavelet energy spectrum map specifically include the following steps: preset the value range of the scale parameter according to the typical physical thickness of the residual layer at night, and generate a scale sequence containing multiple discrete scale values within the value range; based on each scale parameter in the scale sequence and each translation parameter covered by the vertical profile, calculate the wavelet coefficient, and the calculation method is as follows: wherein, is a scale parameter, representing the thickness of the analyzed feature, and taking values in the range 300-1500 meters, is a translation parameter, representing the height position of the window center in the wavelet function, represents the vertical profile after pre-processing, is an integration variable, representing the height, represents the wavelet function; and represent the lowest and highest significant height values of the profile data; iterate through all scale parameters and translation parameters to calculate the corresponding wavelet coefficient and arrange it to obtain a two-dimensional wavelet coefficient matrix; square each element in the two-dimensional wavelet coefficient matrix to obtain the wavelet energy spectrum map.

6. The method of detecting the range of residual layer height at night according to claim 5, characterized by, Extracting an energy spectrum ridge line representing the boundary of the residual layer at night from the wavelet energy spectrum map specifically includes the following steps: for each scale parameter, find the local maximum value point of the wavelet energy spectrum along the height direction as the candidate boundary point under the corresponding scale; connect the candidate boundary points according to the spatial proximity principle between different scales to form a curve that changes continuously in the scale-height space as a candidate energy spectrum ridge line; from all candidate energy spectrum ridge lines, select the two most continuous energy spectrum ridge lines extending along a certain height as the characteristic ridge lines representing the gradient mutation of the lower boundary and the upper boundary of the residual layer at night, respectively.

7. The method of detecting the range of residual layer height at night according to claim 6, characterized by, Inversely deriving a lower boundary height and an upper boundary height from the energy spectrum ridge line specifically includes the following steps: iterate through each point on the energy spectrum ridge line, and perform scale-weighted averaging according to the position of the height coordinate on the energy spectrum ridge line to calculate the weighted average height of the energy spectrum ridge line; The weighted average height of the two energy spectrum ridge lines is taken as the lower boundary height and the upper boundary height of the residual layer respectively, and the height range of the residual layer is the difference between the upper boundary height and the lower boundary height.

8. The night residual layer height range detection method according to any one of claims 1-7, characterized by, The visualization detection result includes a wavelet energy spectrum diagram and a distribution synthesis diagram marked with the lower boundary height and the upper boundary height.

9. The method of detecting the range of residual layer height at night according to any one of claims 1 to 7, characterized by, The vertical profile of the particle extinction coefficient is obtained by real-time observation by a laser radar, query from an atmospheric environment monitoring database, or download from a public scientific research data platform, and has a time resolution of no more than 15 minutes, a vertical resolution of no more than 30 meters, and an effective detection height of no less than 3 kilometers.

10. A nighttime residual layer height range detection system characterized by, The residual layer height range detection system comprises: a data acquisition module for acquiring the vertical profile of the particle extinction coefficient; a preprocessing module for preprocessing the vertical profile to suppress noise and retain gradient change characteristics; a wavelet transform module for performing continuous wavelet covariance transformation on the preprocessed vertical profile according to a selected wavelet function, generating a two-dimensional wavelet coefficient matrix, and calculating a wavelet energy spectrum diagram from the two-dimensional wavelet coefficient matrix; a ridge line extraction and inversion module for extracting energy spectrum ridge lines representing the boundaries of the residual layer from the wavelet energy spectrum diagram, and inversely calculating the lower boundary height and the upper boundary height from the energy spectrum ridge lines; a result calculation and output module for confirming the height range of the residual layer according to the lower boundary height and the upper boundary height, and outputting a visualization detection result according to the height range.

Images