GNSS-based landslide geological hazard monitoring and analysis method using a virtual reference station

Abstract

A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station, comprising: Step 1, which involves determining the continuously operating reference station, i.e., the CORS network, and selecting monitoring points; Step 2, which involves analyzing the multi-system virtual reference station network; Step 3, which involves verifying the quality of virtual observation values, using data from physical reference stations to verify and evaluate the quality of VRS observation values; and Step 4, which involves data processing and result analysis, analyzing monitoring data from different periods, statistically calculating the accuracy of the calculations, and statistically evaluating the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions. [Effects] To reduce the cost of GNSS landslide monitoring, virtual base station data is generated in the monitoring area using the continuously operating satellite positioning base station CORS, and landslide monitoring is performed in place of the physical base station. At the same time, VRS is not limited by geological conditions or distance, and meets the requirements of wide-area, highly versatile, and adaptable to complex environments in landslide monitoring.

Description

[Technical Field]

[0001] The present invention relates to the field of remote sensing disaster identification technology, and more specifically, to a GNSS-based landslide geological disaster monitoring and analysis method based on a virtual reference station. [Background technology]

[0002] Landslide geological disaster monitoring primarily involves both public monitoring and prevention, as well as professional monitoring. Professional monitoring mainly utilizes GNSS technology and sensor technologies such as crack depth gauges, tension gauges, and rain gauges. GNSS technology offers advantages such as all-weather capability, high accuracy, automation, and no line-of-sight requirements. Its relative accuracy in baseline analysis can reach 10⁻⁹, making it an important technological tool in landslide monitoring and early warning.

[0003] Current GNSS landslide monitoring primarily employs the "1+N" mode, which consists of one base station and N monitoring points. Landslide displacement is obtained by analyzing the baseline between the base station in a geologically stable area and the monitoring points within the landslide body. The "1+N" mode requires the installation of at least one base station near each monitoring area, which presents problems such as high construction costs, difficulty in selecting installation locations, instability of the baseline, and waste of resources. [Overview of the project] [Problems that the invention aims to solve]

[0004] The technical problem that this invention aims to solve is to provide a GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station, thereby solving problems in current GNSS landslide monitoring, such as high construction costs, difficulty in selecting installation sites, instability of references, and waste of resources. [Means for solving the problem]

[0005] To solve the above technical problems, the technical solutions employed by the present invention are as follows.

[0006] A GNSS-based landslide geological disaster monitoring and analysis method based on a virtual reference station, Step 1 involves determining the CORS network and selecting monitoring points. This includes determining the coverage status of the CORS network in the construction area, selecting landslide feature monitoring points that require monitoring within the CORS coverage area, and installing monitoring equipment. CORS is defined as continuously operating reference stations. This is an analysis of a multi-system virtual reference station network, and in performing landslide monitoring based on a virtual reference station (VRS), the key to Step 2 is to generate valid VRS observation values ​​near the monitoring point, perform short-baseline analysis on the VRS observation values ​​and monitoring point observation values, and obtain the virtual reference station observation values ​​for the monitoring point. Step 3 is a quality verification of virtual observation values, in which the quality of virtual observation values ​​obtained from VRS is verified and evaluated using physical reference station data. Step 4 involves data processing and result analysis, which includes analyzing monitoring data for different periods, statistically calculating the accuracy of the calculations, and statistically evaluating the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions.

[0007] In Step 1 above, the determination of the CORS network and selection of monitoring points include determining the CORS network coverage area within the construction area and selecting landslide feature monitoring points.

[0008] In Step 2 above, the analysis method for virtual reference station observation values ​​includes the following three parts: First, the baselines within the CORS network are analyzed and the dual phase difference integer ambiguity is fixed; second, the ionospheric delay, tropospheric delay, and overall error of each baseline are calculated using the Beidou observation values, dual phase difference integer ambiguity, and precise reference station position coordinates for each baseline; and third, differential correction information at the virtual reference station position is determined by interpolation calculation based on the virtual reference station coordinates, and a real-time virtual reference service is provided to the geological disaster monitoring terminal.

[0009] The analysis process of integer ambiguity in the analysis of the virtual reference station observations is as follows.

[0010] 1) The carrier phase observation value and the P-code pseudorange form the Melbourne-Wuebbena combination, that is, the M-W combination, and the wide-lane integer ambiguity is calculated and obtained. The specific calculation formula is as follows.

Equation

[0011] In the formula, the superscript letters p and q are the identifiers of the observed satellite p and the reference satellite q, the subscript letter WL is the GNSS wide-lane combination observation value, the subscript letters i and j are the identifiers of the reference stations i and j, λ is the wavelength of the carrier wave, f1 and f2 are the frequencies of the first frequency point L1 and the second frequency point L2 of the GNSS observation values respectively, Δ∇ is the double differential operator, N is the integer ambiguity, P is the pseudorange observation value, and φ is the carrier phase observation value.

[0012] 2) An ionosphere-free combination is performed on the carrier phase observation value and the pseudorange observation value, and analyzed to obtain the real solution of the ionosphere-free ambiguity. The observation equation is as follows.

Equation

[0013] In the formula, ρ is the distance from the measurement point to the satellite, Trop is the troposphere delay error, ε is the carrier phase observation value noise, γ is the pseudorange observation value noise. Here, since GLONASS adopts a signal structure of frequency division multiple access, the GLONASS double differential ambiguity needs to be converted into one double differential integer ambiguity and one single differential integer ambiguity in the double differential observation equation in distance units. The observation equation is as follows.

Number

[0014] In the formula, λ glo , φ glo , P glo , N glo are respectively the wavelength of the GLONASS observation value, the phase observation value, the pseudo-range observation value and the integer ambiguity, p is the observed satellite, and q is the reference satellite.

[0015] 3) Using the wide-lane integer ambiguity analyzed in step 1), the real solution of the ionosphere-free ambiguity analyzed in step 2), and the corresponding variance-covariance matrix, obtain the real solution of the narrow-lane ambiguity and the corresponding variance-covariance matrix, and fix the narrow-lane integer phase ambiguity using the least squares ambiguity decorrelation adjustment method (LAMBDA method). Furthermore, calculate to obtain the GNSS satellite L1 and L2 integer phase ambiguities.

[0016] The calculation processes of the above ionospheric delay, tropospheric delay and total error are as follows.

[0017] In the error calculation of the multi-system virtual reference station, the error is classified into ionospheric delay, tropospheric delay, and total error including second-order ionospheric delay, orbit error and multipath error, etc., and calculated respectively. After fixing the integer ambiguity of each baseline, based on the GNSS observation values of each baseline, the integer ambiguity of the double phase difference, the precise reference station position coordinates, etc., the above three types of errors are calculated and obtained. The specific calculation formulas are as follows.

Number

[0018] In the formula, Δ∇Iono is the double phase difference ionospheric delay, φ1 is the carrier wave observation at the L1 frequency point, φ2 is the carrier wave observation at the L2 frequency point, MF is the tropospheric mapping function, ZWD is the tropospheric zenith moist delay, ZHD is the tropospheric zenith hydrostatic delay, Other is the overall error, λ1 and λ2 are the wavelengths of the L1 and L2 observations, N1 and N2 are the integer ambiguities of the L1 and L2 observations, e is the elevation angle of the observation, Δ∇ is the double phase difference operator, the subscripts H and W represent the tropospheric dry delay and moist delay, respectively, the subscripts n and m represent the reference station and observation point, respectively, and the superscripts s and k represent the reference satellite and observation satellite, respectively.

[0019] Based on the above virtual base station coordinates, the specific process for determining the difference correction information at the virtual base station location through interpolation calculations is as follows:

[0020] For monitoring terminals in a multi-system virtual base station network, the key to achieving real-time high-precision positioning lies in accurately determining distance-related errors. Ionospheric delay error, tropospheric delay error, and overall error have strong spatiotemporal correlations, and the vertical distance between the mobile station and the base station is extremely small compared to the horizontal distance. Therefore, an inverse distance-weighted method is employed to interpolate the three types of errors—ionospheric delay error, tropospheric delay error, and overall error—of the mobile station based on the eastward and northward vectors of the baseline. The observed variances of the interpolation coefficients a1 and a2 corresponding to the eastward and northward vectors are as follows. L = B·X + Δ

[0021] In the equation, L is the observed matrix, B is the design matrix, X is the parameter matrix to be estimated, and △ is the residual matrix, where the weight matrix P of the observation equation is as follows.

number

[0022] ΔX1 and ΔY1 are the eastward and northward distances from each point in the reference station-centered coordinate system of baseline 1 to the reference point, ΔX2 and ΔY2 are the eastward and northward distances from each point in the reference station-centered coordinate system of baseline 2 to the reference point, and ΔX n and ΔY n are the eastward and northward distances from each point in the reference station-centered coordinate system of baseline n to the reference point.

[0023] Based on the least squares principle, the interpolation coefficients [Number] and [Number] corresponding to the eastward and northward vectors can be obtained, and then the error of the mobile station is interpolated by the following formula. [Number]

[0024] In the formula, Err is the baseline error, ΔX and ΔY are the eastward and northward vectors of the baseline, D is the baseline length, a1 and a2 are the interpolation coefficients corresponding to the eastward and northward vectors, Err ν , ΔX ν and ΔY ν are the interpolation error of the mobile station and the eastward and northward vectors of the baseline from the mobile station to the master station, respectively.

[0025] In the above Step 3, the quality verification of the virtual observations is performed. Using the data of the physical reference stations, the quality of the VRS observations is verified and evaluated. It includes three steps. First, the homonym monitoring points are selected and the verification indicators are determined. Then, using the GNSS preprocessing software Anubis, the VRS data and the physical reference station data of the homonym monitoring points are preprocessed and quality-verified respectively. Finally, the quality of the data of each indicator in the two monitoring modes is statistically analyzed to evaluate the VRS data quality.

[0026] Selecting the above homonym monitoring points and determining the verification metrics includes the following: determining the quality verification metrics for the VRS data and physical reference station data of the homonym monitoring points, and the verification metrics include the signal-to-noise ratio (SNR), multipath effects, and cycle slip ratio.

[0027] The method further includes preprocessing and quality verification of VRS data from homonym monitoring points and physical reference station data, respectively. Specifically, the GNSS preprocessing software Anubis is used to process the VRS data from homonym monitoring points and physical reference station observation data, and the numerical values ​​of the quality verification index are determined.

[0028] The method involves statistically analyzing and evaluating the quality of VRS data. This further includes comparing and analyzing quality verification indicators such as the VRS of homonym monitoring points and the signal-to-noise ratio (SNR), multipath effects, and cycle slip ratio of physical reference stations, evaluating the practicality of the VRS analysis data, and determining the data quality.

[0029] The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station provided by the present invention reduces the cost of GNSS landslide monitoring by using a landslide geological disaster monitoring method based on a virtual reference station (VRS), generating virtual reference station data in the monitoring area using the CORS (Continuous Operating Satellite Positioning System), and performing landslide monitoring in place of a physical reference station. At the same time, the VRS is not constrained by geological conditions or distance, and satisfies the requirements of wide-area, highly versatile, and adaptable to complex environments in landslide monitoring. [Brief explanation of the drawing]

[0030] Next, the present invention will be further described with reference to the drawings and embodiments.

[0031] [Figure 1] This is a flowchart of the method of the present invention. [Figure 2] This figure shows the coordinates of the VRS reference station and the corresponding monitoring point in the example. [Figure 3] This is the data analysis result from the example. [Modes for carrying out the invention]

[0032] Next, the technical solutions of the present invention will be described in detail with reference to the drawings and embodiments.

[0033] A GNSS-based landslide geological disaster monitoring and analysis method based on a virtual reference station, Step 1 involves determining the CORS network and selecting monitoring points. This includes determining the coverage status of the CORS network in the construction area, selecting landslide feature monitoring points that require monitoring within the CORS coverage area, and installing monitoring equipment. CORS is defined as continuously operating reference stations. This is an analysis of a multi-system virtual reference station network, and in order to perform landslide monitoring based on a virtual reference station (VRS), the key to Step 2 is to generate an effective VRS near the monitoring point, perform a short baseline analysis on the VRS and the monitoring point, and obtain the displacement of the monitoring point. Step 3 involves quality verification of virtual observation values, using physical reference station data to verify and evaluate the quality of VRS observation values. First, homonym monitoring points are selected and verification indicators are determined. Then, using the GNSS preprocessing software Anubis, preprocessing and quality verification are performed on the VRS data of the homonym monitoring points and the physical reference station data, respectively. Finally, the quality of the data for each indicator in the two monitoring modes is statistically analyzed to evaluate the VRS data quality. Step 4 involves data processing and result analysis, which includes analyzing monitoring data for different periods, statistically calculating the accuracy of the calculations, and statistically evaluating the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions.

[0034] In Step 1 above, the determination of the CORS network and selection of monitoring points include determining the CORS network coverage area within the construction area and selecting landslide feature monitoring points.

[0035] In Step 2 above (which is performed by computer), the method for analyzing virtual reference station observation values ​​includes three parts. First, it analyzes the baselines within the Beidou reference station network and fixes the dual phase difference integer ambiguity. Second, it calculates the ionospheric delay, tropospheric delay, and overall error for each baseline using the Beidou observation values, dual phase difference integer ambiguity, and precise reference station position coordinates for each baseline. Third, it determines the difference correction information at the virtual reference station location by interpolation calculation based on the virtual reference station coordinates and provides a real-time, high-precision virtual reference service to the geological disaster monitoring terminal.

[0036] The analysis process for integer ambiguity in the above-mentioned virtual base station observation values ​​is as follows. 1) The observed carrier phase value and the P-code pseudo-distance constitute a Melbourne-Wuebbena combination, i.e., an MW combination, and are obtained by calculating the wide-lane integer value ambiguity. The specific calculation formula is as follows:

number

[0037] In the formula, the superscripts p and q are identifiers for observation satellite p and reference satellite q, the subscript WL is the GNSS wide lane combination observation value, the subscripts i and j are identifiers for reference stations i and j, λ is the wavelength of the carrier wave, f1 and f2 are the frequencies of the first frequency point L1 and the second frequency point L2 of the GNSS observation value, respectively, Δ∇ is the double phase difference operator, N is the integer ambiguity, P is the pseudo-distance observation value, and φ is the carrier phase observation value.

[0038] 2) Ionospheric free combinations were performed on the carrier phase observations and pseudo-distance observations, and the real solutions to the ionospheric free ambiguity were obtained through analysis. The observed equations are as follows:

number

[0039] In the equation, ρ is the distance from the measurement point to the satellite, Trop is the tropospheric delay error, ε is the carrier phase observation noise, and γ is the pseudo-distance observation noise. Here, since GLONASS employs a frequency-division multiple access signal structure, the GLONASS double phase difference ambiguity, in the distance-unit double phase difference observation equation, needs to be converted from an integer value to one double phase difference integer ambiguity and one single phase difference integer ambiguity, and the observation equation is as follows:

number

[0040] In the formula, λ glo , φ glo , P glo , N glo These are the wavelength, phase, pseudo-distance, and integer ambiguity values ​​of the GLONASS observation, respectively, where p is the observation satellite and q is the reference satellite.

[0041] 3) Using the wide lane integer ambiguity analyzed in step 1), the real solutions of the ionospheric free ambiguity analyzed in step 2), and the corresponding variance-covariance matrices, the real solutions of the narrow lane ambiguity and the corresponding variance-covariance matrices are obtained. Furthermore, the narrow lane integer phase ambiguity is fixed using the LAMBDA method, and the L1 and L2 integer phase ambiguities are calculated.

[0042] The calculation process (performed by computer) for the ionospheric delay, tropospheric delay, and overall error is as follows:

[0043] In calculating the errors of a multi-system virtual reference station, the errors are classified into ionospheric delay, tropospheric delay, secondary ionospheric delay, orbital error, and multipath error, and each is calculated separately. After fixing the integer ambiguity of each baseline, the above three types of errors are calculated and obtained based on the GNSS observation values, dual phase difference integer ambiguity, and precise reference station position coordinates of each baseline. The specific calculation formulas are as follows.

number

[0044] In the formula, Δ∇Iono is the double phase difference ionospheric delay, φ1 is the carrier wave observation at the L1 frequency point, φ2 is the carrier wave observation at the L2 frequency point, MF is the tropospheric mapping function, ZWD is the tropospheric zenith moist delay, ZHD is the tropospheric zenith hydrostatic delay, Other is the overall error, λ1 and λ2 are the wavelengths of the L1 and L2 observations, N1 and N2 are the integer ambiguities of the L1 and L2 observations, e is the elevation angle of the observation, Δ∇ is the double phase difference operator, the subscripts H and W represent the tropospheric dry delay and moist delay, respectively, the subscripts n and m represent the reference station and observation point, respectively, and the superscripts s and k represent the reference satellite and observation satellite, respectively.

[0045] Based on the above virtual reference station coordinates, the specific process (performed by computer) for determining the difference correction information at the virtual reference station location through interpolation calculations is as follows:

[0046] For monitoring terminals in a multi-system virtual base station network, the key to achieving real-time high-precision positioning lies in accurately determining distance-related errors. Ionospheric delay error, tropospheric delay error, and overall error have strong spatiotemporal correlations, and the vertical distance between the mobile station and the base station is extremely small compared to the horizontal distance. Therefore, this specification employs an inverse distance-weighted method to interpolate the three types of errors—ionospheric delay error, tropospheric delay error, and overall error—of the mobile station based on the eastward and northward vectors of the baseline, respectively. Here, the observed variances of the interpolation coefficients a1 and a2 corresponding to the eastward and northward vectors are as follows. L = B·X + Δ

[0047] In the equation, L is the observed matrix, B is the design matrix, X is the parameter matrix to be estimated, and △ is the residual matrix, where the weight matrix P of the observation equation is as follows.

number

[0048] ΔX1 and ΔY1 are the eastward and northward distances from each point to the reference point in the reference station center coordinate system of baseline 1, and ΔX2 and ΔY2 are the eastward and northward distances from each point to the reference point in the reference station center coordinate system of baseline 2, and ΔX n and ΔY n These are the eastward and northward distances from each point in the reference station-centered coordinate system of baseline n to the reference point.

[0049] Interpolation coefficients corresponding to the eastward and northward vectors, based on the least squares principle.

number

number

number

[0050] In the formula, Err is the baseline error, ΔX and ΔY are the east and north vectors of the baseline, D is the baseline length, a1 and a2 are the interpolation coefficients corresponding to the east and north vectors, and Err ν , ΔX ν and ΔY ν These represent the interpolation error of the mobile station and the eastward and northward vectors of the baseline from the mobile station to the main station, respectively.

[0051] This method employs CORS network real-time kinematic positioning (NRTK) technology, enabling the creation of a virtual base station (VRS) at any point within the coverage range of the base station network. This VRS is not subject to geological conditions or distance constraints, reducing the cost of GNSS landslide monitoring, enabling wide-area landslide monitoring, and meeting the requirements of wide coverage, high versatility, and adaptability to complex environments in landslide monitoring.

[0052] In Step 3 above, the quality of virtual observation values ​​is verified. The quality of VRS observation values ​​is verified and evaluated using physical reference station data, and this involves three stages. First, homonym monitoring points are selected and verification indicators are determined. Then, using the GNSS preprocessing software Anubis, preprocessing and quality verification are performed on the VRS data of the homonym monitoring points and the physical reference station data, respectively. Finally, the quality of the data for each indicator of the two monitoring modes is statistically analyzed to evaluate the VRS data quality.

[0053] In selecting the above-mentioned homonym monitoring points and determining the verification indicators, the quality verification indicators for the VRS data of the homonym monitoring points and the physical reference station data are determined, and the verification indicators include the signal-to-noise ratio (SNR), multipath effect, and cycle slip ratio.

[0054] The method further includes preprocessing and quality verification of VRS data from homonym monitoring points and physical reference station data, respectively. Specifically, the GNSS preprocessing software Anubis is used to process the VRS data from homonym monitoring points and physical reference station observation data, and the numerical values ​​of the quality verification index are determined.

[0055] The method involves statistically analyzing and evaluating the quality of VRS data. This further includes comparing and analyzing the VRS data of homonym monitoring points with quality verification indicators such as signal-to-noise ratio (SNR), multipath effects, and cycle slip ratio of physical reference stations, and evaluating the usability of the VRS analysis data and determining the data quality by referring to the technical specification "Technical Specification for the Construction and Acceptance of Beidou Ground Reinforcement System Reference Stations (GBT 39772.2-2021) Part 2: Acceptance Specification".

[0056] In Step 4 above, during data processing and result analysis, monitoring data from different periods is analyzed, calculation accuracy is statistically assessed, and the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions is statistically evaluated.

[0057] (Examples) To verify the present invention, data from the XX project will be used as an example. In this project, the JD38N survey point was selected as a geological hazard monitoring station, and the virtual reference station and the conventional physical reference station were analyzed using VRS to obtain reference station difference data, and the monitoring data was compared. Figure 2 shows the baseline.

[0058] A comparative accuracy analysis was performed using analysis data from 2022 / 09 / 16 16:00:00 to 2022 / 09 / 18 16:00:00. The analysis policy was set to apply a 4-hour Kalman filter to GPS+BDS dual system multi-frequency difference data, with an analysis interval of 5 minutes. The coordinate curves in the east, north, and zenith directions from the analysis results of the two groups are shown in Figure 3.

[0059] By calculating the standard deviation (STD) of the variation in the three-dimensional coordinates of the two groups, the internal agreement accuracy of the positioning results for the two VRS modes and the physical reference station mode can be determined, and the statistical results are as follows. [Table 1]

[0060] To verify the external agreement accuracy of the results from the two groups, this project adopted the single-day static baseline solution as the true value of the baseline vector and calculated the external agreement accuracy of the VRS positioning results from the two groups based on this. The statistical results are as follows. [Table 2]

[0061] The present invention aims to remotely monitor the displacement and trend of landslide areas using sensors and satellite positioning wireless transmission signals. The smaller the difference between the accuracy value obtained by remote calculation and the measurement value by the on-site physical reference station, the higher the accuracy. Figure 3 of the comparative example shows the fitting results between the actual coordinate deviation in three dimensions (east, north, and zenith directions) and the amount of displacement calculated by this application. From the above figure and table data, it can be seen that the deformation monitoring results using difference data from both methods achieve centimeter-level accuracy.

[0062] The actual conditions of the field environment are as follows:

[0063] (1) The baseline is long. The baseline distance between the monitoring point selected for this experiment and the virtual reference station is approximately 5 km, and there is a large difference in altitude.

[0064] (2) In both modes, when generating differential correction data, only the GPS system and the Beidou-2 system are supported, and the Beidou-3 system satellites are not supported, resulting in a limited number of usable satellites during analysis.

[0065] As described above, the current method has achieved centimeter-level accuracy despite the poor actual field conditions (1. long baseline and large height difference, and 2. small number of satellites). Therefore, it is clear that the method of the present invention has better practicality, ease of dissemination, and good performance.

[0066] Furthermore, the results above indicate that the monitoring accuracy of VRS mode approximates that of the physical reference station mode. VRS mode introduces meteorological constraints such as atmospheric errors including ionospheric delay and tropospheric delay. These atmospheric errors exhibit significant spatial correlation, and by incorporating AK25 station data, atmospheric delay information with stronger spatial correlation was integrated into the atmospheric error model construction process. This significantly improved the accuracy of the network RTK's atmospheric error correction number, and furthermore, the VRS accuracy reached a level very close to that of the physical reference station mode.

[0067] (Note) (Note 1) Step 1 involves determining the CORS network and selecting monitoring points, which involves determining the coverage status of the CORS network in the construction area, selecting landslide feature monitoring points that require monitoring within the CORS coverage area, and installing monitoring equipment. This is an analysis of a multi-system virtual reference station network, and involves Step 2 generating a valid virtual reference station VRS near the monitoring point, performing a short baseline analysis on the virtual reference station VRS and the monitoring point, and obtaining virtual observation values ​​for the monitoring point. Step 3 is a quality verification of virtual observation values, in which the quality of virtual observation values ​​obtained from the virtual reference station VRS is verified and evaluated using physical reference station data. A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station, characterized by comprising: data processing and result analysis, including Step 4, which analyzes virtual observation values ​​obtained from VRS over different periods, statistically calculates the calculation accuracy, and statistically evaluates the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions.

[0068] (Note 2) The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in Appendix 1, characterized in that in Step 1, the determination of the CORS network and selection of monitoring points include determining the CORS network coverage area and selecting landslide feature monitoring points in the construction area.

[0069] (Note 3) In Step 2, the method for analyzing virtual reference station observation values ​​comprises three parts: firstly, analyzing the baselines within the CORS network and fixing the dual phase difference integer ambiguity; secondly, using the Beidou observation values, dual phase difference integer ambiguity, and precise reference station position coordinates for each baseline, the ionospheric delay, tropospheric delay, and overall error for each baseline; and thirdly, determining the difference correction information at the virtual reference station position by interpolation calculation based on the virtual reference station coordinates, and providing a real-time virtual reference service to the geological disaster monitoring terminal. This is the GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in Appendix 1.

[0070] (Note 4) The analysis process for integer ambiguity in the analysis of the aforementioned virtual reference station observation values ​​is as follows: 1) The observed carrier phase value and the P-code pseudo-distance constitute a Melbourne-Wuebbena combination, i.e., an MW combination, and are obtained by calculating the wide-lane integer value ambiguity. The specific calculation formula is as follows:

number

number

number

[0071] (Note 5) The calculation process for the ionospheric delay, tropospheric delay, and overall error is as follows: In calculating the errors of a multi-system virtual reference station, the errors are classified into ionospheric delay, tropospheric delay, secondary ionospheric delay, orbital error, and multipath error, and each is calculated separately. After fixing the integer ambiguity of each baseline, the above three types of errors are calculated and obtained based on the GNSS observation values, dual phase difference integer ambiguity, and precise reference station position coordinates of each baseline. The specific calculation formulas are as follows:

number

[0072] (Note 6) The specific process for determining the difference correction information at the virtual reference station location by interpolation calculation based on the aforementioned virtual reference station coordinates is as follows: Using the inverse distance weighting method, the three types of errors of the mobile station—ionospheric delay error, tropospheric delay error, and overall error—are interpolated based on the eastward and northward vectors of the baseline, respectively. Here, the observed variances a1 and a2 of the interpolation coefficients corresponding to the eastward and northward vectors are as follows: L = B·X + Δ In the equation, L is the observed matrix, B is the design matrix, X is the parameter matrix to be estimated, and △ is the residual matrix, where,

number

number

number

number

number

number

[0073] (Note 7) In Step 3, the verification of the quality of virtual observation values ​​involves verifying and evaluating the quality of VRS observation values ​​using data from a physical reference station, and is characterized by three stages: first, selecting a homonym monitoring point and determining the verification indicators; then, using the GNSS preprocessing software Anubis, performing preprocessing and quality verification on the VRS data of the homonym monitoring point and the data from the physical reference station, respectively; and finally, statistically analyzing the quality of the data for each indicator of the two monitoring modes to evaluate the quality of the VRS data, as described in Appendix 1, for GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station.

[0074] (Note 8) A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in Appendix 7, characterized in that, in selecting the homonym monitoring point and determining the verification indicators, quality verification indicators for the VRS data and physical reference station data of the homonym monitoring point are determined, and the verification indicators include the signal-to-noise ratio (SNR), multipath effect, and cycle slip ratio.

[0075] (Note 9) The method further includes preprocessing and quality verification of VRS data from homonym monitoring points and physical reference station data, respectively, and specifically, the GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in Appendix 7, characterized by processing the VRS data from homonym monitoring points and physical reference station observation data using the GNSS preprocessing software Anubis and determining the numerical values ​​of the quality verification index.

[0076] (Note 10) The method involves statistically analyzing and evaluating the quality of VRS data. A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in Appendix 7, further comprising comparing and analyzing VRS data from homonym monitoring points with quality verification index values ​​such as signal-to-noise ratio (SNR), multipath effect, and cycle slip ratio of a physical reference station, to evaluate the practicality of the VRS analysis data and determine the data quality.

Claims

1. Step 1 involves determining the CORS network and selecting monitoring points, which involves determining the coverage status of the CORS network in the construction area, selecting landslide feature monitoring points that require monitoring within the CORS coverage area, and installing monitoring equipment. This is an analysis of a multi-system virtual reference station network, and Step 2 involves generating a valid virtual reference station (VRS) near the monitoring point, performing a short baseline analysis on the virtual reference station (VRS) and the monitoring point, and obtaining virtual observation values ​​for the monitoring point. Step 3 is a quality verification of virtual observation values, where the quality of virtual observation values ​​obtained from the virtual reference station VRS is verified and evaluated using physical reference station data. A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station, characterized by comprising: data processing and result analysis, including Step 4 which analyzes virtual observation values ​​obtained from VRS over different periods, statistically calculates the calculation accuracy, and statistically evaluates the accuracy of the analysis results in the north (N), east (E), and zenith (U) directions.

2. The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station according to claim 1, characterized in that in Step 1, the determination of the CORS network and selection of monitoring points include determining the CORS network coverage area and selecting landslide characteristic monitoring points in the construction area.

3. In Step 2, the method for analyzing virtual reference station observation values ​​comprises three parts: firstly, analyzing baselines within the CORS network and fixing the dual phase difference integer ambiguity; secondly, using the Beidou observation values, dual phase difference integer ambiguity, and precise reference station position coordinates for each baseline, the ionospheric delay, tropospheric delay, and overall error for each baseline; and thirdly, determining differential correction information at the virtual reference station position by interpolation calculation based on the virtual reference station coordinates, and providing a real-time virtual reference service to the geological disaster monitoring terminal. This is the GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station as described in claim 1.

4. The analysis process for integer ambiguity in the analysis of the aforementioned virtual reference station observation values ​​is as follows: 1) The observed carrier phase value and the P-code pseudo-distance constitute a Melbourne-Wuebbena combination, i.e., an M-W combination, and are obtained by calculating the wide lane integer value ambiguity. The specific calculation formula is as follows: [Math 1] In the formula, the superscripts p and q are identifiers for observation satellite p and reference satellite q, the subscript WL is the GNSS wide lane combination observation value, the subscripts i and j are identifiers for reference stations i and j, λ is the wavelength of the carrier wave, and f １ and f ２ L1 and L2 are the frequencies of the first and second frequency points of the GNSS observation, respectively; Δ∇ is the double phase difference operator; N is the integer ambiguity; P is the pseudo-distance observation; and φ is the carrier phase observation. 2) Ionospheric free combinations were performed on the carrier phase observations and pseudo-distance observations, and the real solutions to the ionospheric free ambiguity were obtained through analysis. The observed equations are as follows: [Math 2] In the equation, ρ is the distance from the measurement point to the satellite, Trop is the tropospheric delay error, ε is the carrier phase observation noise, and γ is the pseudo-distance observation noise. Here, because GLONASS employs a frequency-division multiple access signal structure, the GLONASS dual phase-difference ambiguity, in the distance-unit dual phase-difference observation equation, needs to be converted not as an integer value, but into one dual phase-difference integer ambiguity and one single phase-difference integer ambiguity. The observation equation is as follows: [Math 3] In the formula, λ ｇｌｏ , φ ｇｌｏ , P ｇｌｏ , N ｇｌｏ These are the wavelength, phase, pseudo-distance, and integer ambiguity values ​​of the GLONASS observation, respectively, where p is the observation satellite and q is the reference satellite. 3) A GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station, characterized in that the real solutions of the wide lane integer ambiguity analyzed in step 1), the real solutions of the ionospheric free ambiguity analyzed in step 2), and the corresponding variance-covariance matrix are used to obtain the real solutions of the narrow lane ambiguity and the corresponding variance-covariance matrix, the narrow lane integer phase ambiguity is fixed using the least squares ambiguity uncorrelated adjustment method LAMBDA, and further calculations are performed to obtain the GNSS satellite L1 and L2 integer phase ambiguity.

5. The calculation process for the ionospheric delay, tropospheric delay, and overall error is as follows: In calculating the errors of a multi-system virtual reference station, the errors are classified into ionospheric delay, tropospheric delay, secondary ionospheric delay, orbital error, and multipath error, and each is calculated separately. After fixing the integer ambiguity of each baseline, the above three types of errors are calculated and obtained based on the GNSS observation values, dual phase difference integer ambiguity, and precise reference station position coordinates of each baseline. The specific calculation formulas are as follows: [Math 4] where Δ∇Iono is the double differential ionospheric delay, and φ １ is the carrier phase observation value at the L1 frequency point, and φ ２ is the carrier phase observation value at the L2 frequency point, MF is the troposphere mapping function, ZWD is the troposphere zenith wet delay, ZHD is the troposphere zenith hydrostatic delay, Other is the total error, λ １ and λ ２ are the wavelengths of the L1 and L2 observations, N １ and N ２ are the integer value ambiguities of the L1 and L2 observations, e is the elevation angle of the observation value, Δ∇ is the double differential operator, the subscripts H and W represent the dry delay and wet delay of the troposphere respectively, the subscripts n and m represent the reference station and the monitoring point respectively, and the superscripts s and k represent the reference satellite and the observed satellite respectively. The GNSS ground subsidence geological disaster monitoring and analysis method based on the virtual reference station according to claim 4, characterized in that.

6. The specific process for determining the difference correction information at the virtual reference station location by interpolation calculation based on the aforementioned virtual reference station coordinates is as follows: Using the inverse distance weighting method, the three types of errors of the mobile station—ionospheric delay error, tropospheric delay error, and overall error—are interpolated based on the baseline's eastward and northward vectors, respectively. Here, the observed variance a of the interpolation coefficients corresponding to the eastward and northward vectors is used. １ and a ２ The following applies: L = B・X + Δ In the equation, L is the observed matrix, B is the design matrix, X is the parameter matrix to be estimated, and △ is the residual matrix, where, [Math 5] ΔX １ and ΔY １ ΔX is the eastward and northward distance from each point in the coordinate system of the base station 1 to the reference point, and ２ and ΔY ２ ΔX is the eastward and northward distance from each point in the reference station center coordinate system of baseline 2 to the reference point, and ｎ and ΔY ｎ These are the eastward and northward distances from each point in the reference station-centered coordinate system of baseline n to the reference point. Interpolation coefficients corresponding to the east and north vectors, based on the least squares principle. [Math 6] and [Number 7] This can be obtained, and then the error of the mobile station is interpolated using the following formula, [Number 8] In the formula, Err is the baseline error, ΔX and ΔY are the east and north vectors of the baseline, and D is the baseline length. [Number 9] and [Number 10] These are interpolation coefficients corresponding to the east and north vectors, and Err ν , ΔX ν and ΔY ν The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station according to claim 5, characterized in that the interpolation error of the mobile station and the eastward and northward vectors of the baseline from the mobile station to the main station are, respectively.

7. In Step 3, the verification of the quality of virtual observation values ​​involves verifying and evaluating the quality of VRS observation values ​​using data from a physical reference station, and includes three stages: first, selecting homonym monitoring points and determining verification indicators; then, using the GNSS preprocessing software Anubis, performing preprocessing and quality verification on the VRS data of the homonym monitoring points and the data from the physical reference station; and finally, statistically analyzing the quality of the data for each indicator of the two monitoring modes to evaluate the VRS data quality, as described in claim 1, a GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station.

8. The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station according to claim 7, characterized in that, in selecting the homonym monitoring point and determining the verification indicators, quality verification indicators for the VRS data and physical reference station data of the homonym monitoring point are determined, and the verification indicators include the signal-to-noise ratio (SNR), multipath effect, and cycle slip ratio.

9. The method further includes preprocessing and quality verification of the VRS data of the homonym monitoring point and the physical reference station data, respectively, and specifically, the GNSS preprocessing software Anubis is used to process the VRS data of the homonym monitoring point and the physical reference station observation data to determine the numerical value of the quality verification index, characterized in that the GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station is described in claim 7.

10. The method involves statistically analyzing and evaluating the quality of VRS data. The GNSS landslide geological disaster monitoring and analysis method based on a virtual reference station according to claim 7, further comprising comparing and analyzing VRS data from a homonym monitoring point with quality verification index values ​​such as signal-to-noise ratio (SNR), multipath effect, and cycle slip ratio of a physical reference station, evaluating the practicality of the VRS analysis data, and determining the data quality.

Images