Method and system for improving dam GNSS deformation monitoring precision through base station and observation station combined network adjustment

Abstract

The invention relates to the technical field of dam safety monitoring, in particular to a method and system for improving dam GNSS deformation monitoring precision through base station and observation station combined network adjustment. The method comprises the following steps: arranging base stations and monitoring points to form a GNSS monitoring network; gNSS original observation data are collected in real time and preprocessed; constructing a joint network adjustment model, and taking base station coordinates as constraints and monitoring point coordinates as to-be-estimated parameters; performing baseline resolving based on the double-difference carrier phase observed quantity to obtain a baseline vector and a covariance matrix thereof; carrying out overall adjustment on the baseline vector by adopting a robust estimation method, and solving an optimal coordinate estimated value and precision information of the monitoring point; carrying out deformation analysis on the basis of the adjusted coordinate time sequence, and extracting tendency, periodicity and abnormal deformation; and outputting a deformation monitoring result, and carrying out visual display and early warning. According to the method, the precision and reliability of dam deformation monitoring are effectively improved through combination of network adjustment and robust estimation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of engineering safety and deformation monitoring, and particularly relates to a method and system for improving GNSS deformation monitoring precision of a dam by joint network adjustment of base stations and survey stations. BACKGROUND

[0002] As important water conservancy infrastructure, the safe operation of a dam has a significant impact on the safety of life and property of downstream people and social and economic development. Traditional dam deformation monitoring methods mainly include leveling and total station measurement, which have high precision but have limitations such as low efficiency, poor real-time performance, and great influence by weather conditions, and are difficult to meet the requirements of high frequency, automation, and real-time performance for modern dam safety monitoring.

[0003] With the development of global navigation satellite system (GNSS) technology, GNSS-based dam deformation monitoring technology has been widely applied. GNSS technology has the advantages of all-weather, all-day, and high automation, and can obtain real-time three-dimensional coordinate changes of monitoring points. However, in practical application, GNSS deformation monitoring still faces many challenges: first, error sources such as multipath effect, ionospheric delay, and tropospheric delay seriously affect monitoring precision, especially in the dam environment of canyon terrain, the multipath effect is particularly significant; second, single-point positioning mode is difficult to meet the precision requirements of millimeter-level deformation monitoring of the dam, and traditional differential positioning technology is limited by baseline length, and as the baseline increases, the error correlation decreases, and the precision rapidly decreases; in addition, the existing data processing method has insufficient resistance to gross errors, and the overall solution result may be biased due to a few abnormal observation values.

[0004] At present, domestic and foreign scholars have carried out a lot of research on these problems. Some researches adopt precise point positioning (PPP) technology to improve positioning precision through precise ephemeris and clock error products, but the convergence time of PPP is long, which is difficult to meet the real-time monitoring demand. Another research adopts network RTK technology to improve positioning precision through regional error modeling, but this method requires a dense reference station network, which has high cost. In terms of data processing, the least squares method is usually used for parameter estimation, but it is sensitive to gross errors and needs strict data preprocessing. In addition, the existing technology focuses more on monitoring of the plane position, and the improvement of the monitoring precision in the height direction is limited, while the vertical displacement of the dam is often a key indicator for safety evaluation.

[0005] Therefore, there is an urgent need for a new type of GNSS deformation monitoring method that can effectively suppress errors and improve monitoring precision, especially in the height direction, and is suitable for complex dam environments. The present application aims to solve the deficiencies in the existing technology by using joint network adjustment technology of base stations and survey stations combined with robust estimation method, and to provide a high-precision and high-reliability solution for dam safety monitoring. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks.

[0007] In a first aspect, embodiments of this application provide a method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks, the method comprising: S1. Deploy at least two base stations and multiple monitoring points in and around the dam to form a GNSS monitoring network; S2. Real-time acquisition of raw GNSS observation data from each base station and monitoring point, transmission of data to the processing center, and data preprocessing, including cycle slip detection and repair, multipath effect identification and reduction, and outlier removal; S3. Construct a joint network adjustment model for base stations and monitoring points. The model includes observation equations with station coordinates as parameters, where base station coordinates are known or strongly constrained conditions, and monitoring point coordinates are parameters to be estimated. S4. Baseline calculation is performed based on double-difference carrier phase observations to obtain the baseline vector and its covariance matrix; S5. Use the joint network adjustment model to perform overall adjustment of the baseline vector, and use robust estimation method to solve for the optimal coordinate estimate and accuracy information of the monitoring points; S6. Based on the adjusted coordinate time series, perform deformation analysis to extract dam deformation information, including trend deformation, periodic deformation and abnormal deformation; S7. Output deformation monitoring results and provide visualization and early warning judgment.

[0008] Optionally, in one implementation of the first aspect of the present invention, the deployment of base stations and monitoring points in step S1 adopts a hierarchical optimization configuration strategy, specifically including: Multiple base stations are distributed in stable geological areas on both sides of the dam axis. The base stations form a closed network with reasonable spacing, and each base station has line of sight with multiple monitoring points. Monitoring points are evenly distributed along the dam axis, with denser deployment at key locations, forming multiple monitoring sections in areas sensitive to dam deformation. The baseline network layer consists of all base stations, the monitoring network layer consists of all monitoring points, and the encryption network layer adds redundant monitoring points in key areas. The average side length ratio of the network is kept in a reasonable proportion, the network graph strength factor reaches an excellent level, and the point accuracy attenuation factor is controlled at a low level.

[0009] Optionally, in one implementation of the first aspect of the present invention, step S2 specifically includes the following processing procedures: By deploying high-precision GNSS receivers at base stations and monitoring points, raw observation data from multiple systems and frequencies are collected synchronously to ensure the consistency of data time reference and the high sampling rate of observation values. Cycle slips were detected by combining MW and GF combinations, and accurate repair and marking of cycle slips were performed based on polynomial fitting and ionospheric residual methods. An empirical model of multipath error is constructed using signal-to-noise ratio observations, and the weights of the observations are adjusted using a satellite elevation angle-dependent stochastic model, which effectively suppresses the adverse effects of multipath effects. Based on robust statistical theory, data probing methods or weighting function schemes are used to identify and eliminate gross errors, and residual analysis is used to further investigate abnormal observations. Calculate quality indicators such as data integrity rate, multipath error, and signal-to-noise ratio; standardize the observation data and generate a data quality report. The observation data, after cycle slip repair, multipath reduction, and gross error removal, along with the corresponding covariance matrix, are transmitted to the subsequent processing module in real time.

[0010] Optionally, in one implementation of the first aspect of the present invention, when constructing the joint network adjustment model in step S3, a regional network adjustment method based on connection point constraints is adopted, specifically including the following sub-steps: A free network is established using GNSS observation data between each monitoring point and the base station, as well as connection points obtained through image matching or measurement; a preliminary solution is performed using the uncontrolled free network adjustment method to construct a robust network geometry. During the adjustment process, an affine transformation model is used to compensate for systematic errors in GNSS observations or preliminary coordinates. The affine transformation parameters are used as parameters to be estimated and solved together with the station coordinates in the adjustment model. Based on the free network adjustment, known coordinate points evenly distributed in the monitoring area are added as control points to perform control network adjustment; the correction of the coordinates of each monitoring point is calculated by least squares estimation to accurately solve for the coordinates of the monitoring points. The elevation values ​​required during the adjustment process are obtained by interpolation using a digital elevation model; The adjustment results require that the mean square error of the connection points after the adjustment of the free network is better than the set threshold, and the mean square error of the points after the adjustment of the control network meets the accuracy requirements of dam deformation monitoring.

[0011] Optionally, in one implementation of the first aspect of the present invention, based on the affine transformation compensation for system errors, the control network adjustment further employs a weighted hybrid estimation method to optimize the stochastic model and balance the weights of observed information and prior information. The specific method is as follows: Construct the following adaptive weighted estimation optimization model: , in, For GNSS observation vectors, Here, A is the design matrix of the observation equation; H is the design matrix of the prior constraint equation; and X is the vector of parameters to be estimated, including the coordinates of the monitoring points and the affine transformation parameters. , These are the weight matrices for the observed values ​​and the prior constraint values, respectively; This is a weighting factor used to dynamically balance the contributions of observed information and prior constraint information; Optimal weight The trace of the mean square error matrix of the parameter estimates is determined by minimizing the trace of the mean square error matrix, i.e.: , The mean square error matrix is ​​calculated by the following formula: , in, The unit weighted variance estimates of the observed values ​​and prior constraint values ​​are dynamically determined using the following variance component estimation formula: , , in, These represent the number of observations and the number of prior constraints, respectively. , , , The trace of a matrix is ​​the sum of the elements on its main diagonal.

[0012] Optionally, in one implementation of the first aspect of the present invention, step S4 specifically includes the following processing procedures: By integrating observation data from multiple systems, common-view satellites are selected to form inter-station single-difference and inter-satellite double-difference. A double-difference carrier phase observation equation is established with integer ambiguity and baseline vector as parameters. Ionospheric-free combination is used to eliminate first-order ionospheric delay and to uniformly handle the inter-frequency deviation and hardware delay differences between systems. Cycle slips are detected by combining MW and Geometry-free methods, and cycle slips are repaired based on polynomial fitting. For cycle slips that cannot be repaired, data segmentation is performed. The LAMBDA algorithm is used to search for fixed double-difference integer ambiguities. The reliability of ambiguity fixing is confirmed by ratio test. For ambiguities that fail to be fixed, a partial fixing strategy is adopted. Using a fixed integer ambiguity, the baseline vector is solved by the least squares method, and its covariance matrix is ​​calculated. The quality of the observations is verified by residual analysis, the observation weight matrix is ​​adjusted by variance component estimation, and the final baseline vector and its accuracy information are output.

[0013] Optionally, in one implementation of the first aspect of the present invention, step S5 specifically includes the following processing procedures: Using all baseline vectors and their covariance matrices as observed values, monitoring point coordinates as parameters to be estimated, and base station coordinates as fixed or strong constraints, a unified error equation and constraint equation for the entire network are established. The weight matrix of the observations is constructed based on the covariance matrix provided by the baseline solution, and the weights of various observations are adaptively adjusted by the variance component estimation method. The least squares estimation method is used to solve the normal equation. When there are gross errors in the observations, the method is switched to robust estimation, and the influence of gross errors is controlled by the weight function. Calculate the covariance matrix of the parameter estimates to assess the point accuracy, relative accuracy, and error ellipse parameters; The adjustment results were comprehensively evaluated using statistical tests, including network closure error test, parameter significance test, and residual distribution test. Based on the adjustment results, the observation weights, stochastic models, and estimation strategies are adjusted, and multiple iterative adjustments are performed until the results converge.

[0014] Optionally, in one implementation of the first aspect of the present invention, step S7 specifically includes the following processing procedures: Based on the adjusted coordinate time series, time series analysis is used to separate the trend deformation, periodic deformation and random deformation components, and to calculate the deformation rate, acceleration and cumulative deformation of each monitoring point. Generate multi-dimensional deformation monitoring results, including a table of monitoring point coordinates, a deformation process line map, a deformation contour map, and a deformation vector map. Establish an integrated interactive chart and 3D model of the dam to enable dynamic playback, multi-period comparison and spatial query of deformation data; Based on the structural characteristics and design requirements of the dam, a multi-level early warning indicator system was established, and deformation thresholds for different levels were set. Real-time monitoring of deformation data; when the monitored value exceeds the warning threshold, an early warning signal is automatically triggered, and a reliability assessment of the early warning is performed. Generate early warning reports that include the warning level, location, and recommended measures, and disseminate the warning information through multiple channels.

[0015] Secondly, embodiments of this application provide a system for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks, applied to the method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks as described in the second aspect, the system comprising: The network deployment module is used to configure and manage at least two base stations and multiple monitoring points deployed in and around the dam to form the physical topology of the GNSS monitoring network. The data acquisition and transmission module is used to collect raw GNSS observation data from each base station and monitoring point in real time and transmit the data to the processing center. The data preprocessing module is used to perform cycle slip detection and repair, multipath effect identification and reduction, and outlier removal on the received raw GNSS observation data. The joint network adjustment modeling module is used to construct a joint network adjustment model of base stations and monitoring points. The model includes observation equations with station coordinates as parameters, where base station coordinates are known or strong constraints and monitoring point coordinates are parameters to be estimated. The baseline calculation module is used to perform baseline calculation based on double-difference carrier phase observations to obtain the baseline vector and its covariance matrix. The overall adjustment processing module is used to perform overall adjustment of the baseline vector using the joint network adjustment model, and to solve the optimal coordinate estimate and accuracy information of the monitoring points using the robust estimation method. The deformation analysis module is used to perform deformation analysis based on the adjusted coordinate time series, and to extract dam deformation information, including trend deformation, periodic deformation and abnormal deformation. The results output and early warning module is used to output deformation monitoring results and provide visualization and early warning judgments.

[0016] Thirdly, embodiments of this application provide an electronic device, including: processor; Memory used to store processor-executable instructions; The processor is configured to implement, when executing the instructions, a method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks as described in the first aspect.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing a program that instructs a device to perform the method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks as described in the first aspect.

[0018] This invention relates to the field of dam safety monitoring technology, and particularly to a method and system for improving the accuracy of GNSS deformation monitoring of dams through joint network adjustment of base stations and monitoring stations. The method includes: deploying base stations and monitoring points to form a GNSS monitoring network; real-time acquisition and preprocessing of raw GNSS observation data; constructing a joint network adjustment model, using base station coordinates as constraints and monitoring point coordinates as parameters to be estimated; performing baseline calculation based on double-difference carrier phase observations to obtain the baseline vector and its covariance matrix; using a robust estimation method to perform overall adjustment of the baseline vector, solving for the optimal coordinate estimates and accuracy information of the monitoring points; performing deformation analysis based on the adjusted coordinate time series, extracting trend, periodic, and abnormal deformations; and outputting the deformation monitoring results for visualization and early warning. This invention effectively improves the accuracy and reliability of dam deformation monitoring through joint network adjustment and robust estimation.

[0019] Beneficial effects:

[0020] 1. Significantly improve monitoring accuracy: By effectively eliminating common errors through joint network adjustment and suppressing the influence of gross errors by combining robust estimation, the accuracy of monitoring point coordinate calculation is significantly improved, especially the monitoring capability in the elevation direction.

[0021] 2. Enhance system reliability: From data preprocessing to final solution, a full-process quality control is implemented, employing multiple methods such as multi-path reduction, gross error elimination, and robust estimation to ensure the reliability and stability of the results.

[0022] 3. Achieve automated and efficient processing: The method achieves full automation from data acquisition and processing to deformation analysis and early warning, which greatly improves operational efficiency and meets the needs of real-time monitoring of dam safety.

[0023] 4. Provide comprehensive decision support: The final output includes multi-dimensional deformation results and early warning information, providing timely and intuitive scientific basis for dam safety operation assessment and risk prevention and control. Attached Figure Description

[0024] Figure 1 This is a schematic flowchart illustrating a method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks, as provided in an embodiment of this application.

[0025] Figure 2 A system architecture diagram for improving the accuracy of GNSS deformation monitoring of dams through base station and monitoring station joint network adjustment, provided in an embodiment of this application. Figure 3 A schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] 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 a part of the embodiments of this application, and not all of them.

[0027] It should be noted that, in the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0028] It should be noted that in the embodiments of this application, the terms "first," "second," etc., are used only for descriptive purposes and should not be construed as indicating or implying relative importance, nor as indicating or implying order. Features specified as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0029] Based on the embodiments described in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] Example 1

[0031] Figure 1 This is a schematic flowchart illustrating a method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks, provided as an embodiment of this application. Figure 1 As shown, a method for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks includes: S1. Deploy at least two base stations and multiple monitoring points in and around the dam to form a GNSS monitoring network. The deployment of the GNSS monitoring network must consider the placement of base stations and monitoring points to ensure coverage of the dam and its surrounding area. Base stations serve as reference points, and monitoring points are used to monitor deformation. Coverage range, signal reception conditions, and data transmission paths must be considered during deployment.

[0032] Specifically, in this embodiment, the deployment of base stations and monitoring points in step S1 adopts a hierarchical optimization configuration strategy based on terrain and geological conditions and monitoring accuracy requirements, specifically including: The deployment of base stations meets the following conditions: multiple base stations are distributed in areas with stable geological conditions on both sides of the dam axis; the base stations are mutually visible and form a closed network, and their spacing is kept within a reasonable range according to the actual terrain and visibility conditions; each base station maintains good visibility conditions with multiple monitoring points.

[0033] The deployment of monitoring points shall meet the following conditions: monitoring points shall be evenly distributed along the dam axis, with denser deployment at key structural locations such as the dam crest, dam shoulders, and spillway gates; multiple transverse monitoring sections shall be deployed in areas sensitive to dam deformation, each section containing a sufficient number of monitoring points. For example, monitoring points shall be evenly distributed along the dam axis at intervals of 20-50 meters, with denser deployment at intervals of 10-20 meters at key locations such as the dam crest, dam shoulders, and spillway gates; multiple transverse monitoring sections shall be formed in areas sensitive to dam deformation, with at least 3 monitoring points deployed at each section.

[0034] The GNSS monitoring network adopts a layered architecture: the reference network layer consists of all base stations; the monitoring network layer consists of all monitoring points; and the encryption network layer adds redundant monitoring points in key areas. Alternatively, the GNSS monitoring network can adopt a three-layer architecture: the reference network layer consists of all base stations, providing a stable reference frame; and the monitoring network layer consists of all monitoring points, used to capture dam deformation information.

[0035] The network design should meet the following accuracy indicators: the average side length ratio of the network should be maintained at a reasonable level; the network graph strength factor should reach an excellent level; and the point accuracy attenuation factor should be controlled at a low level. For example, the average side length ratio of the network should not be greater than 2:1; the network graph strength factor should be better than 0.8; and the point accuracy attenuation factor (PDOP) value should be controlled below 3.

[0036] S2. Real-time acquisition of raw GNSS observation data from various base stations and monitoring points, transmission of the data to the processing center, and data preprocessing, including cycle slip detection and repair, multipath effect identification and mitigation, and outlier removal. The GNSS data processing workflow includes data preprocessing, such as cycle slip removal, multipath effect detection, and outlier removal. Preprocessing is a crucial step in data processing to ensure data quality.

[0037] Specifically, in this embodiment, step S2 includes the following processing procedures: S2.1. By deploying high-precision GNSS receivers at base stations and monitoring points, raw observation data from multiple systems and frequencies are collected synchronously to ensure the uniformity of data time reference and the high sampling rate of observation values, thereby achieving real-time acquisition and synchronization of multi-source data.

[0038] S2.2. Cycle slips are detected by combining the MW (Melbourne-Wubbena) and GF (Geometry-Free) methods. Cycle slips are accurately repaired and marked based on polynomial fitting and ionospheric residual methods, thus realizing cycle slip detection and repair.

[0039] S2.3. Construct an empirical model of multipath error using signal-to-noise ratio (SNR) observations, and adjust the weights of the observations using a satellite elevation angle-dependent stochastic model to effectively suppress the adverse effects of multipath effects, thereby achieving multipath error modeling and mitigation.

[0040] S2.4 Based on robust statistical theory, the Bararda data detection method or IGGIII scheme is used to identify and eliminate gross errors. Residual analysis is used to further investigate abnormal observations, thereby realizing the identification and elimination of abnormal observations.

[0041] S2.5 Calculate quality indicators such as data integrity rate, multipath error, and signal-to-noise ratio, standardize the observation data, and generate a data quality report to achieve data quality assessment and standardization.

[0042] S2.6. The clean observation data, which has undergone cycle slip repair, multipath reduction and gross error removal, along with the corresponding covariance matrix, is transmitted to the subsequent processing module in real time to realize the real-time output of preprocessing results.

[0043] The data preprocessing process adopts a fully automated pipeline processing method and establishes a strict quality control closed loop to ensure that the preprocessed observation data meets the reliability and integrity requirements of high-precision deformation monitoring.

[0044] S3. Construct a joint network adjustment model for base stations and monitoring points. This model includes observation equations with station coordinates as parameters, where base station coordinates are known or strongly constrained, and monitoring point coordinates are parameters to be estimated. In GNSS data processing, a joint network adjustment model needs to be constructed, with base station coordinates as known or strongly constrained and monitoring point coordinates as parameters to be estimated. This model is used for overall adjustment to improve the accuracy of coordinate calculation.

[0045] First, systematic errors are corrected through affine transformation. Then, based on "cleaner" data, the stochastic model is optimized through weighted mixture estimation to balance the weights of various observations. This embodies the fundamental principle of measurement adjustment: eliminate systematic errors first, then address random errors.

[0046] Specifically, in this embodiment, when constructing the joint network adjustment model in step S3, a regional network adjustment method based on connectivity point constraints is adopted, which specifically includes the following sub-steps: S3.1 Free Network Construction and Preliminary Adjustment: Using GNSS observation data between each monitoring point and the base station, and the connection points automatically extracted between each monitoring point through high-precision image matching technology, a free network with redundant observations is constructed; the uncontrolled free network adjustment method is used for preliminary calculation to obtain the initial estimate of the station coordinates and evaluate the internal consistency of the network, and to preliminarily identify possible gross errors and system deformations.

[0047] S3.2 Affine Transformation System Error Modeling and Compensation: The following affine transformation model is introduced into the adjustment model to compensate for systematic errors related to station location in GNSS observations: , in, These are the original observation coordinates. For the compensated coordinates, The affine transformation parameters are Δx and Δy, which are random error terms. The affine transformation parameters are solved together with the station coordinates in a unified adjustment model as parameters to be estimated, thereby realizing the dynamic elimination of systematic errors.

[0048] S3.3, Control Network Constraint Adjustment and Elevation Constraint Introduction: Based on the free network adjustment and compensation for systematic errors, the coordinates of known high-level control points evenly distributed in key parts of the dam (such as dam abutments, galleries, and spillways) are introduced as strong constraints. The prior elevation information of each point is obtained by interpolation through the digital elevation model (DEM) as a weighted constraint, and the following constraint adjustment model is constructed: Subject to: , Where V is the observation residual vector, P is the observation weight matrix, C is the constraint equation coefficient matrix, d is the constraint constant term, and X is the vector of parameters to be estimated, including coordinate parameters and affine transformation parameters.

[0049] S3.4 Network reinforcement based on connection point constraints: By utilizing the connection points automatically generated between monitoring points through image matching, additional geometric constraints are added to the adjustment model to enhance the rigidity and overall stability of the network, especially improving the accuracy in areas with sparse monitoring points.

[0050] S3.5 Accuracy Assessment and Iterative Optimization: The adjustment process requires that the mean square error of the connection points after the adjustment of the free network be better than 1 pixel, the mean square error of the points after the adjustment of the control network be better than ±1.5 mm, and the horizontal accuracy be better than ±2 mm. The weights of various observations and constraints are adaptively adjusted by the variance component estimation method, and multiple iterations are performed until the results converge and meet the accuracy index.

[0051] For the elevation constraint, the required elevation values ​​during the adjustment process are obtained through interpolation using a digital elevation model. The adjustment results require that the mean square error of the tie points after the free network adjustment is better than a set threshold, and that the mean square error of the points after the control network adjustment meets the accuracy requirements for dam deformation monitoring.

[0052] The first step, affine transformation, corrects systematic errors—"eliminating systematic biases"—handling deterministic and regular errors. Systematic errors in GNSS-RFM (rational function model) orthorectification may originate from: sensor calibration residuals (such as interior orientation elements and lens distortion), imperfections in the atmospheric refraction model, satellite orbit and clock error residuals, and systematic deformations caused by imaging geometry. The affine transformation model can effectively compensate for systematic deformations such as translation, rotation, scaling, and shearing in the image. Based on this, by introducing connection points, the affine transformation parameters are accurately solved, and this model is used to correct the original observation coordinates. This step essentially enhances and corrects the function model of the observations, separating and eliminating the systematic error component in the observations to the greatest extent possible. After this step, the systematic bias in the data is significantly reduced, leaving residuals that are closer to random noise with a mean of zero, providing a "cleaner" data foundation for the next step of stochastic model optimization.

[0053] Furthermore, based on affine transformation compensation for system errors, the control network adjustment further employs a weighted hybrid estimation method to optimize the stochastic model and balance the weights of observed and prior information. The specific method is as follows: Construct the following adaptive weighted estimation optimization model: , in, For GNSS observation vectors, Here, A is the design matrix of the observation equation; H is the design matrix of the prior constraint equation; and X is the vector of parameters to be estimated, including the coordinates of the monitoring points and the affine transformation parameters. , These are the weight matrices for the observed values ​​and the prior constraint values, respectively; This is a weighting factor used to dynamically balance the contributions of observed information and prior constraint information; Optimal weight The trace of the mean square error matrix of the parameter estimates is determined by minimizing the trace of the mean square error matrix, i.e.: , The mean square error matrix is ​​calculated by the following formula: , in, The unit weighted variance estimates of the observed values ​​and prior constraint values ​​are dynamically determined using the following variance component estimation formula: , , in, These represent the number of observations and the number of prior constraints, respectively. , , , The trace of a matrix is ​​the sum of the elements on its main diagonal.

[0054] In this process, a weighted mixture estimation is used to optimize the stochastic model – “balancing randomness” – to handle random and statistical errors and to integrate observation information of different precision and nature. For data after affine transformation, questions arise regarding how to accurately determine its random noise level (variance), how to balance the relative reliability between observations from different sources (e.g., GPS / BDS / GLONASS multi-system observations, observations from different receiver brands, historical prior coordinates and current observations), and how to reasonably integrate the prior information of base station coordinates (usually very precise) as a constraint into the adjustment system, rather than simply treating it as a fixed value. These issues are addressed through variance component estimation or optimal weight selection (e.g., minimizing the trace) to dynamically determine the weight matrix of different observation groups. and This results in high-precision observations contributing more to the adjustment, while low-precision observations contribute less. Simultaneously, through mixed estimation, prior information from the base stations (such as constraints with random errors) is incorporated. ) and GNSS observations at monitoring points (such as L) is placed in the same adjustment model for joint solution. It does not rigidly fix the base station, but rather acknowledges that it also has a certain degree of uncertainty (due to...). This method utilizes high-precision information in a more scientific and gentler way, achieving an optimal statistical description of the stochastic characteristics of the observations. It makes reasonable use of all available information (observations and prior information), making the final solution statistically the optimal linear unbiased estimate, thus improving the robustness of the entire solution system. Even when some observations are of poor quality, the results will not fluctuate drastically.

[0055] This strategy of "affine transformation first, then weighted mixture estimation" functions as a meticulously designed pipeline. First, it addresses the issue of accuracy by correcting the model through affine transformation, ensuring the estimation results are free from systematic bias. Then, it addresses the issue of precision by optimizing the stochastic model through weighted mixture estimation, pursuing the highest estimation accuracy and robustness while maintaining unbiasedness. Both are indispensable, and the order is crucial; together, they constitute a complete and powerful modern measurement adjustment solution.

[0056] S4. Baseline calculation is performed based on double-difference carrier phase observations to obtain the baseline vector and its covariance matrix. The double-difference method is a core method in GNSS data processing, used to eliminate satellite clock errors and receiver errors, improving baseline calculation accuracy. The baseline calculation results are used for subsequent adjustment.

[0057] Specifically, in this embodiment, step S4 includes the following processing procedures: S4.1 Constructing a multi-system double-difference observation equation. By fusing observation data from GPS, BDS, and GLONASS systems, common-view satellites in the same observation period are selected. Inter-station single difference is formed between base stations and monitoring points, and inter-satellite double difference is formed between satellites. A double-difference carrier phase observation equation with integer ambiguity and baseline vector as parameters is established. Ionospheric-free combination is used to eliminate first-order ionospheric delay and uniformly handle the frequency deviation and hardware delay differences between systems. S4.2 Joint Processing of Cycle Slip Detection and Repair. A combination of MW and Geometry-free methods is used to jointly detect cycle slips. Cycle slip repair is performed based on polynomial fitting. Unrepairable cycle slips are segmented, and unreliable observation periods are marked. S4.3 Solving integer ambiguities based on the LAMBDA algorithm. An improved LAMBDA algorithm is used to search for fixed double-difference integer ambiguities. The reliability of ambiguity fixing is confirmed by ratio testing and fixing failure detection. For ambiguities that fail to be fixed, a partial fixing strategy is adopted to improve the success rate of ambiguity fixing.

[0058] Furthermore, the integer ambiguity resolution in step S4.3 employs a multi-strategy collaborative fixing method based on an improved LAMBDA algorithm. Specifically, this includes: using an integer least squares search algorithm with reduced correlation processing to reduce the correlation between ambiguity parameters through integer transformation, constructing an efficient search space, and quickly determining integer ambiguity candidate values; employing a multi-criteria joint verification strategy to confirm the reliability of ambiguity fixing, including: ratio test: assessing the confidence level of ambiguity fixing by comparing the ratio of the sum of squared residuals of the optimal candidate solution and the second-best candidate solution; fixing failure detection: identifying and excluding unreliable ambiguity fixing results based on residual analysis and posterior probability calculation; and employing a partial ambiguity fixing strategy. The system handles ambiguity fixation failures as follows: When all ambiguities fail to be fixed, a subset of ambiguities is automatically selected for fixation; based on variance-covariance matrix information, ambiguity parameters with high reliability are prioritized for fixation; a step-by-step fixation method is adopted to gradually expand the set of fixed ambiguities; multi-system fusion is used to enhance ambiguity fixation: multi-frequency observations are used to form ultra-wide lane and wide lane combinations to fix ambiguities step by step; observation data from multiple satellite systems are fused to increase observation redundancy and improve the fixation success rate; real-time quality control is implemented in the ambiguity resolution process: the search space size and ratio test threshold are dynamically adjusted; the ambiguity fixation success rate and fixation time are recorded; and the confidence index and reliability assessment results of ambiguity fixation are output.

[0059] This improved algorithm effectively enhances the efficiency and reliability of ambiguity fixation in complex environments through the synergy of multiple strategies, providing a guarantee for high-precision baseline solving.

[0060] S4.4 Baseline Vector Co-calculation and Quality Control. Using a fixed integer ambiguity, the baseline vector is solved using the least squares method, and its covariance matrix is ​​calculated. Residual analysis is used to verify the quality of the observations, and variance component estimation is used to adaptively adjust the observation weight matrix, outputting the final baseline vector and its accuracy information.

[0061] Furthermore, the baseline vector collaborative solution and quality control in step S4.4 adopts a multi-source information fusion processing strategy, specifically including: constructing a baseline solution model based on fixed ambiguity: substituting the successfully fixed integer ambiguity as known values ​​into the double-difference observation equation, using the least squares adjustment method to solve for the three coordinate components of the baseline vector, and calculating its complete variance-covariance matrix; implementing residual analysis and data quality control: by analyzing the residual distribution characteristics of the observed values, using standardized residual tests and gross error detection methods to identify abnormal observed values, and reducing or eliminating out-of-limit observed values; using variance components to estimate the adaptive adjustment weight matrix: based on posterior residual information, using Helmert variance analysis... The estimation method adaptively adjusts the weight matrix of observations from different satellite systems and observation periods to optimize the stochastic model; it performs accuracy assessment and uncertainty analysis: calculating the point accuracy, relative length accuracy, and azimuth accuracy of the baseline vector, and analyzing the internal and external conformity accuracy indices of the baseline solution; it outputs complete baseline solution results: including baseline vector estimates, variance-covariance matrix, accuracy assessment indices, ambiguity fixation status, and data quality report, providing reliable observation input for subsequent network adjustment; and it establishes a solution result verification mechanism: cross-validating the baseline solution results through methods such as independent baseline closure loop testing and repeated observation baseline comparison to ensure the correctness and reliability of the solution results.

[0062] The collaborative solution process adopts an iterative approach, and through the collaborative optimization of multiple stages such as observation value screening, weight matrix adjustment and solution verification, it ensures that the final output baseline vector results meet the accuracy requirements of high-precision deformation monitoring.

[0063] S4.5 Optimize network geometric strength. By analyzing the covariance matrix of the baseline vectors, the strength of the network's geometric structure is evaluated. Baselines with weaker geometric strength are weighted to ensure the stability of subsequent adjustment calculations.

[0064] The network geometry strength optimization in step S4.5 adopts a comprehensive evaluation method based on accuracy factors and reliability indicators, specifically including: Network geometric strength quantification assessment: Based on the variance-covariance matrix of baseline vectors, the relative accuracy of length, azimuth accuracy, and elevation accuracy of each baseline are calculated; the strength of the network geometry is quantified using a series of precision decay factor (DOP) indices, including planar precision factor (HDOP), elevation precision factor (VDOP), and spatial precision factor (PDOP); the sensitivity coefficient and reliability index of the baselines are calculated to evaluate the contribution of each baseline to the overall accuracy of the network.

[0065] Identification and processing of geometrically weak baselines: Set an accuracy threshold to identify baselines with weak geometric strength (such as baselines with a length relative accuracy lower than the preset value or a DOP value that is too high); reduce the weight of weak baselines and dynamically adjust their weight in subsequent adjustment according to their accuracy index; for particularly weak baselines, use additional parameters or constraints to enhance their stability.

[0066] Network structure optimization and adjustment: Analyze the weak links in the network and propose optimized deployment schemes for base stations or monitoring points; evaluate the impact of different network shapes on overall accuracy through simulation calculations and select the optimal network configuration; perform virtual optimization on the existing network and propose improvement suggestions.

[0067] Weighting strategy implementation: A weighting matrix is ​​constructed based on baseline accuracy indicators, and a comprehensive strategy of elevation angle dependence and signal-to-noise ratio weighting is adopted; an adaptive weighting method is adopted for different observation periods and observation conditions; iterative weighting processing is implemented to gradually optimize the weight allocation.

[0068] Stability assurance measures: Establish a network stability monitoring mechanism to track changes in network geometry in real time; set network strength early warning thresholds to issue alarms when network geometry strength falls below requirements; provide network optimization suggestions to ensure the long-term stability of the monitored network.

[0069] The network geometric strength optimization process adopts a quantitative evaluation and dynamic adjustment strategy. Through multiple steps such as accuracy factor analysis, reliability assessment and weighted processing, it significantly improves the overall geometric strength of the network, providing a reliable observation basis for subsequent high-precision adjustment calculations.

[0070] S4.6 Data Quality Feedback and Iterative Optimization. Establish a data quality feedback mechanism to feed back the quality indicators of the baseline solution to the front-end data preprocessing stage, and reprocess or remove data segments that do not meet the quality standards, thereby achieving closed-loop optimization of the data processing flow.

[0071] The baseline solution process adopts a multi-stage collaborative processing approach. Through quality control of multiple links such as observation quality assessment, ambiguity fixation reliability verification, and network geometric strength analysis, the baseline solution results are ensured to meet the requirements of high-precision deformation monitoring.

[0072] S5. The baseline vectors are adjusted using the joint network adjustment model, and the optimal coordinate estimates and their accuracy information for the monitoring points are obtained using a robust estimation method. The joint network adjustment model combined with the robust estimation method can effectively handle outliers and systematic errors, improving the reliability of the monitoring point coordinate estimates.

[0073] Specifically, in this embodiment, step S5 includes the following processing procedures: Using all baseline vectors and their covariance matrices as observations, monitoring point coordinates as parameters to be estimated, and base station coordinates as fixed or strong constraints, a unified error equation and constraint equation for the entire network are established. A weight matrix for the observations is constructed based on the covariance matrix provided by the baseline solution, and the weights of various observations are adaptively adjusted using a variance component estimation method. The least squares estimation method is used to solve the normal equations. When gross errors exist in the observations, a robust estimation method is switched to, and the influence of gross errors is controlled through weight functions such as IGGIII. The covariance matrix of the parameter estimates is calculated, and the point accuracy, relative accuracy, and error ellipse parameters are evaluated. The internal and external conformity accuracy of the network as a whole is analyzed. A comprehensive quality assessment of the adjustment results is conducted using various statistical tests, including network closure error test, parameter significance test, and residual distribution test. Based on the adjustment results, the observation weights, stochastic models, and estimation strategies are adjusted, and multiple iterations of adjustment are performed until the results converge.

[0074] The overall adjustment process adopts a modular processing flow, and a data quality feedback mechanism is established between each step to ensure the reliability and stability of the adjustment results. Finally, the optimal coordinate estimate of the monitoring point and its complete accuracy information are output.

[0075] S6. Deformation analysis is performed based on the adjusted coordinate time series to extract dam deformation information, including trend deformation, periodic deformation, and anomalous deformation. Based on the adjusted coordinate time series, deformation trends, periodic deformation, and anomalous deformation can be extracted. By analyzing the deformation trends, dam stability can be assessed.

[0076] Specifically, in this embodiment, step S6 includes the following processing steps: Based on the adjusted coordinate results of each monitoring point, construct the coordinate time series of each monitoring point at different time points; perform data smoothing on the time series, remove gross errors, and compensate for missing data; use linear regression, polynomial fitting, or Kalman filtering methods to extract the long-term deformation trend of the monitoring points, calculate the deformation rate and acceleration, and analyze the spatiotemporal distribution law of deformation; use spectrum analysis, wavelet analysis, or empirical mode decomposition methods to identify periodic components in the deformation (such as temperature cycle, water level cycle) and establish a periodic deformation model; based on statistical process control theory, use control charts, moving averages, or other anomaly detection algorithms to detect abnormal signals exceeding normal deformation in real time and locate the location and time period of the anomaly; combine environmental data (such as water level, temperature, and rainfall) to analyze the correlation between deformation and environmental factors, establish a deformation-environmental response model, and distinguish between deformation caused by load changes and structural deformation; based on the time series analysis results, use time series models or machine learning methods to predict short-term deformation and comprehensively evaluate the overall deformation stability of the dam.

[0077] The deformation analysis process employs a multi-method cross-validation mechanism to ensure the reliability and accuracy of deformation information extraction, ultimately generating a comprehensive deformation analysis result that includes deformation process lines, deformation distribution maps, anomaly alarms, and stability assessment reports.

[0078] S7. Output deformation monitoring results and provide visualization and early warning assessment. Monitoring results can be displayed through a visualization platform, and combined with an early warning mechanism, automatic alarms and emergency responses can be achieved. This system can provide real-time monitoring, early warning, and decision support.

[0079] Specifically, in this embodiment, step S7 includes the following processing procedures: Based on the adjusted coordinate time series, time series analysis is used to separate trend deformation, periodic deformation, and random deformation components, and to calculate the deformation rate, acceleration, and cumulative deformation at each monitoring point. Multi-dimensional deformation monitoring results are generated, including a monitoring point coordinate table, deformation process line graph, deformation contour map, and deformation vector map. The output format supports standard data exchange and manual interpretation. A comprehensive display platform integrating interactive charts and a dam 3D model is established, enabling dynamic playback, multi-period comparison, profile analysis, and spatial query of deformation data. Based on the dam's structural characteristics, design requirements, and historical data, a multi-level early warning indicator system is established, setting deformation thresholds for yellow, orange, and red warnings. Deformation data is monitored in real time; when the monitored value exceeds the warning threshold, an early warning signal is automatically triggered, and Bayesian network methods are used to assess the reliability of the early warning. Early warning reports containing the warning level, location, severity, and recommended measures are generated and disseminated through multiple channels, including audio-visual, SMS, and email, and corresponding emergency plans are activated.

[0080] The deformation monitoring results output and early warning system adopts a collaborative architecture of web and mobile terminals, supports remote real-time access and multi-user concurrent operation, and ensures the timeliness and availability of monitoring information.

[0081] Example 2

[0082] like Figure 2 As shown, this application provides a system architecture diagram for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks. It is applied to the system for improving the accuracy of GNSS deformation monitoring of dams through joint adjustment of base station and monitoring station networks as described in Embodiment 1. The system includes a network deployment module 11, a data acquisition and transmission module 12, a data preprocessing module 13, a joint network adjustment modeling module 14, a baseline calculation module 15, an overall adjustment processing module 16, a deformation analysis module 17, and a result output and early warning module 18.

[0083] The network deployment module 11 is used to configure and manage at least two base stations and multiple monitoring points deployed in and around the dam to form the physical topology of the GNSS monitoring network.

[0084] The data acquisition and transmission module 12 is used to acquire raw GNSS observation data from each base station and monitoring point in real time and transmit the data to the processing center.

[0085] The data preprocessing module 13 is used to perform cycle slip detection and repair, multipath effect identification and reduction, and outlier removal on the received raw GNSS observation data.

[0086] The joint network adjustment modeling module 14 is used to construct a joint network adjustment model of base stations and monitoring points. The model includes observation equations with station coordinates as parameters, where base station coordinates are known or strong constraints and monitoring point coordinates are parameters to be estimated.

[0087] The baseline solution module 15 is used to perform baseline solution based on double-difference carrier phase observations to obtain the baseline vector and its covariance matrix.

[0088] The overall adjustment processing module 16 is used to perform overall adjustment of the baseline vector using the joint network adjustment model, and to solve the optimal coordinate estimate and accuracy information of the monitoring point using the robust estimation method.

[0089] The deformation analysis module 17 is used to perform deformation analysis based on the adjusted coordinate time series, and extract dam deformation information, including trend deformation, periodic deformation and abnormal deformation.

[0090] The result output and early warning module 18 is used to output deformation monitoring results and perform visualization and early warning judgment.

[0091] Figure 3 This is an electronic device provided in one embodiment of this application. For example... Figure 3 As shown, the electronic device includes at least the following components: processor 101 and memory 100, communication interface 103, and bus 102.

[0092] In this embodiment of the application, memory 100 is used to store executable instructions of processor 101, which, when configured to execute instructions, implements the method as described in the first aspect.

[0093] In embodiments of this application, a computer-readable storage medium includes instructions that instruct a device to perform the method as described in the first aspect. For example, the instructions instruct the device to perform... Figure 1 The method is shown in the process steps.

[0094] In one embodiment of this application, the program operating in the electronic device may be a program that controls a central processing unit (CPU) or similar device to achieve the functions of the above-described embodiments of the present invention (a program that enables the computer to function). Information processed by these devices is then temporarily stored in random access memory (RAM) during processing, and subsequently stored in various ROMs such as read-only memory (FlashROM) and hard disk drives (HDDs), and read, corrected, and written by the CPU as needed.

[0095] It should be noted that a portion of the electronic device described above can also be implemented using a computer. In this case, the program for implementing the control function can be recorded on a computer-readable recording medium, and the program recorded on the recording medium can be read into the computer and executed.

[0096] It should be noted that the term "computer" as used here refers to a computer built into an electronic device, employing hardware including an operating system and peripheral devices. Furthermore, "computer-readable recording media" refers to removable media such as floppy disks, magneto-optical disks, ROMs, and CD-ROMs, as well as storage devices such as hard drives built into a computer.

[0097] Furthermore, a "computer-readable recording medium" can include: a medium that dynamically stores a program for a short period of time, such as a communication line used when transmitting a program via a network such as the Internet or a communication line such as a telephone line; or a medium that stores a program for a fixed period of time, such as volatile memory inside a computer that serves as a server or client in this case. In addition, the aforementioned program can be a program used to implement the above-mentioned functions, or it can be a program that can implement the above-mentioned functions by combining with programs already recorded in the computer.

[0098] Furthermore, the electronic device in the above embodiments can also be implemented as an assembly (device group) composed of multiple devices. Each device constituting the device group can possess some or all of the functions or functional blocks of the electronic device in the above embodiments. As a device group, it is sufficient to have all the functions or functional blocks of the electronic device.

[0099] Those skilled in the art should recognize that the above embodiments are only used to illustrate this application and are not intended to limit this application. Any appropriate changes and variations made to the above embodiments within the essential spirit and scope of this application fall within the scope of protection claimed in this application.

Claims

1. A method for improving the precision of dam GNSS deformation monitoring by joint network adjustment of base stations and stations, characterized in that, The method comprises: S1. Deploying at least two base stations and a plurality of monitoring points on the dam and its surrounding area to form a GNSS monitoring network; S2. Real-time acquisition of GNSS raw observation data of each base station and monitoring point, transmission of the data to a processing center, and data preprocessing, including cycle slip detection and repair, multi-path effect identification and weakening, and outlier rejection; S3. Constructing a joint network adjustment model of the base stations and monitoring points, the model comprising observation equations with site coordinates as parameters, wherein the base station coordinates are known or strong constraints, and the monitoring point coordinates are to be estimated parameters; S4. Baseline solution based on double-difference carrier phase observations to obtain baseline vectors and their covariance matrices; S5. Overall adjustment of the baseline vectors using the joint network adjustment model, and solving the optimal coordinate estimates and precision information of the monitoring points using a robust estimation method; S6. Deformation analysis based on the time series of the adjusted coordinates to extract dam deformation information, including trend deformation, periodic deformation, and abnormal deformation; S7. Output of the deformation monitoring results, and visual display and early warning judgment.

2. The method for improving the precision of dam GNSS deformation monitoring of a base station and survey station combined network adjustment according to claim 1, characterized in that, The deployment of base stations and monitoring points in step S1 adopts a hierarchical optimization configuration strategy, specifically including: A plurality of base stations are distributed in the stable geological area on both sides of the dam axis, forming a closed network between the base stations, and the distance between them is kept within a reasonable range, and each base station maintains line-of-sight conditions with a plurality of monitoring points; Monitoring points are evenly distributed along the dam axis direction, and are densely deployed at key positions to form a plurality of monitoring sections in the dam deformation sensitive area; The reference network layer is composed of all base stations, the monitoring network layer is composed of all monitoring points, and the redundant monitoring points are added at key positions in the encryption network layer; The network average edge length ratio is kept at a reasonable proportion, the network graph intensity factor reaches an excellent level, and the point accuracy decay factor is controlled at a low level.

3. The method of claim 1, wherein the method further comprises: Step S2 specifically includes the following processing procedures: Synchronize the acquisition of raw observation data of multiple systems and multiple frequencies through high-precision GNSS receivers deployed at base stations and monitoring points to ensure the unity of the data time reference and the high sampling rate of the observation values; Detect cycle slips using MW combination and GF combination, and accurately repair and mark cycle slips based on polynomial fitting method and ionospheric residual method; Construct a multi-path error empirical model using signal-to-noise ratio observation values, adjust observation value weights using a satellite elevation angle-dependent random model, and effectively suppress the adverse effects of multi-path effects; Based on robust statistical theory, identify and reject outliers using data detection method or weight function scheme, and further investigate abnormal observation values through residual analysis; Calculate quality indicators such as data integrity, multi-path error, and signal-to-noise ratio, standardize the observation data, and generate a data quality report; Real-time transmission of the observation data and the corresponding covariance matrix after cycle slip repair, multi-path weakening, and outlier rejection to the subsequent processing module.

4. The method of claim 1, wherein the method further comprises: In step S3, when constructing the joint network adjustment model, a regional network adjustment method based on connection point constraints is used, specifically including the following sub-steps: The GNSS observation data between each monitoring point and the base station and the connecting points obtained through image matching or measurement are used to establish a free network; a free network adjustment method is used for preliminary calculation to construct a robust network geometry; In the adjustment process, the affine transformation model is used to compensate for the systematic errors of GNSS observation values or the coordinates of the preliminary calculation, and the affine transformation parameters are used as the estimated parameters and the station coordinates are solved in the adjustment model; On the basis of the free network adjustment, known coordinate points uniformly distributed in the monitoring area are added as control points for control network adjustment; the correction numbers of the coordinates of each monitoring point are calculated through least squares estimation to accurately solve the coordinates of the monitoring points; The elevation values required in the adjustment process are obtained through digital elevation model interpolation; The point position errors of the connecting points after the free network adjustment are required to be better than the set threshold, and the point position errors after the control network adjustment meet the accuracy requirements of dam deformation monitoring.

5. The method for improving the precision of dam GNSS deformation monitoring of a base station and survey station combined network adjustment according to claim 4, characterized in that, On the basis of the affine transformation compensation for systematic errors, the control network adjustment further adopts a weighted mixed estimation method to optimize the stochastic model and balance the weights of the observation information and the prior information, and the specific method is as follows: An adaptive weighted estimation optimization model is constructed as follows: ， wherein, is a GNSS observation vector, is a priori constraint residual vector, A is an observation equation design matrix; H is a priori constraint equation design matrix; X is a vector of parameters to be estimated, including monitoring point coordinates and affine transformation parameters; , are weight matrices of observation values and a priori constraint values, respectively; is a weight factor, used to dynamically balance the contributions of observation information and a priori constraint information; optimal weight The optimal weight is determined by minimizing the trace of the mean square error matrix of the parameter estimates, i.e.: ， The mean square error matrix is calculated as follows: ， wherein are unit weight variance estimates of the observation and prior constraint values, respectively, dynamically determined by the following variance component estimation formula: ， ， wherein are the number of observations and the number of prior constraints, respectively, , , , denotes the trace of a matrix, i.e. the sum of the main diagonal elements of the matrix.

6. The method of claim 4, wherein the method further comprises: The step S4 specifically includes the following processing process: Fusion of multi-system observation data, selection of common-view satellites to form inter-station single difference and inter-satellite double difference, establishment of double-difference carrier phase observation equation with integer ambiguity and baseline vector as parameters, elimination of first-order ionospheric delay by using ionosphere-free combination, and unified processing of inter-frequency bias and hardware delay difference among systems; Combination of MW combination and Geometry-free combination is adopted to detect cycle slip, and polynomial fitting method is used for cycle slip repair, and data segmentation processing is performed on the cycle slip that cannot be repaired; LAMBDA algorithm is used to search for fixed double-difference integer ambiguity, and the reliability of ambiguity fixing is confirmed through ratio test, and partial fixing strategy is adopted for the ambiguity that fails to be fixed; Using the fixed integer ambiguity, the least squares method is used to solve the baseline vector and calculate the covariance matrix thereof; The quality of the observation values is tested through residual analysis, the observation weight matrix is adjusted through variance component estimation, and the final baseline vector and accuracy information thereof are output.

7. The method of claim 6, wherein the method further comprises: The step S5 specifically includes the following processing process: All baseline vectors and covariance matrices thereof are used as observation values, the coordinates of the monitoring points are used as estimated parameters, and the coordinates of the base stations are used as fixed or strong constraint conditions to establish error equations and constraint condition equations of the whole network; The weight matrix of the observation values is constructed based on the covariance matrix provided by the baseline solution, and the variance component estimation method is used to adaptively adjust the weights of various observation values; The least squares estimation method is used to solve the equations, and when there is a gross error in the observation values, the robust estimation method is switched to, and the gross error influence is controlled through the weight function; The covariance matrix of the parameter estimates is calculated to evaluate the point position accuracy, relative accuracy and error ellipse parameters; Statistical test methods such as network closure error test, parameter significance test and residual distribution test are used to comprehensively evaluate the quality of the adjustment results. According to the adjustment result feedback, the observation weight, the random model and the estimation strategy are adjusted, and multiple iteration adjustment is performed until the result converges.

8. The method of claim 6, wherein the method further comprises: The step S7 specifically includes the following processing process: Based on the adjusted coordinate time series, the time series analysis method is used to separate the trend deformation, the periodic deformation and the random deformation components, and the deformation rate, the acceleration and the cumulative deformation of each monitoring point are calculated; Multi-dimensional deformation monitoring results including monitoring point coordinate result table, deformation hydrograph, deformation contour map and deformation vector map are generated; A comprehensive display platform integrating interactive charts and three-dimensional dam models is established to realize dynamic playback, multi-period comparison and spatial query of deformation data; According to the dam structure characteristics and design requirements, a multi-level early warning index system is established, and deformation threshold values of different levels are set; Real-time monitoring of deformation data, when the monitoring value exceeds the early warning threshold, an early warning signal is automatically triggered, and early warning reliability evaluation is performed; An early warning report containing early warning level, position and recommended measures is generated, and early warning information is released through multiple channels.

9. A system for improving the precision of GNSS deformation monitoring of a dam by joint network adjustment of base stations and survey stations, applied to the method for improving the precision of GNSS deformation monitoring of a dam by joint network adjustment of base stations and survey stations according to any one of claims 1 to 8, characterized in that, The system comprises: A network layout module for configuring and managing at least two base stations and a plurality of monitoring points laid in the dam and its surrounding area to form a physical topology of the GNSS monitoring network; A data acquisition and transmission module for acquiring GNSS raw observation data of each base station and monitoring point in real time and transmitting the data to the processing center; A data preprocessing module for performing cycle slip detection and repair, multi-path effect identification and weakening, and outlier rejection processing on the received GNSS raw observation data; A joint network adjustment modeling module for constructing a joint network adjustment model of the base stations and the monitoring points, the model including observation equations with site coordinates as parameters, wherein the base station coordinates are known or strong constraint conditions, and the monitoring point coordinates are to-be-estimated parameters; A baseline solution module for performing baseline solution based on double-difference carrier phase observations to obtain baseline vectors and their covariance matrices; An overall adjustment processing module for performing overall adjustment on the baseline vectors using the joint network adjustment model, and solving the optimal coordinate estimates and their precision information of the monitoring points by using a robust estimation method; A deformation analysis module for performing deformation analysis based on the adjusted coordinate time series to extract dam deformation information, including trend deformation, periodic deformation and abnormal deformation; A result output and early warning module for outputting deformation monitoring results and performing visual display and early warning judgment.

10. A computer-readable storage medium, characterized in that, The instructions instruct the device to perform the method of claim 1 to 8. The instructions instruct the device to perform the method of claim 1 to 8.

Images