Water conservancy facility deformation monitoring and early warning method, system and device based on GNSS and medium

Abstract

The invention relates to a GNSS-based water conservancy facility deformation monitoring and early warning method, system and device and a medium, and the method comprises the steps: obtaining GNSS monitoring network data of a target part of a water conservancy facility, and carrying out the real-time dynamic differential processing, and obtaining corresponding three-dimensional coordinate data; acquiring environmental factor data, and performing multi-source feature fusion on the environmental factor data and the three-dimensional coordinate data to obtain a corresponding deformation feature vector; performing prediction analysis on the deformation feature vector according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result; setting a multi-level early warning threshold according to the deformation feature vector, and performing early warning analysis on a deformation trend prediction result to obtain a corresponding initial early warning level; and obtaining standard deformation characteristic information of the deformation rate and the acceleration of the water conservancy facility, and performing comprehensive evaluation on the standard deformation characteristic information based on a deformation trend prediction result to obtain a corresponding early warning index. Potential risks can be found and processed in time, and the possibility of hydraulic engineering accidents is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy facility monitoring, and in particular to a GNSS-based water conservancy facility deformation monitoring and early warning method, system, device and medium. Background Art

[0002] As global climate change and the impact of human activities become increasingly severe, water conservancy facilities are facing increasingly complex environmental challenges, and their deformation monitoring and early warning have become a key link in ensuring the safe operation of water conservancy projects. Traditional methods for monitoring deformation of water conservancy facilities mainly rely on regular manual inspections and single sensor data collection. This method has problems such as low monitoring frequency, limited data accuracy, and inability to reflect the status of facilities in real time. With the rapid development of global satellite navigation system (GNSS) technology, GNSS has shown great potential in the field of high-precision positioning and deformation monitoring. However, most existing methods currently only focus on the processing of GNSS data, ignoring the impact of environmental factors on deformation monitoring, which may lead to deviations in monitoring results. Summary of the invention

[0003] The main purpose of the present invention is to provide a GNSS-based water conservancy facility deformation monitoring and early warning method, system, device and medium, which can timely discover and deal with potential risks and reduce the possibility of water conservancy project accidents.

[0004] To achieve the above object, the present invention provides a GNSS-based water conservancy facility deformation monitoring and early warning method, comprising: Obtain GNSS monitoring network data of target parts of water conservancy facilities, perform real-time dynamic differential processing, and obtain corresponding three-dimensional coordinate data; Acquire environmental factor data, and perform multi-source feature fusion with the three-dimensional coordinate data to obtain a corresponding deformation feature vector; Predicting and analyzing the deformation feature vector according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result; Setting a multi-level warning threshold according to the deformation feature vector, and performing a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; The standard deformation characteristic information of the deformation rate and acceleration of the water conservancy facility is obtained, and the standard deformation characteristic information is comprehensively evaluated based on the deformation trend prediction result to obtain the corresponding early warning index.

[0005] Furthermore, the GNSS monitoring network data of the target part of the water conservancy facility is obtained, and real-time dynamic differential processing is performed to obtain the corresponding three-dimensional coordinate data, including: Acquire data from GNSS monitoring points located at target locations of the water conservancy facilities to obtain original observation data; Performing satellite precision calculation on the original observation data to obtain precise ephemeris data; Performing cycle slip detection and repair on the original observation data to obtain corrected observation data; Performing positioning and fixing analysis based on the corrected observation data and the precise ephemeris data to obtain a relative positioning solution set; Performing coordinate transformation and projection on the relative positioning solution set to obtain first coordinate data; Performing time series analysis on the first coordinate data to obtain corresponding displacement change trends; Performing outlier detection on the GNSS monitoring point according to the displacement change trend to obtain second coordinate data; The first coordinate data and the second coordinate data are subjected to precision fusion processing to obtain the three-dimensional coordinate data.

[0006] Furthermore, the environmental factor data is obtained and multi-source feature fusion is performed with the three-dimensional coordinate data to obtain a corresponding deformation feature vector, including: Collecting meteorological data of the water conservancy facilities to obtain a set of environmental parameters including temperature, humidity, air pressure and rainfall; Performing coordinate time series analysis on the three-dimensional coordinate data to obtain a coordinate change trend; Calculating the correlation coefficient matrix of the environmental parameters and the coordinate change trends to obtain the influence weight of the environmental factors on the coordinate change; Compensating the three-dimensional coordinate data for environmental factors according to the influence weights to obtain compensated coordinate data; Performing singular value decomposition on the compensation coordinate data to obtain target deformation direction amplitude data; Performing structural feature analysis on the three-dimensional coordinate data to obtain corresponding structural feature information; A vector is constructed based on the structural feature information in combination with the target deformation direction and amplitude data to obtain a deformation feature vector, wherein the deformation feature vector includes the target deformation direction, deformation amplitude, deformation rate and deformation acceleration.

[0007] Furthermore, the deformation feature vector is predicted and analyzed according to the preset whale optimization algorithm to obtain the corresponding deformation trend prediction result, including: Initializing the deformation feature vector according to preset whale optimization parameters to obtain an initial whale population; Iteratively optimize the initial whale population, including: Calculating an updated encirclement position value in the encirclement prey stage according to the time series characteristics of the deformation feature vector; Based on the spatial distribution characteristics of the deformation feature vector, the bubble position update value in the bubble net predation stage is calculated; Calculating a search position update value in the prey search phase according to the historical change trend of the deformation feature vector; Comprehensively processing the enclosing position update value, the bubble position update value and the search position update value, and combining the weight of the deformation trend prediction result to obtain an individual new position; Performing boundary iterative optimization processing on the initial whale population according to the new position of the individual to obtain an optimized deformation feature vector; Constructing a function based on the optimized deformation feature vector to obtain a deformation trend prediction function; Based on the deformation trend prediction function, a deformation trend analysis is performed to obtain a preliminary prediction result; The historical deformation data is obtained, and the upper and lower limits of the preliminary prediction result are calibrated to obtain the deformation trend prediction result.

[0008] Furthermore, the multi-level warning threshold is set according to the deformation feature vector, and the deformation trend prediction result is subjected to warning analysis to obtain the corresponding initial warning level, including: Calculating the deformation energy distribution of the deformation feature vector to obtain an energy concentration index; Extracting factor indicators from the environmental factor data to obtain an environmental factor impact index; Adaptively segmenting the energy concentration index to obtain an initial warning threshold interval; The initial warning threshold interval is optimized according to the preset warning conditions to obtain four levels of warning thresholds, including a caution level threshold, a warning level threshold, a danger level threshold and an emergency level threshold; Matching the deformation trend prediction result with the four-level warning threshold to obtain a preliminary level; Conducting a persistence analysis on the preliminary grade to obtain a time grade coefficient; The preliminary level is analyzed for early warning based on the time level coefficient and the environmental factor impact index to obtain a corresponding initial early warning level.

[0009] Furthermore, the standard deformation characteristic information of the deformation rate and acceleration of the water conservancy facility is obtained, and the standard deformation characteristic information is comprehensively evaluated based on the deformation trend prediction result to obtain the corresponding early warning index, including: Performing adaptive piecewise linear fitting on the standard deformation feature information to obtain corresponding piecewise features; Performing Hilbert-Huang transform processing on the segmented features to obtain corresponding instantaneous frequency and amplitude information; Performing nonlinear and non-stationary characteristic analysis on the standard deformation characteristic information according to the instantaneous frequency and the amplitude information to obtain corresponding independent characteristic components; The deformation trend prediction result is regressed and evaluated according to the independent characteristic components to obtain the early warning indicator.

[0010] Furthermore, the method further comprises: Performing a warning score on the initial warning level according to the warning indicators to obtain a corresponding comprehensive warning score; Applying the quantile analysis method according to the comprehensive warning score to obtain the corresponding comprehensive warning index; Activating the comprehensive warning index according to a preset piecewise activation function to obtain a corresponding warning activation value; Performing trigger condition analysis on the warning activation value to obtain corresponding warning trigger conditions; Based on the warning triggering condition and the warning activation value, rule division is performed to obtain corresponding warning decision rules; The warning triggering condition and the warning decision rule are integrated to obtain the warning information push strategy.

[0011] The present invention also provides a GNSS-based water conservancy facility deformation monitoring and early warning system, characterized in that the GNSS-based water conservancy facility deformation monitoring and early warning method applied to any one of the above-mentioned methods comprises: A collection module, which is used to obtain GNSS monitoring network data of target parts of water conservancy facilities, perform real-time dynamic differential processing, and obtain corresponding three-dimensional coordinate data; An analysis module, the analysis module is used to obtain environmental factor data, perform multi-source feature fusion with the three-dimensional coordinate data, and obtain a corresponding deformation feature vector; An association module, wherein the association module is used to perform prediction analysis on the deformation feature vector according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result; A processing module, the processing module is used to set a multi-level warning threshold according to the deformation feature vector, and perform a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; A control module is used to obtain standard deformation characteristic information of deformation rate and acceleration of the water conservancy facility, and to comprehensively evaluate the standard deformation characteristic information based on the deformation trend prediction result to obtain corresponding early warning indicators.

[0012] The present invention also provides a GNSS-based water conservancy facility deformation monitoring and early warning device, comprising: Memory, used to store programs; The processor is used to execute the program to implement each step of the GNSS-based water conservancy facility deformation monitoring and early warning method described in any one of the above.

[0013] The present invention also provides a medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute any of the above methods.

[0014] The present invention provides a GNSS-based water conservancy facility deformation monitoring and early warning method, system, device and medium, which have the following beneficial effects: By processing GNSS monitoring network data in real-time dynamic differential and fusing multi-source features with environmental factor data, the deformation state of water conservancy facilities can be evaluated more comprehensively and accurately, the accuracy and reliability of deformation monitoring can be improved, and a more reliable data basis can be provided for the safety management of water conservancy facilities. The deformation feature vector is predicted and analyzed by the whale optimization algorithm, which realizes the accurate prediction of complex deformation trends, overcomes the limitation of traditional prediction methods that it is difficult to capture complex deformation trends, and provides strong support for the timely discovery of potential risks. By setting multi-level warning thresholds and combining the deformation trend prediction results for early warning analysis, the deformation state of water conservancy facilities can be finely managed, the probability of false alarms and missed alarms can be effectively reduced, and the reliability and practicality of the early warning system can be improved. Based on the standard deformation feature information of deformation rate and acceleration, the deformation trend prediction results are comprehensively evaluated to obtain more scientific early warning indicators, making the early warning mechanism more comprehensive and accurate, and able to adapt to the monitoring needs of different types of water conservancy facilities. By comprehensively analyzing the initial warning level and warning indicators, an intelligent early warning information push strategy is formulated, which realizes the accurate push of warning information, improves the timeliness and pertinence of warning information, and provides more effective decision support for water conservancy facility managers. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a GNSS-based water conservancy facility deformation monitoring and early warning method provided by the present invention; Figure 2 This is a structural diagram of a GNSS-based water conservancy facility deformation monitoring and early warning system provided by the present invention; Figure 3 This is a structural diagram of a GNSS-based water conservancy facility deformation monitoring and early warning device provided by the present invention.

[0016] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0019] Reference Figure 1 As shown, the present invention provides a GNSS-based water conservancy facility deformation monitoring and early warning method, comprising: Step S1: Acquire GNSS monitoring network data of the target part of the water conservancy facility, perform real-time dynamic differential processing, and obtain corresponding three-dimensional coordinate data; Step S2: Acquire environmental factor data, perform multi-source feature fusion with the three-dimensional coordinate data, and obtain the corresponding deformation feature vector; Step S3: predict and analyze the deformation feature vector according to the preset whale optimization algorithm to obtain the corresponding deformation trend prediction result; Step S4: setting a multi-level warning threshold according to the deformation feature vector, and performing a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; Step S5: Obtain standard deformation characteristic information of deformation rate and acceleration of the water conservancy facility, conduct a comprehensive evaluation on the standard deformation characteristic information based on the deformation trend prediction result, and obtain corresponding early warning indicators.

[0020] Based on the above steps, the detailed process is as follows: Step S1: Select appropriate monitoring points at the target locations of the water conservancy facilities to ensure coverage of the main structures and potential deformation areas. The target locations may include the dam crest, dam body, spillway, etc.

[0021] Set the sampling rate of the GNSS receiver, usually 1Hz or higher, to ensure sufficient time resolution. A higher sampling rate helps capture subtle changes and dynamic characteristics of deformation. Collect raw observation data for 24 hours continuously, including pseudorange, carrier phase and other information. All-weather data collection helps to fully understand the deformation characteristics.

[0022] The observation data of the base station network is collected synchronously to ensure the time consistency of the base station and monitoring point data. The raw observation data is quality checked to remove outliers and cycle slips. The quality check uses statistical methods or filtering techniques to improve data reliability. Multipath effect and ionospheric delay correction are performed, which is achieved by using dual-frequency signals and model correction.

[0023] Apply precise ephemeris and clock data to improve positioning accuracy. Precise ephemeris and clock data can be obtained from organizations such as the International GNSS Service (IGS). Use double-difference or triple-difference algorithms for baseline solution. The double-difference algorithm improves relative positioning accuracy by eliminating common errors, and the triple-difference algorithm further eliminates asynchronous errors. Use Kalman filtering technology for dynamic positioning solution, which can update and optimize positioning results in real time. Solve the three-dimensional coordinates of the monitoring point in real time to generate high-precision location information. Convert GNSS coordinate systems (such as WGS84) to local coordinate systems for engineering applications and analysis. Calculate coordinate accuracy indicators, such as PDOP values ​​and standard deviations, to evaluate the accuracy of positioning results.

[0024] Step S2: Collect data such as temperature, humidity, air pressure, and rainfall. These meteorological data help to understand the impact of environmental conditions on the deformation of water conservancy facilities. Water level changes are recorded through water level monitoring equipment. Water level changes are an important factor affecting the deformation of water conservancy facilities such as dams. Additional deformation information can be obtained through auxiliary monitoring equipment such as inclinometers and strain gauges to supplement the deficiencies of GNSS data.

[0025] The environmental factor data is synchronized with the GNSS observation data to ensure that all data have the same time reference. Outlier detection and smoothing are performed on the environmental data to remove outliers and reduce the impact of noise.

[0026] Calculate the correlation coefficient between GNSS coordinate changes and environmental factors to identify environmental factors that have a significant impact on deformation. Apply principal component analysis (PCA) to reduce data dimensions, extract main features, and reduce data redundancy. Construct a weight matrix of the impact of environmental factors on deformation to quantify the impact of each environmental factor.

[0027] Extract displacement, velocity, acceleration and other features from GNSS coordinate data to describe the dynamic characteristics of deformation. Extract periodicity and trend features from environmental data to reflect the changing laws of environmental factors. Combine GNSS and environmental data to construct time series features and comprehensively describe the deformation process.

[0028] The Dempster-Shafer evidence theory is used for feature-level fusion, which integrates data from different sources to improve the comprehensiveness of features. The adaptive weight algorithm is used to adjust the importance of different features, and the weights are dynamically adjusted according to the contribution of the features. A comprehensive feature vector containing GNSS deformation information and environmental impact is generated to fully reflect the deformation state and its influencing factors.

[0029] Step S3: Set the population size, maximum number of iterations, and search space range to ensure that the algorithm runs within reasonable computing resources. Randomly generate an initial whale population, each individual represents a set of potential prediction parameters. Define a fitness function to evaluate the prediction effect, usually the root mean square error (RMSE) or mean absolute error (MAE).

[0030] Calculate the distance between the current optimal solution and each individual whale. Update the position of the whale, simulate the behavior of surrounding the prey, and gradually approach the optimal solution. Apply the shrinking and surrounding mechanism to gradually narrow the search range and improve the convergence speed.

[0031] Generate spiral update positions to simulate the spiral swimming of whales and improve global search capabilities. Randomly select between the shrinking and spiraling models to increase diversity. Update the position information of individual whales and explore new solution spaces.

[0032] Randomly select whales for global search to avoid falling into local optimality. Update the position of the selected individuals to further expand the search range. Balance local development and global exploration to ensure the stability and robustness of the algorithm.

[0033] Based on the optimal parameters obtained by the whale optimization algorithm, the deformation feature vector is predicted and analyzed. The historical deformation feature vector is used for training to verify the performance of the prediction model. The model performance is evaluated through cross-validation, and the parameters are fine-tuned to ensure the accuracy of the prediction results.

[0034] The optimized prediction model is used to predict deformation trends at future time points. The confidence interval of the prediction results is calculated to quantify the uncertainty of the prediction. The short-term (24 hours), medium-term (7 days) and long-term (30 days) deformation trend prediction results are generated to provide prediction information at multiple time scales.

[0035] Step S4: Set multiple warning levels (such as normal, concern, warning, and danger) based on historical deformation data and engineering experience. Determine the threshold range for each warning level, which is usually set based on indicators such as deformation amplitude, rate, and acceleration. Larger deformation amplitudes or rates correspond to higher levels of warning. Consider the impact of environmental factors on deformation and adjust the threshold to adapt to different environmental conditions. For example, extreme weather may cause increased deformation, and the warning threshold should be appropriately lowered.

[0036] According to the set threshold, the deformation feature vectors acquired in real time are classified to determine the warning level to which they belong. The threshold and classification model are updated regularly to ensure their adaptability and flexibility.

[0037] Analyze the deformation trend prediction results to determine the changing trend of the deformation feature vector in the future. Combine the real-time monitoring data and prediction results to conduct a comprehensive assessment to identify potential dangers and abnormal situations. Generate an initial warning report, including the current and predicted deformation status and its corresponding warning level.

[0038] Adjust and verify the initial warning level according to different engineering safety requirements and environmental conditions. Consider factors such as public safety and facility importance, and appropriately increase or decrease the warning level. Compare with historical warning records to verify the rationality and accuracy of the current warning level.

[0039] Step S5: Collect and organize the design specifications and safety standards of water conservancy facilities, and determine the standard thresholds of deformation rate and acceleration. Extract the deformation rate and acceleration under normal operating conditions from historical monitoring data as benchmark features. Define the limit value of abnormal deformation and clarify the corresponding measures to be taken when it exceeds the standard range.

[0040] The deformation rate and acceleration are calculated using the obtained three-dimensional coordinate data. The time derivative of the displacement is calculated using the differential method to obtain the rate, and the time derivative of the rate is further calculated to obtain the acceleration. Filtering technology is applied to reduce measurement noise and improve the accuracy of the calculation results.

[0041] Compare the deformation trend prediction results with the standard deformation feature information, identify potential deviations and anomalies, conduct a comprehensive assessment, and quantify the severity and possibility of deviations. Combine environmental factors and facility operation status to conduct a multi-dimensional comprehensive analysis.

[0042] Based on the comprehensive assessment results, detailed early warning indicators are generated, including deformation amplitude, rate, acceleration and its changing trend. The confidence interval of each early warning indicator is determined to quantify uncertainty and risk. Combined with multi-level early warning thresholds, the final early warning level and response measures are determined.

[0043] Compare early warning indicators with historical early warning records to evaluate the accuracy and reliability of current early warning analysis. Collect feedback information, optimize prediction models and early warning algorithms, and improve the intelligence level of the system.

[0044] The present invention provides a GNSS-based water conservancy facility deformation monitoring and early warning method. By processing GNSS monitoring network data in real-time dynamic differential and fusing multi-source features with environmental factor data, the deformation state of water conservancy facilities can be evaluated more comprehensively and accurately, the accuracy and reliability of deformation monitoring can be improved, and a more reliable data basis can be provided for the safety management of water conservancy facilities. The deformation feature vector is predicted and analyzed by the whale optimization algorithm, and the accurate prediction of complex deformation trends can be achieved, which overcomes the limitation that traditional prediction methods are difficult to capture complex deformation trends, and provides strong support for timely discovery of potential risks. By setting multi-level early warning thresholds and combining early warning analysis with deformation trend prediction results, the refined management of the deformation state of water conservancy facilities is achieved, the probability of false alarms and missed alarms is effectively reduced, and the reliability and practicality of the early warning system are improved. Based on the standard deformation feature information of deformation rate and acceleration, the deformation trend prediction results are comprehensively evaluated to obtain more scientific early warning indicators, making the early warning mechanism more comprehensive and accurate, and able to adapt to the monitoring needs of different types of water conservancy facilities. Through comprehensive analysis of the initial warning level and warning indicators, an intelligent warning information push strategy was formulated, which achieved accurate push of warning information, improved the timeliness and pertinence of warning information, and provided more effective decision-making support for water conservancy facility managers.

[0045] In one embodiment, GNSS monitoring network data of the target part of the water conservancy facility is obtained, and real-time dynamic differential processing is performed to obtain corresponding three-dimensional coordinate data, including: The data acquisition process of the GNSS monitoring network begins with the GNSS monitoring points deployed at the target locations of the water conservancy facilities. These monitoring points continuously collect satellite signals through high-precision GNSS receivers and record raw observation data containing pseudorange, carrier phase and other information. The collection frequency of raw observation data is usually set to once every 1 second or 30 seconds to ensure the continuity and integrity of the data.

[0046] After obtaining the original observation data, it is subjected to satellite precision calculation. This step uses the precise ephemeris products provided by the International GNSS Service (IGS) and combines them with the satellite orbit information in the observation data to calculate more accurate satellite position and clock error data. The use of precise ephemeris data significantly improves the accuracy of subsequent positioning calculations, especially in long baseline observations.

[0047] Cycle slip detection and repair are performed on the original observation data. A cycle slip is a sudden jump of a whole cycle or multiple cycles in the carrier phase observation value, which can cause significant errors in the positioning results. Cycle slip detection uses a combination of multiple algorithms, such as the Melbourne-Wübbena combination, the ionospheric residual method, etc., to identify possible cycle slip points. Once a cycle slip is detected, the data is repaired through interpolation or other mathematical methods to obtain continuous and reliable corrected observation data.

[0048] Based on the corrected observation data and precise ephemeris data, the positioning fix analysis is performed. This process uses double or triple difference methods to eliminate common errors, and uses ambiguity fixation technology to solve the integer ambiguity to obtain a high-precision relative positioning solution set. In this process, a series of quality control parameters are set, such as the minimum number of satellites, the maximum PDOP value, the minimum elevation angle, etc., to ensure the reliability of the solution results.

[0049] The relative positioning solution set is converted into plane coordinates under local coordinates through coordinate transformation and projection processing to form the first coordinate data. The coordinate transformation usually adopts a seven-parameter or four-parameter model, and the projection selects a suitable projection method according to the geographical location of the monitoring area, such as Gauss-Kruger projection.

[0050] Perform time series analysis on the first coordinate data, and use methods such as wavelet analysis and Kalman filtering to extract displacement trends. This step aims to identify the long-term deformation trend of the monitoring point and eliminate the influence of short-term fluctuations and noise. According to the extracted displacement trend, combined with the preset thresholds (such as daily maximum displacement, cumulative displacement, etc.), perform outlier detection on the GNSS monitoring point. The detected outliers will be marked or eliminated, and the remaining data will constitute the second coordinate data.

[0051] The first coordinate data and the second coordinate data are accurately fused. The fusion process uses weighted averaging or Bayesian estimation methods to comprehensively consider the accuracy characteristics and reliability of the two sets of data to obtain the final three-dimensional coordinate data. The purpose of this step is to maximize the accuracy and reliability of the coordinate data while retaining the original observation information.

[0052] This embodiment significantly improves the quality and reliability of monitoring data by performing satellite precision calculation and cycle slip detection and repair on the original observation data, laying a solid foundation for subsequent analysis. The cycle slip detection method combining multiple algorithms is used to effectively identify and repair data anomalies, ensuring the continuity and accuracy of the observation data. The use of precise ephemeris data for positioning and fixation analysis has greatly improved the accuracy of relative positioning, especially in long baseline observations. The introduction of time series analysis and outlier detection helps to identify long-term deformation trends and eliminate the impact of short-term fluctuations, thereby improving the reliability of monitoring results. Finally, by accurately fusing the coordinate data of different processing stages and comprehensively utilizing multi-source information, the accuracy of the final three-dimensional coordinate data is further optimized. This multi-step, multi-algorithm combination method not only improves the accuracy and reliability of deformation monitoring of water conservancy facilities, but also enhances the system's adaptability to complex environments, providing strong support for the safety management of water conservancy projects.

[0053] In one embodiment, environmental factor data is obtained and multi-source feature fusion is performed with three-dimensional coordinate data to obtain a corresponding deformation feature vector, including: By placing weather stations around water conservancy facilities, a set of environmental parameters including temperature, humidity, air pressure and rainfall are collected in real time. GNSS receivers are used to continuously collect three-dimensional coordinate data of key points of water conservancy facilities. These three-dimensional coordinate data are subjected to time series analysis, and the trend of coordinate changes over time is obtained by fitting polynomials or using sliding windows. Coordinate time series analysis is the process of processing three-dimensional coordinate data collected by GNSS. This step uses mathematical methods such as polynomial fitting or sliding windows to model the changes of coordinates over time. 30 days are usually selected as an analysis period to balance short-term fluctuations and long-term trends. The analysis results include the rate of change and acceleration of the coordinates in the three directions of X, Y, and Z.

[0054] The correlation coefficient matrix between environmental parameters and coordinate change trends is calculated to determine the influence weights of various environmental factors on coordinate changes. Based on these influence weights, environmental factor compensation is performed on the original three-dimensional coordinate data to eliminate the errors caused by environmental changes and obtain more accurate compensated coordinate data.

[0055] Singular value decomposition (SVD) is performed on the compensated coordinate data to extract the main deformation direction and amplitude information. Singular value decomposition (SVD) is a mathematical method for processing the compensated coordinate data and extracting the main deformation characteristics. SVD decomposes the coordinate change matrix into left singular vectors, singular values, and right singular vectors. Among them, the left singular vector corresponding to the maximum singular value represents the main deformation direction, and the size of the singular value reflects the amplitude of the deformation.

[0056] Perform structural feature analysis on the three-dimensional coordinate data, such as calculating the principal curvature and normal vector, to obtain the structural feature information of the water conservancy facilities. Combine this structural feature information with the deformation direction amplitude data obtained through SVD to construct a comprehensive deformation feature vector The obtained deformation feature vector contains key information such as target deformation direction, deformation amplitude, deformation rate and deformation acceleration. Among them, the target deformation direction reflects the main deformation trend of the water conservancy facility, the deformation amplitude indicates the degree of deformation, and the deformation rate and acceleration reflect the speed and change trend of deformation respectively.

[0057] This embodiment uses a multi-source data fusion method to comprehensively consider GNSS measurement data, environmental factors and structural characteristics, thereby significantly improving the accuracy and reliability of deformation monitoring of water conservancy facilities. The environmental factor compensation mechanism effectively eliminates the interference of external conditions such as temperature and humidity on the measurement results, ensuring the accuracy of the monitoring data. The introduction of structural feature analysis makes deformation monitoring more in line with the actual situation of water conservancy facilities, and improves the pertinence and effectiveness of monitoring. The construction of deformation feature vectors includes key information such as the main deformation direction, amplitude, rate and acceleration, providing rich data support for a comprehensive assessment of the deformation state of water conservancy facilities. This comprehensive analysis method can not only detect potential risks in a timely manner, but also provide a scientific basis for the formulation of preventive measures, greatly improving the safety management level and early warning capabilities of water conservancy facilities. At the same time, the flexibility of this method makes it easy to adapt to water conservancy facilities of different types and sizes, and has broad application prospects.

[0058] In one embodiment, the deformation feature vector is predicted and analyzed according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result, including: The deformation feature vector is initialized according to the preset whale optimization parameters to generate the initial whale population. An initial solution space is created by mapping the deformation feature vector to the position and velocity of individual whales.

[0059] After obtaining the initial whale population, the method enters the iterative optimization phase. The iterative optimization process includes three main phases: encircling prey, bubble net predation, and searching for prey. In the encircling prey phase, the algorithm calculates the updated encirclement position value based on the time series characteristics of the deformation feature vector. This step is based on the change law of deformation data over time, which enables the algorithm to more accurately capture the temporal characteristics of the deformation trend.

[0060] The bubble net predation phase calculates the bubble position update value based on the spatial distribution characteristics of the deformation feature vector. This phase makes full use of the spatial distribution information of the deformation data, which helps the algorithm better understand the spatial pattern and correlation of the deformation. The search prey phase calculates the search position update value based on the historical change trend of the deformation feature vector, which enables the algorithm to learn and use past deformation patterns to predict future trends.

[0061] The calculated updated values ​​of the enclosing position, bubble position, and search position are processed comprehensively and combined with the weight of the deformation trend prediction results to finally obtain the new individual position. This step integrates the optimization results of multiple stages.

[0062] The initial whale population is iteratively optimized based on the new positions of the individuals to obtain the optimized deformation feature vector. This process ensures that the optimization result is always within the valid solution space, improving the reliability and accuracy of the prediction.

[0063] Based on the optimized deformation feature vector, a deformation trend prediction function is constructed. This function captures the complex nonlinear relationship contained in the deformation feature vector and provides a mathematical basis for subsequent deformation trend analysis. Using this prediction function, the method analyzes the deformation trend and obtains preliminary prediction results.

[0064] By obtaining historical deformation data and calibrating the upper and lower limits of the preliminary prediction results based on the historical deformation data, and by introducing historical data as a reference, the prediction results are reasonably constrained and adjusted to obtain more reliable deformation trend prediction results.

[0065] This embodiment uses the whale optimization algorithm to predict and analyze the deformation feature vector. This method can fully consider the time series characteristics, spatial distribution characteristics and historical change trends of the deformation data, thereby improving the accuracy and reliability of the deformation trend prediction. The multi-stage optimization process, including encircling prey, bubble net predation and searching for prey, enables the algorithm to fully capture the complex nonlinear relationship of the deformation data, effectively improving the adaptability and robustness of the prediction model. Through the comprehensive processing of the encirclement position update value, bubble position update value and search position update value, and combined with the weight of the deformation trend prediction result, a comprehensive consideration of multiple influencing factors is achieved, thereby obtaining a more accurate individual new position. The boundary iterative optimization process ensures that the optimization result is always within the effective solution space, further improving the reliability of the prediction. By introducing historical deformation data to calibrate the upper and lower limits of the preliminary prediction results, the historical experience is effectively combined with the prediction model, which significantly improves the practicality and credibility of the prediction results.

[0066] In one embodiment, a multi-level warning threshold is set according to the deformation feature vector, and a warning analysis is performed on the deformation trend prediction result to obtain a corresponding initial warning level, including: The deformation energy distribution of the deformation feature vector is calculated to obtain the energy concentration index, and the factor index of the environmental factor data is extracted to obtain the environmental factor impact index. The energy concentration index is adaptively segmented to obtain the initial warning threshold interval, and then the initial warning threshold interval is optimized according to the preset warning conditions to obtain four levels of warning thresholds, including attention level threshold, warning level threshold, danger level threshold and emergency level threshold.

[0067] The deformation trend prediction results are matched with the four-level warning threshold to obtain the preliminary level, and the preliminary level is subjected to a continuous analysis to obtain the time level coefficient. The preliminary level is subjected to a warning analysis based on the time level coefficient and the environmental factor impact index to obtain the corresponding initial warning level.

[0068] Among them, the deformation energy distribution calculation can use methods such as wavelet transform or Fourier transform, and the energy concentration index can be calculated by analyzing the energy distribution of the deformation feature vector at different frequencies or scales. The index extraction of environmental factor data can consider factors such as temperature, humidity, and rainfall, and obtain a comprehensive environmental factor impact index through normalization and weighted summation.

[0069] In the adaptive segmentation process, clustering algorithms or dynamic programming can be used to automatically determine appropriate segmentation points based on the distribution characteristics of energy concentration indicators, thereby obtaining the initial warning threshold range. Threshold optimization can be combined with historical data to fine-tune and optimize the initial threshold through fuzzy comprehensive evaluation, and finally obtain the four-level warning threshold.

[0070] The deformation trend prediction results are matched with the four-level warning thresholds by using a simple interval judgment method to correspond the interval into which the prediction results fall to the corresponding warning level. The persistence analysis calculates the time level coefficient by setting a time window, counting the frequency and duration of each warning level within a certain period of time.

[0071] During the early warning analysis phase, a weighted average is used to comprehensively consider the preliminary level, time level coefficient and environmental factor impact index to arrive at the final initial warning level.

[0072] This implementation calculates the deformation energy distribution of the deformation feature vector to obtain the energy concentration index, which can more comprehensively reflect the deformation state of the water conservancy facilities and improve the accuracy of monitoring. The environmental factor data is combined to extract indicators and obtain the environmental factor impact index, which effectively considers the impact of external conditions on deformation and enhances the reliability of early warning. The energy concentration index is processed using an adaptive segmentation method to obtain the initial warning threshold interval, and further optimized to obtain the four-level warning threshold, making the division of warning levels more scientific and reasonable, and more adaptable. By matching the deformation trend prediction results with the warning threshold and performing a continuous analysis to obtain the time level coefficient, the interference of short-term fluctuations on the warning results is effectively avoided. Finally, according to the time level coefficient and the environmental factor impact index In one embodiment, standard deformation characteristic information of deformation rate and acceleration of water conservancy facilities is obtained, and the standard deformation characteristic information is comprehensively evaluated based on the deformation trend prediction result to obtain corresponding early warning indicators, including: Adaptive piecewise linear fitting is performed on the standard deformation feature information to obtain the corresponding segment features. The adaptive piecewise linear fitting adopts a dynamic programming algorithm to determine the optimal segmentation point by minimizing the weighted sum of the fitting error and the number of segments.

[0073] After obtaining the segmented features, the Hilbert-Huang transform is performed on them to obtain the corresponding instantaneous frequency and amplitude information. The Hilbert-Huang transform is a time-frequency analysis method that can effectively extract the time-varying characteristics of non-stationary signals. The segmented features are decomposed into a series of intrinsic mode functions, and the Hilbert spectrum of each mode function is calculated to obtain the instantaneous frequency and amplitude information.

[0074] Based on the instantaneous frequency and amplitude information obtained, the standard deformation feature information is analyzed for nonlinear and non-stationary characteristics to obtain the corresponding independent characteristic components. This step uses the empirical mode decomposition method to decompose the complex deformation characteristics into a finite number of intrinsic mode functions. Each intrinsic mode function represents an independent characteristic component in the original signal, reflecting the deformation characteristics at different scales.

[0075] The deformation trend prediction results are regressed and evaluated based on the independent characteristic components to obtain the early warning indicators. The regression evaluation uses the support vector regression algorithm, takes the independent characteristic components as input variables, and the deformation trend prediction results as target variables to establish a nonlinear regression function. Through this function, the influence of the independent characteristic components on the deformation trend is quantified, so as to obtain the early warning indicators that comprehensively consider multiple factors.

[0076] This embodiment can effectively simplify complex nonlinear deformation characteristics and improve the efficiency and accuracy of data processing by adopting adaptive piecewise linear fitting technology. Combined with advanced signal processing technologies such as Hilbert-Huang transform and empirical mode decomposition, a refined analysis of the deformation characteristics of water conservancy facilities is achieved, and non-stationary and nonlinear deformation characteristics can be accurately captured. This multi-level analysis method significantly improves the accuracy and reliability of deformation feature extraction. By introducing the support vector regression algorithm for regression evaluation, the influence of multiple independent feature components on the deformation trend can be comprehensively considered, thereby obtaining a more comprehensive and accurate early warning indicator.

[0077] In one embodiment, a GNSS-based water conservancy facility deformation monitoring and early warning method further includes: The initial warning level is quantified in the early warning scoring stage. The initial warning level is scored according to the pre-set early warning indicators, and each indicator is assigned a corresponding weight. The scoring process considers multiple dimensions such as deformation degree, deformation rate, and environmental factors, and a comprehensive early warning score is obtained through weighted calculation. This step converts the qualitative early warning level into a quantitative value, laying the foundation for subsequent analysis.

[0078] The calculation of the comprehensive early warning index adopts the quantile analysis method. Collect historical monitoring data and establish a probability distribution model for early warning scores. Perform quantile analysis on the current comprehensive early warning score to determine its position in the historical data, thereby obtaining the corresponding comprehensive early warning index. This step takes into account the statistical characteristics of historical data, making the early warning results more objective and comparable.

[0079] The calculation of the warning activation value introduces a piecewise activation function. The comprehensive warning index is divided into multiple intervals according to the preset threshold, and each interval corresponds to a different activation function. The warning activation value is calculated by substituting the comprehensive warning index into the corresponding activation function. This step introduces nonlinear mapping, which enhances the sensitivity of the warning to different degrees of risk.

[0080] The trigger condition analysis phase conducts an in-depth evaluation of the warning activation value. Multiple trigger thresholds are set and the warning activation value is compared with these thresholds. Based on the comparison results, it is determined whether the warning trigger conditions are met and the specific type of trigger. This step clarifies the criteria for warning triggering and provides a basis for subsequent decision-making.

[0081] The formulation of early warning decision rules is based on early warning trigger conditions and early warning activation values. According to different trigger conditions and activation value ranges, corresponding decision rules are formulated. These rules cover aspects such as the determination of early warning levels, the selection of response measures, and resource allocation strategies. The design of decision rules takes into account the characteristics of water conservancy facilities and actual management needs to ensure the practicality of early warnings.

[0082] The formation of early warning information push strategy is the integration of early warning trigger conditions and decision rules. According to the urgency of the trigger conditions and the requirements of the decision rules, formulate the corresponding information push strategy. This includes the determination of the push object, the selection of the push method, the setting of the push frequency, etc. The design of the push strategy is to ensure that relevant personnel can receive early warning information in a timely and accurate manner and take corresponding actions.

[0083] This embodiment introduces a warning scoring mechanism to convert qualitative warning levels into quantitative comprehensive warning scores, thereby improving the objectivity and accuracy of the warning process. The quantile analysis method is used to calculate the comprehensive warning index, making full use of the statistical characteristics of historical data to make the warning results more comparable and valuable for reference. The introduction of a piecewise activation function to calculate the warning activation value enhances the system's sensitivity to risks of varying degrees and improves the accuracy of the warning. Through trigger condition analysis and decision rule formulation, the warning process is made intelligent and personalized, and more reasonable warning decisions can be made according to different situations. Finally, by integrating warning trigger conditions and decision rules, a targeted information push strategy is formed to ensure the timeliness and effectiveness of warning information.

[0084] Reference Figure 2 As shown, the present invention also provides a GNSS-based water conservancy facility deformation monitoring and early warning system, which is applied to any one of the above-mentioned GNSS-based water conservancy facility deformation monitoring and early warning methods, including: The acquisition module is used to obtain the GNSS monitoring network data of the target parts of the water conservancy facilities, perform real-time dynamic differential processing, and obtain the corresponding three-dimensional coordinate data; Analysis module, which is used to obtain environmental factor data, perform multi-source feature fusion with three-dimensional coordinate data, and obtain the corresponding deformation feature vector; The association module is used to predict and analyze the deformation feature vector according to the preset whale optimization algorithm to obtain the corresponding deformation trend prediction result; A processing module is used to set a multi-level warning threshold according to the deformation feature vector, and to perform a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; The control module is used to obtain the standard deformation characteristic information of the deformation rate and acceleration of the water conservancy facilities, conduct a comprehensive evaluation of the standard deformation characteristic information based on the deformation trend prediction results, and obtain the corresponding early warning indicators.

[0085] The present invention provides a GNSS-based water conservancy facility deformation monitoring and early warning system, which can more comprehensively and accurately evaluate the deformation state of water conservancy facilities by processing GNSS monitoring network data in real-time dynamic differential processing and fusing multi-source features with environmental factor data, thereby improving the accuracy and reliability of deformation monitoring and providing a more reliable data basis for the safety management of water conservancy facilities. The whale optimization algorithm is used to predict and analyze the deformation feature vector, which realizes the accurate prediction of complex deformation trends, overcomes the limitation that traditional prediction methods are difficult to capture complex deformation trends, and provides strong support for timely discovery of potential risks. By setting multi-level early warning thresholds and combining early warning analysis with deformation trend prediction results, the refined management of the deformation state of water conservancy facilities is realized, the probability of false alarms and missed alarms is effectively reduced, and the reliability and practicality of the early warning system are improved. Based on the standard deformation feature information of deformation rate and acceleration, the deformation trend prediction results are comprehensively evaluated to obtain more scientific early warning indicators, making the early warning mechanism more comprehensive and accurate, and able to adapt to the monitoring needs of different types of water conservancy facilities. Through comprehensive analysis of the initial warning level and warning indicators, an intelligent warning information push strategy was formulated, which achieved accurate push of warning information, improved the timeliness and pertinence of warning information, and provided more effective decision-making support for water conservancy facility managers.

[0086] Reference Figure 3 As shown, the present invention also provides a GNSS-based water conservancy facility deformation monitoring and early warning device, comprising: Memory, used to store programs; The processor is used to execute the program to implement each step of any one of the above-mentioned methods for monitoring and early warning of deformation of water conservancy facilities based on GNSS.

[0087] In this embodiment, the processor and the memory may be connected via a bus or other means. The memory may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk, or a solid-state drive. The processor may be a general-purpose processor, such as a central processing unit, a digital signal processor, an application-specific integrated circuit, or one or more integrated circuits configured to implement the embodiments of the present invention.

[0088] The present invention also provides a medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute any of the above methods.

[0089] It should be noted that technicians in the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process of the system and each module described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0090] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A GNSS-based water conservancy facility deformation monitoring and early warning method, characterized in that: include: Obtain GNSS monitoring network data of target parts of water conservancy facilities, perform real-time dynamic differential processing, and obtain corresponding three-dimensional coordinate data; Acquire environmental factor data, and perform multi-source feature fusion with the three-dimensional coordinate data to obtain a corresponding deformation feature vector; Predicting and analyzing the deformation feature vector according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result; Setting a multi-level warning threshold according to the deformation feature vector, and performing a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; The standard deformation characteristic information of the deformation rate and acceleration of the water conservancy facility is obtained, and the standard deformation characteristic information is comprehensively evaluated based on the deformation trend prediction result to obtain the corresponding early warning index.

2. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The method of obtaining GNSS monitoring network data of the target part of the water conservancy facility and performing real-time dynamic differential processing to obtain corresponding three-dimensional coordinate data includes: Acquire data from GNSS monitoring points located at target locations of the water conservancy facilities to obtain original observation data; Performing satellite precision calculation on the original observation data to obtain precise ephemeris data; Performing cycle slip detection and repair on the original observation data to obtain corrected observation data; Performing positioning and fixing analysis based on the corrected observation data and the precise ephemeris data to obtain a relative positioning solution set; Performing coordinate transformation and projection on the relative positioning solution set to obtain first coordinate data; Performing time series analysis on the first coordinate data to obtain corresponding displacement change trends; Performing outlier detection on the GNSS monitoring point according to the displacement change trend to obtain second coordinate data; The first coordinate data and the second coordinate data are subjected to precision fusion processing to obtain the three-dimensional coordinate data.

3. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The acquiring of the environmental factor data and the multi-source feature fusion of the three-dimensional coordinate data to obtain the corresponding deformation feature vector includes: Collecting meteorological data of the water conservancy facilities to obtain a set of environmental parameters including temperature, humidity, air pressure and rainfall; Performing coordinate time series analysis on the three-dimensional coordinate data to obtain a coordinate change trend; Calculating the correlation coefficient matrix of the environmental parameters and the coordinate change trends to obtain the influence weight of the environmental factors on the coordinate change; Compensating the three-dimensional coordinate data for environmental factors according to the influence weights to obtain compensated coordinate data; Performing singular value decomposition on the compensation coordinate data to obtain target deformation direction amplitude data; Performing structural feature analysis on the three-dimensional coordinate data to obtain corresponding structural feature information; A vector is constructed based on the structural feature information in combination with the target deformation direction and amplitude data to obtain a deformation feature vector, wherein the deformation feature vector includes the target deformation direction, deformation amplitude, deformation rate and deformation acceleration.

4. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The deformation feature vector is predicted and analyzed according to the preset whale optimization algorithm to obtain the corresponding deformation trend prediction result, including: Initializing the deformation feature vector according to preset whale optimization parameters to obtain an initial whale population; Iteratively optimize the initial whale population, including: Calculating an updated encirclement position value in the encirclement prey stage according to the time series characteristics of the deformation feature vector; Based on the spatial distribution characteristics of the deformation feature vector, the bubble position update value in the bubble net predation stage is calculated; Calculating a search position update value in the prey search phase according to the historical change trend of the deformation feature vector; Comprehensively processing the enclosing position update value, the bubble position update value and the search position update value, and combining the weight of the deformation trend prediction result to obtain an individual new position; Performing boundary iterative optimization processing on the initial whale population according to the new position of the individual to obtain an optimized deformation feature vector; Constructing a function based on the optimized deformation feature vector to obtain a deformation trend prediction function; Based on the deformation trend prediction function, a deformation trend analysis is performed to obtain a preliminary prediction result; The historical deformation data is obtained, and the upper and lower limits of the preliminary prediction result are calibrated to obtain the deformation trend prediction result.

5. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The multi-level warning threshold is set according to the deformation feature vector, and the deformation trend prediction result is subjected to warning analysis to obtain the corresponding initial warning level, including: Calculating the deformation energy distribution of the deformation feature vector to obtain an energy concentration index; Extracting factor indicators from the environmental factor data to obtain an environmental factor impact index; Adaptively segmenting the energy concentration index to obtain an initial warning threshold interval; The initial warning threshold interval is optimized according to the preset warning conditions to obtain four levels of warning thresholds, including a caution level threshold, a warning level threshold, a danger level threshold and an emergency level threshold; Matching the deformation trend prediction result with the four-level warning threshold to obtain a preliminary level; Conducting a persistence analysis on the preliminary grade to obtain a time grade coefficient; The preliminary level is analyzed for early warning based on the time level coefficient and the environmental factor impact index to obtain a corresponding initial early warning level.

6. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The step of obtaining the standard deformation characteristic information of the deformation rate and acceleration of the water conservancy facility and comprehensively evaluating the standard deformation characteristic information based on the deformation trend prediction result to obtain the corresponding early warning indicator includes: Performing adaptive piecewise linear fitting on the standard deformation feature information to obtain corresponding piecewise features; Performing Hilbert-Huang transform processing on the segmented features to obtain corresponding instantaneous frequency and amplitude information; Performing nonlinear and non-stationary characteristic analysis on the standard deformation characteristic information according to the instantaneous frequency and the amplitude information to obtain corresponding independent characteristic components; The deformation trend prediction result is regressed and evaluated according to the independent characteristic components to obtain the early warning indicator.

7. The GNSS-based water conservancy facility deformation monitoring and early warning method according to claim 1 is characterized in that: The method further comprises: Performing a warning score on the initial warning level according to the warning indicators to obtain a corresponding comprehensive warning score; Applying the quantile analysis method according to the comprehensive warning score to obtain the corresponding comprehensive warning index; Activating the comprehensive warning index according to a preset piecewise activation function to obtain a corresponding warning activation value; Performing trigger condition analysis on the warning activation value to obtain corresponding warning trigger conditions; Based on the warning triggering condition and the warning activation value, rule division is performed to obtain corresponding warning decision rules; The warning triggering condition and the warning decision rule are integrated to obtain the warning information push strategy.

8. A GNSS-based water conservancy facility deformation monitoring and early warning system, characterized in that: The GNSS-based water conservancy facility deformation monitoring and early warning method applied to any one of claims 1 to 7 above comprises: A collection module, which is used to obtain GNSS monitoring network data of target parts of water conservancy facilities, perform real-time dynamic differential processing, and obtain corresponding three-dimensional coordinate data; An analysis module, the analysis module is used to obtain environmental factor data, perform multi-source feature fusion with the three-dimensional coordinate data, and obtain a corresponding deformation feature vector; An association module, wherein the association module is used to perform prediction analysis on the deformation feature vector according to a preset whale optimization algorithm to obtain a corresponding deformation trend prediction result; A processing module, the processing module is used to set a multi-level warning threshold according to the deformation feature vector, and perform a warning analysis on the deformation trend prediction result to obtain a corresponding initial warning level; A control module is used to obtain standard deformation characteristic information of deformation rate and acceleration of the water conservancy facility, and to comprehensively evaluate the standard deformation characteristic information based on the deformation trend prediction result to obtain corresponding early warning indicators.

9. A GNSS-based water conservancy facility deformation monitoring and early warning device, characterized in that: include: Memory, used to store programs; The processor is used to execute the program to implement the various steps of the GNSS-based water conservancy facility deformation monitoring and early warning method as described in any one of claims 1 to 7.

10. A medium, characterized in that Computer instructions are stored, and the computer instructions are used to make a computer execute the method according to any one of claims 1 to 7.

Images