Dam deformation space-time prediction method and device based on Kriging interpolation

Abstract

The invention discloses a dam deformation space-time prediction method and device based on Kriging interpolation, and relates to the technical field of dam safety monitoring, and the method comprises the steps: carrying out the cleaning and unified processing of dam multi-source heterogeneous deformation monitoring data, including unifying a coordinate reference and carrying out the data fusion through a linear relation, forming a data set of a unified time vector; then, fitting data of each monitoring point and predicting future moments by using regression models such as Fourier series and the like; and after the consistency of the prediction result is checked, the kriging interpolation algorithm is adopted to carry out spatial interpolation on the predicted data passing the check, the space-time continuous deformation field of the dam is finally generated, and when newly added data exist, the process is repeated to update the model, the problems of space-time discretization and difficult fusion of the deformation data in the prior art are effectively solved, and the prediction accuracy of the dam is improved. High-precision and time-space continuous dam deformation prediction can be realized, and a reliable basis is provided for dam safety analysis.

Description

Technical Field

[0001] This invention relates to the field of dam safety monitoring technology, specifically to a method and device for spatiotemporal prediction of dam deformation based on Kriging interpolation. Background Technology

[0002] Dam deformation monitoring is crucial for ensuring its safe operation. Currently, it is mainly achieved by deploying various monitoring devices (such as leveling points, static leveling instruments, and tension wires) on the dam body to obtain deformation data at discrete points. To analyze deformation patterns, existing technologies typically employ two independent processing paths: one is to fit historical data of a single point based on regression analysis models (such as Gaussian functions and Fourier series) to achieve continuous prediction in the time domain; the other is to use spatial interpolation algorithms such as inverse distance weighted average (IDW) and Kriging interpolation to estimate the deformation value of unknown points, considering spatial correlation, in order to construct a spatially continuous field.

[0003] However, due to uneven distribution of monitoring points and diverse types of instruments leading to varying observation frequencies and accuracies, the acquired data is inherently discrete in the spatiotemporal domain. Existing technologies that separate time-series prediction and spatial interpolation cannot effectively integrate multi-source heterogeneous data. Their fundamental limitation lies in the failure to organically unify the processing of continuity in the temporal and spatial dimensions, resulting in the difficulty in obtaining a temporally continuous deformation field. This hinders the accurate analysis of the overall deformation patterns of the dam and the timely detection of safety hazards. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method and apparatus for spatiotemporal prediction of dam deformation based on Kriging interpolation, which can effectively integrate multi-source monitoring data and achieve continuous spatiotemporal prediction.

[0005] In a first aspect, embodiments of the present invention provide a spatiotemporal prediction method for dam deformation based on kriging interpolation, comprising: Deformation monitoring data from multiple monitoring points on the dam are acquired, and the monitoring data is cleaned to obtain a cleaned dataset. The cleaned monitoring data is processed in a unified manner, including unifying the coordinate reference and data fusion, to form a unified time vector and monitoring dataset; Based on the monitoring dataset, a regression analysis model was used to fit the deformation data of each monitoring point to obtain the fitting model. The deformation data of each monitoring point is predicted using the fitting model to obtain the predicted deformation data sequence of each monitoring point at future times; The predicted deformation data sequence is checked for consistency with the corresponding real monitoring data. Based on the predicted deformation data sequence that passes the check, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. When new monitoring data is added, repeat the above steps and update the model.

[0006] In some embodiments, cleaning the monitoring data includes: Coarse screening process: Manual judgment is performed based on environmental data, dam structure type, and data consistency of nearby or same location measuring points to eliminate gross errors; Cleaning process: The recent historical data is jointly judged using the trend method and the extreme value method to identify abnormal data points and form a cleaned dataset; wherein, the trend method is used to determine whether the data points deviate from the long-term trend, and the extreme value method is used to determine whether the data points exceed the reasonable range based on historical statistics.

[0007] In some embodiments, the data fusion includes: For multiple monitoring instruments deployed at the same location, the data sequence measured by the monitoring instrument with the highest accuracy shall be used as the benchmark; Establish a linear relationship between the data sequences measured by other monitoring instruments and the reference data sequence; The parameters of each linear relationship are obtained through regression analysis; The average of the baseline data and each transformed data is taken as the fused deformation data of that part at the corresponding time.

[0008] In some embodiments, the unified time vector is constructed by extracting unique time points from the original observation time series of all monitoring points and sorting them chronologically.

[0009] In some embodiments, the expression of the regression analysis model is:

[0010] In the formula, This is the amount of deformation. For time variables, , , w represents the model parameters, and H represents the number of harmonics.

[0011] In some embodiments, in the step of fitting the deformation data of each monitoring point using a regression analysis model to obtain a fitted model, the historical data of the monitoring points are fitted using the least squares method to determine the model parameters, and the optimal model is selected from models with different harmonic counts based on the root mean square error (RMSE). The formula for calculating the root mean square error (RMSE) is as follows:

[0012] In the formula, For the sample size, These are measured values. These are the fitted values ​​for the model.

[0013] In some embodiments, the spatial interpolation using the Kriging interpolation algorithm is calculated using the following formula:

[0014] In the formula, For the part to be interpolated The estimated value, For known locations The deformation measurement value, The interpolation weights are the corresponding parts. This represents the number of samples.

[0015] Secondly, embodiments of the present invention provide a spatiotemporal prediction device for dam deformation based on Kriging interpolation, the device comprising: The acquisition module is used to acquire deformation monitoring data from multiple monitoring points of the dam and clean the monitoring data to obtain a cleaned dataset. The processing module is used to perform unified processing on the cleaned monitoring data, including unified coordinate reference and data fusion, to form a unified time vector and monitoring dataset. The fitting module is used to fit the deformation data of each monitoring point using a regression analysis model based on the monitoring dataset to obtain a fitting model. The prediction module is used to predict the deformation data of each monitoring point using the fitting model, so as to obtain the predicted deformation data sequence of each monitoring point at future time. The verification module is used to verify the consistency between the predicted deformation data sequence and the corresponding real monitoring data. Based on the predicted deformation data sequence that has passed the verification, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. The update module is used to repeat the above steps and update the model when new monitoring data is added.

[0016] Thirdly, embodiments of this application provide an electronic device, including a memory and a processor, wherein the memory stores program code that can run on the processor, and when the program code is executed by the processor, it implements a spatiotemporal prediction method for dam deformation based on Kriging interpolation as described in any embodiment of the first aspect.

[0017] Fourthly, embodiments of this application provide a computer storage medium storing one or more programs, which can be executed by an electronic device as described in the third aspect to implement a spatiotemporal prediction method for dam deformation based on Kriging interpolation as described in any embodiment of the first aspect.

[0018] This invention provides a method and apparatus for spatiotemporal prediction of dam deformation based on Kriging interpolation. The method includes: acquiring deformation monitoring data from multiple monitoring points of the dam and cleaning the monitoring data to obtain a cleaned dataset; performing unified processing on the cleaned monitoring data, including unifying coordinate references and data fusion to form a unified time vector and monitoring dataset; fitting the deformation data of each monitoring point using a regression analysis model based on the monitoring dataset to obtain a fitted model; predicting the deformation data of each monitoring point using the fitted model to obtain a predicted deformation data sequence for each monitoring point at future times; and sequencing the predicted deformation data. The model is consistent with the corresponding real monitoring data. Based on the predicted deformation data sequence that passes the test, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. When new monitoring data is added, the above steps are repeated and the model is updated. By fusing multiple monitoring data, the spatial density of monitoring points and the temporal density of deformation data are improved. By combining regression analysis and the Kriging interpolation algorithm to form a temporally continuous deformation field, the model can accurately draw the deformation contour maps of the dam's transverse and longitudinal profiles, analyze the changes in deformation gradients, etc., and assist in the analysis of dam deformation patterns and the identification of safety hazards. The algorithm has the characteristics of simple calculation, high interpolation accuracy, and accurate prediction.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] The invention will now be described in more detail with reference to embodiments and the accompanying drawings.

[0021] Figure 1 A schematic diagram of an exemplary spatiotemporal prediction method for dam deformation based on Kriging interpolation, proposed in one embodiment of the present invention, is shown. Figure 2 A schematic diagram of an exemplary implementation process according to one embodiment of the present invention is shown; Figure 3 A schematic diagram illustrating an exemplary method for determining a unified time vector is shown in one embodiment of the present invention; Figure 4 The diagram shows a structural block diagram of a dam deformation spatiotemporal prediction device based on Kriging interpolation proposed in one embodiment of the present invention. Figure 5 This paper shows a structural block diagram of an electronic device for performing a kriging interpolation-based spatiotemporal prediction method for dam deformation according to an embodiment of this application. Figure 6This application illustrates a computer-readable storage medium for storing or carrying a method for spatiotemporal prediction of dam deformation based on Kriging interpolation, according to an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0023] Dams, as key structures in water conservancy and hydropower projects, play a vital role in flood control, power generation, irrigation, navigation, and regulation. The geological structure of dam sites is usually quite complex, and may contain fault zones, weak interlayers, etc. As a result, during long-term operation, dams are not only affected by their own weight, water pressure, temperature, and earthquakes, but also constrained by the complexity of geological conditions, causing the vertical or horizontal displacement of the dam to exhibit certain deformation patterns.

[0024] In related technologies, various monitoring devices, such as horizontal observation piers, leveling points, plumb lines, tension wires, static leveling, bimetallic markers, and laser collimators, are deployed in the dam foundation, inside the dam body, and on the dam surface to form monitoring sections. These sections acquire deformation data at different locations and elevations of the dam, allowing for the summarization and analysis of dam deformation patterns, resulting in visualized comprehensive analysis results. However, limitations in instrument and manpower costs make it difficult to achieve uniform and continuous spatial distribution of monitoring equipment. Furthermore, the varying observation frequencies and methods at different monitoring points, along with the discrete deformation data in both time and space, significantly hinder the construction of the deformation field and the effective analysis of deeper deformation patterns.

[0025] Dam deformation prediction and spatial interpolation can effectively address this problem. By using regression analysis models such as Gaussian functions, Fourier series, and polynomial functions, combined with historical data from monitoring points, and performing least-squares fitting to obtain model parameters, continuous time-domain prediction can be carried out. Furthermore, spatial interpolation algorithms such as inverse distance weighted average (IDW) and Kriging interpolation consider the spatial correlation between monitoring points, enabling the estimation of deformation data at unknown points using monitoring point data, transforming discrete spatial deformation data into a continuous deformation field. However, time-series deformation prediction and spatial interpolation alone are insufficient to obtain a time-continuous deformation field, and the accuracy of data from various monitoring points varies, while the observation frequency is affected by the observation method, making effective unification difficult.

[0026] Based on this, the existing dam deformation monitoring instruments vary in type, location, and quantity, and there are differences in observation frequency and accuracy. The deformation monitoring data is discrete in both time and space, and there is a lack of an effective overall algorithm for fusing multiple monitoring data into spatiotemporal prediction.

[0027] This invention provides a spatiotemporal prediction method for dam deformation based on Kriging interpolation. The method involves: first, compiling and statistically analyzing dam deformation monitoring instruments to obtain the temporal deformation sequence and station number (planar coordinates) of each monitoring point, and comprehensively determining the time vector; second, cleaning the data to remove outliers and abnormal measurements to ensure accuracy; third, using regression analysis to determine the fitting function model for the monitoring points and predicting deformation data at those points; and finally, using Kriging interpolation to obtain a temporally continuous deformation field, which can accurately plot the deformation contour maps of the dam's cross-sections and analyze deformation gradient changes, thus assisting in the analysis of dam deformation patterns and the identification of safety hazards. This algorithm features simple calculation, high interpolation accuracy, and precise prediction.

[0028] One of the methods for spatiotemporal prediction of dam deformation based on Kriging interpolation will be described in detail in the following embodiments.

[0029] The following describes an application scenario of a spatiotemporal prediction method for dam deformation based on Kriging interpolation provided by an embodiment of the present invention: Please see Figure 1 , Figure 1 This is a schematic diagram of a spatiotemporal prediction method for dam deformation based on Kriging interpolation provided in an embodiment of the present invention. In this embodiment, a spatiotemporal prediction method for dam deformation based on Kriging interpolation can be applied to, for example... Figure 4 The dam deformation spatiotemporal prediction device 300 based on Kriging interpolation shown is neutralized. Figure 5 In the electronic device 200 shown, the following is specifically for... Figure 1 The process shown is described in detail. A spatiotemporal prediction method for dam deformation based on Kriging interpolation may include S110 to S160.

[0030] S110: Obtain deformation monitoring data from multiple monitoring points on the dam, and clean the monitoring data to obtain a cleaned dataset.

[0031] In this embodiment, the deformation monitoring data includes data acquired by various monitoring devices deployed in the dam foundation, inside the dam body, and on the dam surface. Examples include vertical or horizontal displacement data measured by monitoring instruments such as observation piers, leveling points, plumb lines, tension lines, hydrostatic leveling, bimetallic markers, and laser collimators. This data is typically discrete in both time and space. Data cleaning involves removing gross errors and outliers to ensure measurement accuracy and create a reliable initial dataset.

[0032] In some embodiments, S110 cleansing the monitoring data includes: S111: Coarse screening process: Manually judge based on environmental data, dam structure type, and data consistency of nearby or same location measuring points to eliminate gross errors.

[0033] In this embodiment, coarse screening is the first step in data cleaning. Specifically, based on environmental data, such as reservoir water level, temperature, and general patterns of dam deformation, such as the positive correlation between the radial horizontal displacement of concrete gravity dams and arch dams downstream and the rise in reservoir water level and the decrease in temperature, as well as the concurrent data from adjacent measuring points and measuring points at the same location, the monitoring data are manually judged, and gross data that obviously does not conform to the pattern are removed.

[0034] S112: Cleaning process: The trend method and the extreme value method are used to jointly judge the recent historical data and identify abnormal data points to form a cleaned dataset; among them, the trend method is used to judge whether the data points deviate from the long-term trend, and the extreme value method is used to judge whether the data points exceed the reasonable range based on historical statistics.

[0035] In this embodiment of the application, fine washing is the second step of data cleaning. Specifically, it involves selecting data from the past three years, using the trend method to determine whether the data points deviate from the long-term trend, and the extreme value method to determine whether the data points exceed the reasonable range based on historical statistics. This joint judgment identifies abnormal data points, and then a second manual review is conducted to finally form the initial cleaned dataset.

[0036] S120: Perform unified processing on the cleaned monitoring data, including unified coordinate reference and data fusion, to form a unified time vector and monitoring dataset.

[0037] In this embodiment, a unified coordinate benchmark refers to sorting out the deformation data of different monitoring instrument measuring points, obtaining the time deformation sequence and station number of each monitoring point, such as plane coordinates, and unifying the coordinate benchmark. Combining as-built drawings and field survey data, three-dimensional spatial positioning of the dam at different elevations and plane locations is performed to determine the plane coordinates and elevations of all known parts. Among them, data fusion is used to address situations where multiple monitoring instruments may be deployed at the same location. By establishing linear relationships between data sequences of different accuracies and transforming them, the average value is finally taken to form a unique and reliable deformation data sequence for that location.

[0038] In some embodiments, data fusion includes: For multiple monitoring instruments deployed at the same location, the data sequence measured by the monitoring instrument with the highest accuracy shall be used as the benchmark.

[0039] In this embodiment of the application, during data fusion, the monitoring instrument with the highest theoretical accuracy among different types of measuring points at the same location is first determined. For example, the measurement accuracy of a precision leveling point is better than that of a static leveling point, and its measured data sequence is used. As a benchmark.

[0040] Establish a linear relationship between the data sequences measured by other monitoring instruments and the reference data sequences.

[0041] In this embodiment, a linear functional relationship is established between the data sequence measured by a monitoring instrument with slightly lower accuracy and the reference data sequence. .

[0042] The parameters of each linear relationship are obtained through regression analysis, and the proportional coefficients and other parameters in each of the above linear relationships are obtained through regression analysis.

[0043] The average of the baseline data and each transformed data is taken as the fused deformation data of that part at the corresponding time.

[0044] After obtaining the parameters, the baseline data The average value is taken from the data obtained by linear transformation and used as the final fused deformation data for that part at the corresponding time.

[0045] For example, the deformation trends of monitoring data from different types of measuring points in the same location should generally be consistent. Measuring point types are classified based on theoretical accuracy. For instance, the measurement accuracy of precision leveling points is better than that of static leveling points. Static leveling typically uses data from adjacent precision leveling points as the starting point for calculating the absolute vertical displacement. However, static leveling can significantly increase the frequency of observations through automation. Therefore, the deformation data with the highest accuracy and the deformation data with slightly lower accuracy exhibit the following linear functional relationship. , ,......, After obtaining the parameters through regression analysis, and The average value is taken as the monitoring data for this part, and the deformation data sequence for this part can be obtained by summarizing the data. The time dimension includes the observation time of different types of measuring points.

[0046] A unified time vector is constructed by extracting unique time points from the original observation time series of all monitoring points and sorting them chronologically. After data fusion, a dimensional monitoring dataset is finally formed, in which the unified time vector is constructed by extracting unique time points from the original observation time series of all monitoring points and sorting them chronologically.

[0047] For example, see Figure 3 As shown, assume there are three monitoring instruments JC1, JC2, and JC3, and a, b, and c units are deployed respectively, with their time series as follows: ; ; Its integrated time vector is .

[0048] Considering that multiple measuring points may be set up in the same location, the location where measuring points exist is: Through time series unification and data fusion, the dataset becomes... .

[0049] S130: Based on the monitoring dataset, a regression analysis model is used to fit the deformation data of each monitoring point to obtain the fitted model.

[0050] In this embodiment, the regression analysis model is preferably a Fourier series model. Least squares fitting is performed on the historical data of each monitoring point, and its expression is as follows:

[0051] In the formula, This is the amount of deformation. For time variables, , , w represents the model parameters, and H represents the number of harmonics.

[0052] The above expression can determine the deformation fitting function model at each monitoring point to characterize its change over time.

[0053] In some implementations, a regression analysis model is used to fit the deformation data at each monitoring point to obtain the fitted model. In this step, the historical data of the monitoring points is fitted using the least squares method to determine the model parameters, and the optimal model is selected from models with different harmonic counts based on the root mean square error (RMSE). The formula for calculating the root mean square error (RMSE) is as follows:

[0054] In the formula, For the sample size, These are measured values. These are the fitted values ​​for the model.

[0055] The parameters of the Fourier series model and the root mean square error were obtained. and determination coefficient ,in, The coefficient of determination represents the magnitude of the difference between the measured value and the fitted value, while RMSE represents the average deviation between the fitted value and the monitoring data. The closer the value is to 1, the better the regression model matches the monitoring results. A comprehensive comparison is made of the sums obtained from the fitting of each Fourier series model, and the model with the smaller sum is selected first.

[0056] S140: Use the fitting model to predict the deformation data of each monitoring point, and obtain the predicted deformation data sequence of each monitoring point at future times.

[0057] In this embodiment of the application, the time vector is considered based on the fitted regression model. For each The measuring point is located at Estimate the missing deformation data at each time point and predict the future. Predicting based on deformation data at each time point, obtaining... Temporal deformation sequence of each measuring point .

[0058] S150: The consistency of the predicted deformation data sequence with the corresponding real monitoring data is checked. Based on the predicted deformation data sequence that passes the check, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam.

[0059] In this embodiment, Kriging interpolation is a commonly used geostatistical interpolation algorithm. It describes the spatial structural changes of regionalized variables through a semivariogram, and then performs linear unbiased optimal estimation. Its interpolation formula is as follows:

[0060] In the formula, For the part to be interpolated The estimated value, For known locations The deformation measurement value, This is the interpolation weight for that part. . For the observation location and observation location The semi-variogram value, For the observation location and the part to be interpolated The semi-variogram value.

[0061] Solve the system of equations The interpolation weights can then be obtained. and Lagrange multiplier Then, the deformation estimate of the part to be interpolated is obtained.

[0062] S160: When new monitoring data is available, repeat the above steps and update the model.

[0063] In this embodiment of the application, Comparing the predicted deformation data of each measuring point with the actual monitoring data, when the deformation trends are consistent and If the value is less than the set threshold, the predicted data is considered accurate. If there is newly added measured deformation data, repeat steps S110 to S150.

[0064] This invention provides a spatiotemporal prediction method for dam deformation based on Kriging interpolation. By fusing multiple monitoring data, it improves the spatial density of monitoring points and the temporal density of deformation data. By combining regression analysis and Kriging interpolation algorithm to form a temporally continuous deformation field, it can accurately draw deformation contour maps of the dam's transverse and longitudinal profiles, analyze deformation gradient changes, etc., and assist in the analysis of dam deformation patterns and the identification of safety hazards. This method has the characteristics of simple calculation, high interpolation accuracy, and accurate prediction.

[0065] Please see Figure 4 , Figure 4 This invention provides a structural block diagram of a dam deformation spatiotemporal prediction device based on Kriging interpolation. The device includes: an acquisition module 310, a processing module 320, a fitting module 330, a prediction module 340, a verification module 350, and an update module 350, wherein: The acquisition module 310 is used to acquire deformation monitoring data from multiple monitoring points of the dam and clean the monitoring data to obtain a cleaned dataset. The processing module 320 is used to perform unified processing on the cleaned monitoring data, including unified coordinate reference and data fusion, to form a unified time vector and monitoring dataset. The fitting module 330 is used to fit the deformation data of each monitoring point based on the monitoring dataset using a regression analysis model to obtain a fitting model; The prediction module 340 is used to predict the deformation data of each monitoring point using a fitting model, and obtain the predicted deformation data sequence of each monitoring point at future times. The verification module 350 is used to verify the consistency between the predicted deformation data sequence and the corresponding real monitoring data. Based on the predicted deformation data sequence that has passed the verification, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. The update module 360 ​​is used to repeat the above steps and update the model when new monitoring data is added.

[0066] It should be noted that the device embodiments in this invention correspond to the aforementioned method embodiments. The specific principles in the device embodiments can be found in the content of the aforementioned method embodiments, and will not be repeated here.

[0067] In the several embodiments provided in this example, the coupling between modules can be electrical, mechanical, or other forms of coupling.

[0068] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0069] Please see Figure 5 , Figure 5 The present application provides a structural block diagram of an electronic device 200 that can perform the above-described spatiotemporal prediction method for dam deformation based on Kriging interpolation. The electronic device 200 may be a smartphone, tablet computer, computer, or portable computer.

[0070] The electronic device 200 also includes a processor 202 and a memory 204. The memory 204 stores programs that can execute the contents of the foregoing embodiments, and the processor 202 can execute the programs stored in the memory 204.

[0071] The processor 202 may include one or more cores for data processing and message matrix units. The processor 202 connects to various parts within the electronic device 200 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 204, and by calling data stored in the memory 204. Optionally, the processor 202 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 202 may integrate one or more of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem / decoder. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem / decoder handles wireless communication. It is understood that the modem / decoder may also be implemented separately as a communication chip, without being integrated into the processor.

[0072] Memory 204 may include random access memory (RAM) or read-only memory (ROM). Memory 204 can be used to store instructions, programs, code, code sets, or instruction sets. Memory 204 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (e.g., instructions for a user to obtain random numbers), instructions for implementing the various method embodiments described below, etc. The data storage area may also store data (e.g., random numbers) created by the terminal during use.

[0073] Electronic device 200 may also include a network module and a screen. The network module is used to receive and transmit electromagnetic waves, converting electromagnetic waves into electrical signals, thereby enabling communication with communication networks or other devices, such as audio playback devices. The network module may include various existing circuit elements used to perform these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, SIM cards, memory, etc. The network module can communicate with various networks such as the Internet, corporate intranets, and wireless networks, or communicate with other devices via wireless networks. The aforementioned wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The screen can display interface content and facilitate data interaction.

[0074] Please refer to Figure 6 , Figure 6 This diagram illustrates a structural block diagram of a computer-readable storage medium according to an embodiment of this application. The computer-readable storage medium 400 stores program code 410, which can be called by a processor to execute the methods described in the above method embodiments.

[0075] The computer-readable storage medium 400 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium 400 has storage space for program code 410 that performs any of the method steps described above. This program code 410 can be read from or written to one or more computer program products. The program code 410 may be compressed, for example, in a suitable form.

[0076] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform a kriging interpolation-based spatiotemporal prediction method for dam deformation described in the various optional implementations above.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A spatiotemporal prediction method for dam deformation based on Kriging interpolation, characterized in that, include: Deformation monitoring data from multiple monitoring points on the dam are acquired, and the monitoring data is cleaned to obtain a cleaned dataset. The cleaned monitoring data is processed in a unified manner, including unifying the coordinate reference and data fusion, to form a unified time vector and monitoring dataset; Based on the monitoring dataset, a regression analysis model was used to fit the deformation data of each monitoring point to obtain the fitting model. The deformation data of each monitoring point is predicted using the fitting model to obtain the predicted deformation data sequence of each monitoring point at future times; The predicted deformation data sequence is checked for consistency with the corresponding real monitoring data. Based on the predicted deformation data sequence that passes the check, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. When new monitoring data is added, repeat the above steps and update the model.

2. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 1, characterized in that, The cleaning of the monitoring data includes: Coarse screening process: Manual judgment is performed based on environmental data, dam structure type, and data consistency of nearby or same location measuring points to eliminate gross errors; Cleaning process: The recent historical data is jointly judged using the trend method and the extreme value method to identify abnormal data points and form a cleaned dataset; wherein, the trend method is used to determine whether the data points deviate from the long-term trend, and the extreme value method is used to determine whether the data points exceed the reasonable range based on historical statistics.

3. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 1, characterized in that, The data fusion includes: For multiple monitoring instruments deployed at the same location, the data sequence measured by the monitoring instrument with the highest accuracy shall be used as the benchmark; Establish a linear relationship between the data sequences measured by other monitoring instruments and the reference data sequence; The parameters of each linear relationship are obtained through regression analysis; The average of the baseline data and each transformed data is taken as the fused deformation data of that part at the corresponding time.

4. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 1, characterized in that, The unified time vector is constructed by extracting unique time points from the original observation time series of all monitoring points and sorting them in chronological order.

5. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 1, characterized in that, The expression for the regression analysis model is: In the formula, This is the amount of deformation. For time variables, , , w represents the model parameters, and H represents the number of harmonics.

6. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 5, characterized in that, In the step of fitting the deformation data of each monitoring point using a regression analysis model to obtain the fitted model, the historical data of the monitoring points are fitted using the least squares method to determine the model parameters, and the optimal model is selected from models with different harmonic counts based on the root mean square error (RMSE). The formula for calculating the root mean square error (RMSE) is as follows: In the formula, For the sample size, These are measured values. These are the fitted values ​​for the model.

7. The spatiotemporal prediction method for dam deformation based on Kriging interpolation according to claim 1, characterized in that, The spatial interpolation using the Kriging interpolation algorithm is calculated using the following formula: In the formula, For the part to be interpolated The estimated value, For known locations The deformation measurement value, The interpolation weights are the corresponding parts. This represents the number of samples.

8. A spatiotemporal prediction device for dam deformation based on Kriging interpolation, the device comprising: The acquisition module is used to acquire deformation monitoring data from multiple monitoring points of the dam and clean the monitoring data to obtain a cleaned dataset. The processing module is used to perform unified processing on the cleaned monitoring data, including unified coordinate reference and data fusion, to form a unified time vector and monitoring dataset. The fitting module is used to fit the deformation data of each monitoring point using a regression analysis model based on the monitoring dataset to obtain a fitting model. The prediction module is used to predict the deformation data of each monitoring point using the fitting model, so as to obtain the predicted deformation data sequence of each monitoring point at future time. The verification module is used to verify the consistency between the predicted deformation data sequence and the corresponding real monitoring data. Based on the predicted deformation data sequence that has passed the verification, spatial interpolation is performed using the Kriging interpolation algorithm to obtain the spatiotemporal continuous deformation field of the dam. The update module is used to repeat the above steps and update the model when new monitoring data is added.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores program code that can run on the processor. When the program code is executed by the processor, it implements a spatiotemporal prediction method for dam deformation based on Kriging interpolation as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code that can be invoked by one or more processors to execute a spatiotemporal prediction method for dam deformation based on Kriging interpolation as described in any one of claims 1-7.

Images