Method and device for predicting dam deformation

Abstract

The invention provides a method and device for predicting dam deformation, and belongs to the technical field of deformation prediction. The method comprises the following steps: fitting monitoring displacements of a plurality of measuring points in a historical time period in a measuring area of a dam to obtain fitting displacements of the plurality of measuring points; displacement residual errors of the multiple measuring points are obtained through the monitoring displacement and the fitting displacement of the multiple measuring points; further optimizing the displacement residual errors of the plurality of measuring points to obtain optimized residual errors of the plurality of measuring points; and predicting the displacement of the plurality of measuring points in the future time period according to the fitting displacement and the optimized displacement residual of the plurality of measuring points in the future time period. According to the method, the displacement residual error of each measuring point of the dam in the future time period can be predicted more accurately, so that the displacement of each measuring point in the dam deformation process is predicted based on the displacement residual error, and the prediction precision of dam deformation can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of deformation prediction, and in particular to a method and a device for predicting dam deformation. Background Art

[0002] As a concrete structure, dams are usually built in rivers, canyons and other terrains. Dams block water flow to form reservoirs. Dams may be deformed due to various factors. At present, the displacement of the dam at each measuring point is obtained through monitoring equipment arranged at each measuring point of the dam. The displacement of the internal structure of the dam in the future is predicted by the displacement of each measuring point, so that maintenance or reinforcement measures can be taken in time to prevent dam failure accidents.

[0003] However, due to the complex operating environment of the dam (for example, diverse geological structures, etc.), the factors that cause the deformation of the dam's internal structure are difficult to analyze. Therefore, the existing prediction methods have low prediction accuracy for dam deformation. Summary of the invention

[0004] The present invention proposes a method and device for predicting dam deformation. In the process of predicting dam deformation, the displacement residuals of each measuring point of the dam in the future time period can be predicted more accurately, so that the displacement of each measuring point in the dam deformation process can be predicted based on the displacement residual, which can improve the prediction accuracy of dam deformation.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] In the first aspect, the present invention provides a method for dam deformation, comprising: based on the monitored displacements of multiple measuring points in a measuring area of ​​the dam in a first historical time period and the monitored displacements of multiple measuring points in a second historical time period, fitting the fitted displacements of multiple measuring points in a second historical time period. The multiple measuring points in the measuring area have the same deformation law; the first historical time period is before the second historical time period. Then, based on the monitored displacements and the fitted displacements of the multiple measuring points in the second historical time period, the displacement residuals of the multiple measuring points in the measuring area are determined. The displacement residuals of the multiple measuring points in the measuring area are then optimized to obtain the optimized displacement residuals of the multiple measuring points. And based on the fitted displacements of the multiple measuring points in the future time period and the optimized displacement residuals of the multiple measuring points, the displacements of the multiple measuring points in the future time period are determined; the fitted displacements of the multiple measuring points in the future time period are obtained by data fitting based on the monitored displacements of the multiple measuring points in the measuring area in the historical time period.

[0007] In the process of determining the prediction residuals, the method for predicting dam deformation provided by the present invention can more accurately predict the displacement residuals of each measuring point of the dam in the future time period by optimizing the optimized residuals of multiple measuring points after optimization processing, thereby predicting the displacement of each measuring point in the dam deformation process based on the displacement residuals, and improving the prediction accuracy of dam deformation.

[0008] In an implementation of the first aspect and the second aspect, any one of the multiple measuring points in the measuring area is taken as a target measuring point, and the optimized displacement residual of the target measuring point is determined, including: based on the displacement residual of the target measuring point, the displacement residual of each of the multiple measuring points is decomposed to obtain multiple modal residuals corresponding to the multiple measuring points; wherein each measuring point corresponds to at least one modal residual. Multiple related modal residuals are screened out from the multiple modal residuals corresponding to the multiple measuring points; the correlation coefficients between the displacement residual of the target measuring point and the multiple related modal residuals are all greater than a preset threshold. From the multiple modal residual groups, the target modal residual group with the strongest correlation with the displacement residual of the target measuring point is screened out; the multiple modal residual groups are obtained by combining multiple related modal residuals, and each modal residual group includes N related modal residuals. The related modal residuals in the target modal residual group are weighted summed to obtain the optimized displacement residual of the target measuring point.

[0009] From the above content, it can be seen that the process of optimizing the displacement residual of the target measuring point can be divided into three steps.

[0010] In the first step, based on the modal number of the target measuring point, the displacement residuals of multiple measuring points in the measuring area are decomposed to obtain the modal residuals corresponding to the multiple measuring points, which provides more effective information that is closer to the deformation trend of the internal structure of the dam for the optimization processing of the displacement residuals of the target measuring point.

[0011] In the second step, the modal residuals corresponding to multiple measuring points are screened to obtain the target modal residual group, which eliminates the noise information for the optimal processing of the displacement residuals of the target measuring points and screens out the target modal residual group containing as much effective information as possible.

[0012] The third step is to reconstruct the target modal residual group to obtain the optimized residual, so that the reconstructed optimized residual is as consistent as possible with the change law of the displacement residual of the target measuring point.

[0013] In summary, the optimized residuals obtained through optimization processing contain more limited information and are closer to the changing law of the displacement residuals of the target measuring points, which can make the prediction accuracy of the displacement of the target measuring points in the future time period predicted by the optimized residuals higher.

[0014] In an implementation of the first aspect, based on the displacement residual of the target measuring point, the displacement residual of each measuring point in the plurality of measuring points is decomposed to obtain a plurality of modal residuals corresponding to the plurality of measuring points, including: determining the modal number of the displacement residual according to the central frequency of the displacement residual of the target measuring point. The displacement residual of each measuring point in the plurality of measuring points is decomposed using a variational modal decomposition method to obtain a plurality of modal residuals corresponding to the plurality of measuring points; wherein the number of modal residuals corresponding to each measuring point is equal to the modal number.

[0015] As is known to all, the displacement residuals can be decomposed by the variational modal decomposition method to extract effective information from the displacement residuals. Then, under the premise that multiple measuring points in the measurement area have the same deformation law, the displacement residuals of the target measuring point are decomposed by the modal number of the displacement residuals of the target measuring point, and effective information from the displacement residuals of the target measuring point can be extracted; and the displacement residuals of other measuring points in the measurement area are decomposed by the modal number of the displacement residuals of the target measuring point, and effective information similar to and / or related to the displacement residuals of the target measuring point can be extracted from the displacement residuals of other measuring points. Therefore, when the factors causing the deformation of the internal structure of the dam due to the complex operating environment of the dam are difficult to analyze, the above processing process can provide more effective information that is closer to the deformation trend of the internal structure of the dam for the optimization of the displacement residuals of the target measuring point, and the displacement of the target measuring point in the future time period can be predicted by the optimized displacement residuals obtained based on these effective information, with higher prediction accuracy.

[0016] In an implementation of the first aspect, a weighted summation is performed on multiple related modal residuals in a target modal residual group to obtain an optimized displacement residual of a target measuring point, including: calculating multiple weighted summation values ​​of related modal residuals in a target modal residual group under multiple groups of residual weight coefficients; wherein a group of residual weight coefficients corresponds to one weighted summation value. Among the multiple weighted summation values, the weighted summation value with the strongest correlation with the displacement residual of the target measuring point is used as the optimized displacement residual of the target measuring point.

[0017] From the above content, it can be seen that the above processing process simulates various situations of the target modal residual group reconstructed residuals by adjusting the weighting coefficients of multiple related modal residuals in the target modal residual group; and through the correlation between multiple weighted sum values ​​(corresponding to multiple groups of weighting coefficients) and the displacement residuals of the target measuring point, it indicates the similarity between the residual reconstructed by the target modal residual group and the displacement residual change law of the target measuring point (the stronger the correlation between the weighted sum value and the displacement residual of the target measuring point, the higher the similarity between the residual reconstructed by the target modal residual group and the displacement residual change law of the target measuring point). On this basis, the weighted sum value with the strongest correlation with the displacement residual of the target measuring point is selected as the optimized displacement residual of the target measuring point, and the optimized residual that is most similar to the displacement residual change law of the target measuring point can be obtained.

[0018] In an implementation of the first aspect, any one of the multiple measuring points in the measuring area is used as a target measuring point, and the displacement of the target measuring point in the future time period is determined, including: using an autoregressive integrated moving average model to process the optimized displacement residual of the target measuring point, and predicting the displacement residual of the target measuring point in the future time period. The fitted displacement and the displacement residual of the target measuring point in the future time period are superimposed to obtain the displacement of the target measuring point in the future time period.

[0019] From the above content, it can be seen that the optimized displacement residual of the measuring point obtained by the above optimization processing is obtained according to the displacement of the measuring point in the second historical time period. The optimized displacement residual can more accurately reflect the change of the displacement residual in the second historical time period. Based on this, in the above prediction process, the optimized displacement residual is further analyzed and processed to more reasonably and accurately predict the displacement residual of the measuring point in the future time period, which helps to more accurately predict the displacement of the measuring point in the future time period.

[0020] In an implementation manner of the first aspect, the method further includes: classifying all measuring points deployed in the dam to obtain multiple measuring areas.

[0021] In a second aspect, the present invention provides a device for predicting dam deformation, including a fitting module, a determination module, an optimization module and a prediction module.

[0022] The fitting module is used to fit the fitting displacements of multiple measuring points in a measuring area of ​​the dam in the first historical time period based on the monitoring displacements of multiple measuring points in a measuring area of ​​the dam in the second historical time period; wherein the multiple measuring points in the measuring area have the same deformation law; and the first historical time period is before the second historical time period;

[0023] The determination module is used to determine the displacement residuals of the plurality of measuring points in the measurement area according to the monitoring displacements and the fitting displacements of the plurality of measuring points in the second historical time period;

[0024] The optimization module is used to optimize the displacement residuals of multiple measuring points in the measuring area to obtain the optimized displacement residuals of the multiple measuring points;

[0025] The prediction module is used to determine the displacement of multiple measuring points in the future time period based on the fitted displacement of multiple measuring points in the future time period and the optimized displacement residuals of multiple measuring points; the fitted displacement of multiple measuring points in the future time period is obtained by data fitting based on the monitored displacement of multiple measuring points in the measuring area in the historical time period.

[0026] In an implementation of the second aspect, any one of the multiple measuring points in the measuring area is used as a target measuring point, and the above-mentioned optimization module is specifically used to: based on the displacement residual of the target measuring point, decompose the displacement residual of each of the multiple measuring points to obtain multiple modal residuals corresponding to the multiple measuring points; wherein each measuring point corresponds to at least one modal residual. Filter out multiple related modal residuals from the multiple modal residuals corresponding to the multiple measuring points; the correlation coefficients between the displacement residual of the target measuring point and the multiple related modal residuals are all greater than a preset threshold. From the multiple modal residual groups, filter out the target modal residual group with the strongest correlation with the displacement residual of the target measuring point; the multiple modal residual groups are obtained by combining multiple related modal residuals, and each modal residual group includes N related modal residuals.

[0027] In an implementation of the second aspect, the prediction module is specifically used to: process the optimized displacement residual of the target measuring point using an autoregressive integrated moving average model to predict the displacement residual of the target measuring point in the future time period. Superimpose the fitted displacement and displacement residual of the target measuring point in the future time period to obtain the displacement of the target measuring point in the future time period.

[0028] In an implementation of the second aspect, the device further includes a classification module. The classification module is used to classify all measurement points deployed in the dam to obtain multiple measurement areas.

[0029] In an implementation of the first aspect, the optimization module is further specifically used to: based on the displacement residual of the target measuring point, decompose the displacement residual of each measuring point in the multiple measuring points to obtain multiple modal residuals corresponding to the multiple measuring points, including: determining the modal number of the displacement residual according to the center frequency of the displacement residual of the target measuring point. Decompose the displacement residual of each measuring point in the multiple measuring points using a variational modal decomposition method to obtain multiple modal residuals corresponding to the multiple measuring points; wherein the number of modal residuals corresponding to each measuring point is equal to the modal number.

[0030] In an implementation of the second aspect, the optimization module is further specifically used to: calculate multiple weighted sum values ​​of the relevant modal residuals in the target modal residual group under multiple groups of residual weight coefficients; wherein one group of residual weight coefficients corresponds to one weighted sum value. Among the multiple weighted sum values, the weighted sum value with the strongest correlation with the displacement residual of the target measuring point is used as the optimized displacement residual of the target measuring point.

[0031] In an implementation of the second aspect, the prediction module is specifically used to: process the optimized displacement residual of the target measuring point using an autoregressive integrated moving average model to predict the displacement residual of the target measuring point in the future time period. Superimpose the fitted displacement and displacement residual of the target measuring point in the future time period to obtain the displacement of the target measuring point in the future time period.

[0032] In an implementation manner of the first aspect and the second aspect, N is equal to the number of modes.

[0033] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device performs the method described in the first aspect or any one of its implementations.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium, comprising computer program instructions, which, when executed by a computer, enable the computer to execute the method as described in the first aspect or any one of its implementations.

[0035] In a fifth aspect, the present invention provides a computer program product, comprising computer program instructions, which, when executed on a computer, enable the computer to execute the method as described in the first aspect or any one of its implementations.

[0036] The technical effects corresponding to the above-mentioned second to fifth aspects and their possible implementations can refer to the above-mentioned description of the technical effects of the first aspect and its possible implementations, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is one of the schematic diagrams of a method for predicting dam deformation provided in an embodiment of the present application;

[0038] Figure 2 is a schematic diagram of clustering results of all measuring points deployed in the dam provided in an embodiment of the present application;

[0039] Figure 3 is a schematic diagram of the change of the monitoring displacement of four measuring points in the first measuring area over time provided in an embodiment of the present application;

[0040] Figure 4 It is a schematic diagram of the monitored displacement, fitted displacement and displacement residual of the measuring point LS1 in the second historical time period provided in an embodiment of the present application;

[0041] Figure 5 This is a second schematic diagram of a method for predicting dam deformation provided in an embodiment of the present application;

[0042] Figure 6 This is a third schematic diagram of a method for predicting dam deformation provided in an embodiment of the present application;

[0043] Figure 7 is a schematic diagram of multiple modal residuals corresponding to four measuring points in the first measuring area provided in an embodiment of the present application;

[0044] Figure 8 This is a fourth schematic diagram of a method for predicting dam deformation provided in an embodiment of the present application;

[0045] Fig. 9 This is a fifth schematic diagram of a method for predicting dam deformation provided in an embodiment of the present application;

[0046] Fig.10 This is one of the structural schematic diagrams of a device for predicting dam deformation provided in an embodiment of the present application;

[0047] Fig.11 This is the second structural schematic diagram of a device for predicting dam deformation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The terms "first" and "second" etc. in the description and claims of the present invention are used to distinguish different objects rather than to describe a specific order of the objects.

[0049] "And / or" in the embodiments of the present application indicates the relationship between objects. For example, similar and / or related valid information can represent the following three situations: the existence of similar valid information alone, the existence of related valid information alone, and the existence of similar valid information and related valid information at the same time.

[0050] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0051] In the description of the present invention, unless otherwise specified, the meaning of "multiple" refers to two or more than two. For example, multiple measuring points in a measuring area refers to two or more measuring points belonging to one measuring area.

[0052] The method and device provided in the embodiments of the present application can be used to predict the displacement caused by dam deformation. Specifically, the displacement data of multiple measuring points arranged on the dam in a historical time period can be used to predict the respective displacement data of the multiple measuring points in a future time period.

[0053] It is understandable that a dam is an important water-blocking structure in water bodies such as reservoirs and rivers, and its main functions include flood control, power generation, irrigation, water supply and shipping. As the dam's service life increases, the dam is affected by factors such as the geological environment and natural disasters and deforms, which is specifically reflected in the displacement of some points on the dam, resulting in a certain displacement. In the embodiment of the present application, the prediction of dam deformation is specifically to predict the displacement of each measuring point of the dam. On this basis, maintenance or reinforcement measures can be taken for the dam according to the deformation prediction results to prevent the occurrence of dam breach accidents.

[0054] In order to solve the problem of low prediction accuracy of dam deformation in the background technology, the embodiment of the present application provides a method and device for predicting dam deformation. In this technical solution, the monitored displacements of multiple measuring points in a measuring area of ​​the dam in the first historical time period and the second historical time period are fitted to obtain the fitted displacements of the multiple measuring points in the second historical time period; and the displacement residuals of the multiple measuring points are obtained through the monitored displacements and the fitted displacements of the multiple measuring points in the second historical time period; further, the displacement residuals of the multiple measuring points are optimized to obtain the optimized residuals of the multiple measuring points; and the displacements of the multiple measuring points in the future time period are predicted based on the fitted displacements and the optimized displacement residuals of the multiple measuring points in the future time period. This technical solution can make a more accurate prediction of the displacement residuals of each measuring point of the dam in the future time period, thereby predicting the displacement of each measuring point in the dam deformation process based on the displacement residual, and can improve the prediction accuracy of the dam deformation.

[0055] Exemplarily, the method for predicting dam deformation provided in the embodiment of the present invention may be executed by an electronic device having a processing function, for example, the electronic device may be a computer, a server, etc. Taking the electronic device as a computer as an example, the hardware part of the computer may include: a processor, a memory, a network interface, a user interface, a communication bus, etc.

[0056] The processor is used to control the electronic device to perform related processing and computing tasks, for example, to determine the fitting displacement, displacement residual, optimized displacement residual, and displacement of multiple measurement points in a future time period. The processor may include a central processing unit (CPU) or other processors, and the processor may be single-core or multi-core, for example, the processor may include multiple CPUs.

[0057] The memory is used to store computer instructions and related data, for example, to store the monitored displacements, fitted displacements, displacement residuals, and optimized displacement residuals of multiple measuring points. The memory can be a random access memory (RAM), a read only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, or an optical memory, a disk storage medium or other magnetic storage device, or any other medium that can be used to store program codes or data that can be accessed by a computer. Optionally, the memory can be integrated into the processor, and the memory can also be independent of the processor.

[0058] The network interface is used for the computer to communicate with other devices or communication networks. The network interface can be a transceiver with transceiver functions. Optionally, the network interface can include a standard wired interface, a wireless interface (such as a WI-FI interface, a Bluetooth interface, a 5G interface).

[0059] The communication bus is used to realize the connection and communication between various components. For example, the processor, memory, network interface and user interface mentioned above can be interconnected through the communication bus.

[0060] The user interface may include a display screen and an input unit (such as a keyboard). Optionally, the user interface may also include a standard wired interface and a wireless interface.

[0061] Those skilled in the art will appreciate that the above-mentioned computer may also include more or fewer components, or a combination of certain components, or different arrangements of components, which is not limited in the embodiments of the present application.

[0062] like Figure 1 As shown, the method for predicting dam deformation provided in the embodiment of the present application includes S101-S105.

[0063] S101. Classify all measuring points deployed in the dam to obtain multiple measuring areas.

[0064] It can be understood that during the long-term operation of the dam, the displacement of each measuring point deployed in the dam (hereinafter referred to as monitoring displacement) can be monitored, and the deformation of the dam can be predicted based on the monitoring displacement of each measuring point. On this basis, the law followed by the displacement of each measuring point can be analyzed based on the monitoring displacement of each measuring point on the dam, that is, the deformation law of the measuring point, and the measuring points with the same deformation law among all the measuring points deployed in the dam can be classified into one category (that is, divided into one measuring area), and multiple measuring areas are obtained.

[0065] Optionally, Ward's agglomerative hierarchical clustering method is used to cluster all the measuring points deployed in the dam, so as to partition all the deployed measuring points and obtain multiple measuring areas of the dam. The specific steps of the above clustering process are as follows.

[0066] Step 1: Treat each measuring point deployed in the dam as a class. At this time, the level of all classes is 1 .

[0067] Step 2: Calculate the sum of squares of errors between the monitoring displacement of each class and the monitoring displacement of other classes (hereinafter referred to as the sum of squares of errors).

[0068] The above error sum of squares ESS(C) satisfies: Among them, x is the monitoring displacement of this class, and μ(C) is the average value of the monitoring displacements of other classes.

[0069] Step 3: For each class among all classes, exhaustively enumerate the binomial classes of the class (i.e., the combination of the class and another class among all classes except the class), merge the binomial classes with the smallest incremental sum of squared errors, and reduce the number of classes after merging by 1.

[0070] In the embodiment of the present application, for the i-th class, the i-th class is combined with the j-th class in other classes into a binomial class, and the error square increment of the binomial class (denoted as ΔESS ij )satisfy:

[0071] ΔESS ij =ESS(C t )-ESS(C i )-ESS(C j ),

[0072] Among them, ESS(C t ) is the sum of squared errors of the class after merging the i-th class with the j-th class, ESS(C i ) is the error sum of squares of the i-th class, ESS(C j ) is the sum of squared errors of the jth class.

[0073] Step 4: Add 1 to the level where the class merged in step 3 is located.

[0074] Repeat steps 2 to 4 above until the number of classes corresponding to the highest level is 1, which means that the convergence of all classes (corresponding to all measuring points deployed in the dam) has been completed, and the classes corresponding to all levels are the clustering results of Ward's aggregate hierarchical clustering method. Then, according to the actual situation, select some of the levels of classes as measurement areas, each of which contains multiple measurement points.

[0075] Figure 2 The clustering results of the above clustering method are shown. Figure 2 It can be seen that the clustering method is used to divide a certain number of measuring points deployed in the dam (the measuring points are denoted as LS) into six measuring areas (corresponding to six clusters), among which the first measuring area includes LS1, LS2, LS3, and LS4; the second measuring area includes LS20, LS22, LS23, LS21, LS24, LS25, and LS26; the third measuring area includes LS8, LS9, LS10, and LS11; the fourth measuring area includes LS5, LS6, and LS7; the fifth measuring area includes LS18, LS19, LS27, and LS28; and the sixth measuring area includes LS13, LS15, LS14, LS16, and LS17.

[0076] The following takes a measurement area of ​​the dam as an example to describe in detail the prediction process of the dam deformation.

[0077] S102: fitting the displacements of the plurality of measuring points in a measuring area of ​​the dam in the first historical period based on the monitoring displacements of the plurality of measuring points in the second historical period and the monitoring displacements of the plurality of measuring points in the second historical period to obtain the fitting displacements of the plurality of measuring points in the second historical period.

[0078] Among them, multiple measuring points in the measuring area have the same deformation law, and the first historical time period is before the second historical time period.

[0079] In the embodiment of the present application, the displacement of the dam in one of the historical time periods can be fitted by combining the monitored displacements of the dam in multiple historical time periods (referred to as fitted displacement). For example, the fitted displacement of the dam in the second historical time period is fitted based on the monitored displacements of multiple measuring points (the multiple measuring points belong to one measuring area) of the dam in the first historical time period and the second historical time period.

[0080] As an implementation method, the fitted displacement of the dam in a certain historical period can also be directly fitted based on the monitored displacement of the dam in the historical period. For example, the fitted displacement of multiple measuring points in a measuring area of ​​the dam in the second historical period can be fitted to obtain the fitted displacement of multiple measuring points in the second historical period.

[0081] The above deformation law refers to the law of monitoring displacement changes over time. Figure 3 Indicated Figure 2 The change pattern of the monitored displacement of the four measuring points (LS1, LS2, LS3, LS4) in the first measuring area over time is shown in the clustering results. Figure 3 It can be seen that the variation trends of the monitored displacements of the above four measuring points are almost the same. Therefore, the deformation laws of the four measuring points in the first measuring area are the same.

[0082] In the embodiment of the present application, the firefly-random forest algorithm can be used to perform displacement fitting to obtain the fitting displacement of multiple measuring points in the second historical time period. Specifically, the parameters of the random forest (RF) are first optimized using the firefly algorithm (Fire-fly Algorithm, FA), and then the monitoring displacements of multiple measuring points in a measuring area in the first historical time period and the second historical time period are fitted by the random forest after parameter optimization to obtain the fitting displacements of multiple measuring points in the second historical time period. Since the firefly-random forest algorithm is a common algorithm in the technical field, the embodiment of the present application does not further elaborate on the above process.

[0083] In one case, the Firefly algorithm optimizes the parameters of the random forest, such as the number of variable selections (Mtry) and the number of running trees (Ntree). The optimized Mtry is 3 and Ntree is 100.

[0084] Optionally, other fitting algorithms may also be used to fit the fitting displacements of multiple measuring points in the second historical time period, which is not limited in the embodiment of the present application.

[0085] S103: Determine displacement residuals of multiple measuring points in the measuring area according to the monitored displacements and fitted displacements of multiple measuring points in the second historical time period.

[0086] The displacement residuals of multiple measuring points in the measuring area in the second historical period are the differences between the monitoring displacements and the fitting displacements of the multiple measuring points in the second historical period. Figure 2 Take the measurement point LS1 in the first measurement area as an example. Figure 4 The monitored displacement, fitted displacement and displacement residual of the measuring point LS1 in the second historical time period are illustrated.

[0087] S104, optimizing the displacement residuals of multiple measuring points in the measuring area to obtain optimized displacement residuals of the multiple measuring points.

[0088] In the embodiment of the present application, the optimization processing method of the displacement residual of each of the multiple measuring points is the same, combined with Figure 1 ,like Figure 5 As shown, for the convenience of description, any measuring point among the multiple measuring points in the measuring area is taken as the target measuring point. The following describes the process of optimizing the displacement residual of the target measuring point to obtain the optimized displacement residual of the target measuring point through S1041-S1044.

[0089] S1041. Based on the displacement residual of the target measuring point, decompose the displacement residual of each measuring point in the multiple measuring points to obtain multiple modal residuals corresponding to the multiple measuring points; wherein each measuring point corresponds to at least one modal residual.

[0090] Optionally, combined Figure 5 ,like Figure 6 As shown, S1041 includes S1041A-S1041B.

[0091] S1041A. Determine the mode number of the displacement residual according to the center frequency of the displacement residual of the target measuring point.

[0092] In one implementation, the bit mode number is set within an integer range of 2 to 10, and the displacement residual is modally decomposed in ascending order of the mode number. After each decomposition, the center frequency of each modal residual is obtained. When the center frequencies of adjacent modal residuals are similar, it is considered that this modal decomposition is an over-decomposition of the signal, and the mode number -1 corresponding to this modal decomposition is determined as the mode number of the displacement residual.

[0093] S1041B. Decompose the displacement residual of each of the multiple measuring points using a variational modal decomposition method to obtain multiple modal residuals corresponding to the multiple measuring points, wherein the number of modal residuals corresponding to each measuring point is equal to the modal number (the modal number is greater than or equal to 1, therefore, each measuring point corresponds to at least one modal residual).

[0094] The above-mentioned variational mode decomposition (VMD) is a signal processing method that can decompose a multimodal signal (in the embodiment of the present application, the multimodal signal is the displacement residual of each measuring point) into a group of single-modal components IMF of different modes (i.e., the modal residual corresponding to each measuring point).

[0095] In one implementation, for each of the multiple measuring points, the displacement residual of the measuring point is decomposed by the variational mode decomposition method according to the mode number K of the displacement residual, to obtain K mode residuals corresponding to the measuring point. Specifically, the variational mode decomposition method realizes the decomposition of the displacement residual by minimizing the following objective function.

[0096] The objective function is:

[0097] Among them, u k (t) represents the kth modal residual in the modal residual corresponding to the measurement point, ω k represents the central frequency of the kth displacement residual, and x(t) represents the monitoring displacement of the measuring point in the second historical time period; It represents the difference between the kth modal residual in the modal residual corresponding to the measuring point and the monitored displacement of the measuring point in the second historical time period. Indicates that u k (t) and ω k The smallest value among them is taken as u k (t) Substitute The calculation results obtained later; Represents the smoothness constraint term, which is used to adjust the smoothness of the modal residual.

[0098] Still Figure 2 Taking the measuring point LS1 in the first measuring area shown as an example, if the mode number K of the displacement residual is determined to be 4 according to the central frequency of the displacement residual of the measuring point LS1, then according to the mode number K, the displacement residuals of the measuring points LS1, LS2, LS3 and LS4 in the first measuring area are respectively decomposed into 4 IMFs by the variational mode decomposition method to obtain 16 IMFs. For example, the displacement residual of LS1 is decomposed into IMF1~IMF4, the displacement residual of LS2 is decomposed into IMF5~IMF8, the displacement residual of LS3 is decomposed into IMF9~IMF12, and the displacement residual of LS4 is decomposed into IMF13~IMF16. Figure 7Figure 1 shows the evolution of the 16 IMFs over time.

[0099] From the above content, it can be seen that S1041 decomposes the displacement residuals of multiple measuring points in the measuring area where the target measuring point is located based on the modal number of the displacement residual of the target measuring point. Subsequently, multiple modal residuals corresponding to the multiple measuring points obtained by the decomposition are used as optimization conditions for the displacement residual of the target measuring point to optimize the displacement residual of the target measuring point.

[0100] As is known to all, the displacement residuals can be decomposed by the variational modal decomposition method to extract effective information from the displacement residuals. Then, under the premise that multiple measuring points in the measuring area have the same deformation law, the displacement residuals of the target measuring point are decomposed by the modal number of the displacement residuals of the target measuring point, and effective information from the displacement residuals of the target measuring point can be extracted; and the displacement residuals of other measuring points in the measuring area are decomposed by the modal number of the displacement residuals of the target measuring point, and effective information similar to and / or related to the displacement residuals of the target measuring point can be extracted from the displacement residuals of other measuring points. Therefore, when the factors causing the deformation of the internal structure of the dam due to the complex operating environment of the dam are difficult to analyze, S1041 can provide more effective information that is closer to the deformation trend of the internal structure of the dam for the optimization of the displacement residuals of the target measuring point, and the displacement of the target measuring point in the future time period can be predicted by the optimized displacement residuals obtained based on these effective information, with higher prediction accuracy.

[0101] S1042, selecting a plurality of related modal residuals from a plurality of modal residuals corresponding to a plurality of measuring points; correlation coefficients between the displacement residual of the target measuring point and the plurality of related modal residuals are all greater than a preset threshold.

[0102] In one implementation, the correlation coefficient between the displacement residual of the target measuring point and the multiple modal residuals corresponding to the multiple measuring points is the Pearson correlation coefficient. Exemplarily, the preset threshold can be set to 0.3174. Since the calculation method of the Pearson correlation coefficient belongs to the commonly used method in the technical field, it will not be described here.

[0103] Continue with Figure 2 Taking the measuring point LS1 in the first measuring area as an example, the Pearson correlation coefficients of each modal residual (IMF) in IMF1 to IMF16 and the displacement residual of the measuring point LS1 are calculated respectively. When the Pearson correlation coefficient of a modal residual and the displacement residual of the measuring point LS1 is greater than 0.3174, the modal residual is taken as the correlated modal residual.

[0104] Optionally, the correlation coefficient between the displacement residual of the target measuring point and the multiple modal residuals corresponding to the multiple measuring points may also be a Spearman rank correlation coefficient or a Kendall rank correlation coefficient, etc. The above preset threshold may also be other values ​​between 0 and 1 that meet the requirements, which is not limited in the embodiments of the present application.

[0105] From the above content, it can be seen that S1042 calculates the correlation coefficient between the displacement residual of the target measuring point and the multiple modal residuals corresponding to the multiple measuring points, determines the modal residuals with a correlation coefficient less than or equal to a preset threshold as noise information in the displacement residuals and screens them out, determines the modal residuals with a correlation coefficient greater than the preset threshold as related modal residuals containing valid information, and uses the related modal residuals to optimize the displacement residuals of the target measuring point, which can improve the optimization effect of the displacement residuals of the target measuring point and make the displacement residuals of the target measuring point more accurate.

[0106] S1043. Filter out a target modal residual group having the strongest correlation with the displacement residual of the target measuring point from the multiple modal residual groups; the multiple modal residual groups are obtained by combining multiple related modal residuals, and each modal residual group includes N related modal residuals.

[0107] According to the above S1041, the number N of related modal residuals in each modal residual group is equal to the above modal number K.

[0108] In one implementation, the correlation between the modal residual group and the displacement residual of the target measurement point satisfies:

[0109]

[0110] Among them, f1 represents the correlation coefficient between the modal residual group and the displacement residual of the target measurement point, r i represents the correlation coefficient between the ith relevant modal residual in the modal residual group and the displacement residual of the target measurement point, It represents the sum of the correlation coefficients between each relevant modal residual in the modal residual group and the displacement residual of the target measurement point.

[0111] by Figure 2 Taking the measuring point LS1 in the first measuring area shown as an example, the modal number K=4, after screening to obtain multiple related modal residuals (for example, IMF2, IMF4, IMF9, IMF10, IMF13), 4 related modal residuals are selected from the above multiple related modal residuals for combination, and a total of 5 modal residual groups are obtained. Then, the correlation coefficient between each modal residual group and the displacement residual of the target measuring point is calculated, and the modal residual group with the largest correlation coefficient (that is, the strongest correlation) is used as the target modal residual group (for example, {IMF2, IMF9, IMF10, IMF13}).

[0112] From the above content, it can be seen that S1043 combines multiple related modal residuals according to the modal number K to obtain multiple modal residual groups containing K related modal residuals, among which the modal residual group with the strongest correlation with the displacement residual of the target measuring point contains the most effective information, and the modal residual group containing the most effective information is used as the target modal residual group, which can provide more effective information that is closer to the deformation trend of the internal structure of the dam for the optimization of the displacement residual of the target measuring point.

[0113] S1044. Perform weighted summation on the relevant modal residuals in the target modal residual group to obtain the optimized displacement residual of the target measuring point.

[0114] Optionally, combined Figure 6 ,like Figure 8 As shown, S1044 includes S1044A-S1044B.

[0115] S1044A. Calculate multiple weighted sum values ​​of related modal residuals in the target modal residual group under multiple groups of residual weight coefficients; wherein one group of residual weight coefficients corresponds to one weighted sum value.

[0116] The sum of each group of residual weight coefficients in the above multiple groups of residual weight coefficients is 1. Assume that for a group of residual weight coefficients (a1, a2....a K ), where a1 represents the weight of the first relevant modal residual x1, a2 represents the weight of the second relevant modal residual x2, and a K represents the Kth related modal residual x K The weight of the above weighted sum satisfies

[0117] S1044B. Among the multiple weighted sum values, the weighted sum value having the strongest correlation with the displacement residual of the target measuring point is used as the optimized displacement residual of the target measuring point.

[0118] In the embodiment of the present application, the correlation coefficients of the multiple weighted sum values ​​and the displacement residuals of the target measuring points are calculated. The maximum value of the correlation coefficient indicates the strongest correlation. The correlation coefficient of the displacement residual with the target measuring point satisfies:

[0119]

[0120] Among them, R 2 Represents the weighted sum value Correlation coefficient with the displacement residual of the target measuring point, a i represents the weight of the residual of the i-th related mode, x i represents the i-th relevant modal residual, Represents the average value of K related modal residuals corresponding to the target measurement point.

[0121] by Figure 2 Taking the measuring point LS1 in the first measuring area as an example, after obtaining the target modal residual group (for example, {IMF2, IMF9, IMF10, IMF13}), the target modal residual group is weighted and summed based on multiple groups of residual weight coefficients to obtain multiple weighted sum values. Exemplarily, among the multiple groups of residual weight coefficients, the weighted sum value corresponding to the weight residual coefficient (0.1, 0.2, 0.3, 0.4) (i.e., 0.1×IMF2+0.2×IMF9+0.3×IMF10+0.4×IMF13) has the largest correlation coefficient with the optimized displacement residual of the target measuring point.

[0122] From the above content, it can be known that S1044 simulates various situations of the target modal residual group reconstructed residuals by adjusting the weighting coefficients of multiple related modal residuals in the target modal residual group; and indicates the similarity between the residual reconstructed by the target modal residual group and the displacement residual change law of the target measuring point through the correlation between multiple weighted sum values ​​(corresponding to multiple groups of weighting coefficients) and the displacement residual of the target measuring point (the stronger the correlation between the weighted sum value and the displacement residual of the target measuring point, the higher the similarity between the residual reconstructed by the target modal residual group and the displacement residual change law of the target measuring point). On this basis, the weighted sum value with the strongest correlation with the displacement residual of the target measuring point is selected as the optimized displacement residual of the target measuring point, and the optimized residual that is most similar to the displacement residual change law of the target measuring point can be obtained.

[0123] In summary, the process of optimizing the displacement residual of the target measuring point by S104 can be divided into three steps.

[0124] In the first step, based on the modal number of the target measuring point, the displacement residuals of multiple measuring points in the measuring area are decomposed to obtain the modal residuals corresponding to the multiple measuring points; the first step corresponds to S1041, which provides more effective information that is closer to the deformation trend of the internal structure of the dam for the optimization processing of the displacement residuals of the target measuring point.

[0125] The second step is to screen the modal residuals corresponding to multiple measuring points to obtain a target modal residual group; the second step corresponds to S1042-S1043, which is to optimize the displacement residuals of the target measuring points, eliminate noise information, and screen out a target modal residual group containing as much valid information as possible.

[0126] The third step is to reconstruct the target modal residual group to obtain the optimized residual; the third step corresponds to S1044, so that the reconstructed optimized residual is as consistent as possible with the change law of the displacement residual of the target measuring point.

[0127] In summary, the optimized residual obtained by the optimization process in S104 contains more limited information and is closer to the variation law of the displacement residual of the target measuring point, which can make the prediction accuracy of the displacement of the target measuring point in the future time period predicted by the optimized residual higher.

[0128] S105 , determining the displacements of the multiple measuring points in the future time period according to the fitted displacements of the multiple measuring points in the future time period and the optimized displacement residuals of the multiple measuring points.

[0129] The fitted displacements of multiple measuring points in the future time period are obtained by data fitting based on the monitored displacements of multiple measuring points in the measuring area in the historical time period. The specific data fitting process can refer to the relevant description of data fitting in S102, which will not be repeated here.

[0130] Exemplary, combined Figure 8 ,like Fig. 9 As shown, S105 includes S1051-S1052.

[0131] S1051. Use an autoregressive integrated moving average model to process the optimized displacement residual of the target measuring point, and predict the displacement residual of the target measuring point in the future time period.

[0132] It should be understood that the above-mentioned Autoregressive Integrated Moving Average model, referred to as ARIMA model, is a classic statistical model for time series data forecasting. The basic form of the ARIMA model can be expressed as ARIMA (p, d, q). Among them, p is the order of the autoregressive term (AR), which indicates the number of lagged observations used in the model; d is the number of differences, which indicates the number of difference operations performed to make the time series stable; q is the order of the moving average term (MA), which indicates the number of lagged values ​​of the past forecast errors used in the forecast error.

[0133] In one implementation, the ARIMA model satisfies:

[0134]

[0135] θ(B)=1-θ1B-θ2B 2 -…-θ q B q

[0136] φ(B)=1-φ1B-φ2B 2 -…-φ p B p

[0137] Among them, B represents the delay operator, Y t represents the optimized displacement residual of the target measuring point, represents the displacement residual of the target measurement point in the future time period, θ(B) represents the autoregressive operator that introduces the delay operator B in the autoregressive term (AR), φ(B) represents the moving average operator that introduces the delay operator B in the moving average term (MA), e t represents random error; θ1 represents the first-order autoregressive operator, θ2 represents the second-order autoregressive operator, B 2 represents the second-order delay operator, θ q represents the q-order autoregressive operator, B q represents the q-order delay operator; φ1 represents the first-order moving average operator, φ2 represents the second-order moving average operator, φ p represents the p-order moving average operator, B p represents the p-order delay operator. In the ARIMA model, θ1, θ2…θ q and φ1, φ2…θ q The quantity to be solved.

[0138] In the embodiment of the present application, for the ARIMA model, the values ​​of the parameters p, d and q in ARIMA (p, d, q) are first set, and then the above formula is used to solve θ1, θ2…θ q and φ1, φ2…θ q , and then solve for θ(B) and φ(B), through The displacement residual of the target measuring point in the future time period is predicted.

[0139] Optionally, the basic form of the ARIMA model may be ARIMA (0, 0, 2) or other forms, which is not further limited in the embodiments of the present application.

[0140] In an embodiment of the present application, the optimized displacement residual of the measuring point obtained in the above S104 is obtained based on the displacement of the measuring point in the second historical time period. The optimized displacement residual can more accurately reflect the change of the displacement residual in the second historical time period. Based on this, in S1051, the optimized displacement residual is further analyzed and processed to more reasonably and accurately predict the displacement residual of the measuring point in the future time period, which helps to more accurately predict the displacement of the measuring point in the future time period.

[0141] S1052: Superimpose the fitted displacement and displacement residual of the target measuring point in the future time period to obtain the displacement of the target measuring point in the future time period.

[0142] It can be understood that the above superposition is to add the fitted displacement and displacement residual of the target measuring point in the future time period in chronological order.

[0143] In summary, the embodiment of the present application fits the monitored displacements of multiple measuring points in a measuring area of ​​the dam in the first historical time period and the second historical time period to obtain the fitted displacements of the multiple measuring points in the second historical time period; obtains the displacement residuals of the multiple measuring points through the monitored displacements and the fitted displacements of the multiple measuring points in the second historical time period; then optimizes the displacement residuals of the multiple measuring points to obtain the optimized residuals of the multiple measuring points; and predicts the displacements of the multiple measuring points in the future time period based on the fitted displacements and the optimized displacement residuals of the multiple measuring points in the future time period. In the process of predicting the displacements of multiple measuring points in the future time period, the embodiment of the present application optimizes the displacement residuals of the multiple measuring points, so that the displacements of the multiple measuring points in the future time period obtained through the optimized displacement residuals of the multiple measuring points can more reasonably and accurately predict the displacement residuals of the measuring points in the future time period, which helps to more accurately predict the displacements of the measuring points in the future time period, thereby improving the prediction accuracy of the dam deformation.

[0144] Accordingly, the embodiment of the present application provides a device for predicting dam deformation, such as Fig.10 As shown, it includes a fitting module 901 , a determination module 902 , an optimization module 903 and a prediction module 904 .

[0145] The fitting module 901 is used to fit the fitting displacements of multiple measuring points in a measuring area of ​​the dam in the first historical time period based on the monitoring displacements of multiple measuring points in a measuring area of ​​the dam in the second historical time period; wherein the multiple measuring points in the measuring area have the same deformation law; and the first historical time period is before the second historical time period. For example, the fitting module 901 is used to implement S102 of the above-mentioned method for predicting dam deformation.

[0146] The determination module 902 is used to determine the displacement residuals of the multiple measuring points in the measurement area according to the monitoring displacements and the fitting displacements of the multiple measuring points in the second historical time period. For example, the determination module 902 is used to implement S103 of the above-mentioned method for predicting dam deformation.

[0147] The optimization module 903 is used to optimize the displacement residuals of multiple measuring points in the measurement area to obtain the optimized displacement residuals of the multiple measuring points. For example, the optimization module 903 is used to implement S104 of the above-mentioned method for predicting dam deformation.

[0148] The prediction module 904 is used to determine the displacement of multiple measuring points in the future time period according to the fitted displacement of multiple measuring points in the future time period and the optimized displacement residuals of the multiple measuring points; the fitted displacement of multiple measuring points in the future time period is obtained by data fitting based on the monitoring displacement of multiple measuring points in the measurement area in the historical time period. For example, the prediction module 904 is used to implement S105 of the above-mentioned method for predicting dam deformation.

[0149] Optionally, the optimization module 903 is specifically used for: based on the displacement residual of the target measuring point, decomposing the displacement residual of each measuring point in the multiple measuring points to obtain multiple modal residuals corresponding to the multiple measuring points; wherein each measuring point corresponds to at least one modal residual. Filter out multiple related modal residuals from the multiple modal residuals corresponding to the multiple measuring points; the correlation coefficients between the displacement residual of the target measuring point and the multiple related modal residuals are all greater than a preset threshold. Filter out the target modal residual group with the strongest correlation with the displacement residual of the target measuring point from the multiple modal residual groups; the multiple modal residual groups are obtained by combining multiple related modal residuals, and each modal residual group includes N related modal residuals. Perform weighted summation on the related modal residuals in the target modal residual group to obtain the optimized displacement residual of the target measuring point. For example, the optimization module 903 is specifically used to implement S1041-S1044 of the above-mentioned method for predicting dam deformation.

[0150] Optionally, the prediction module 904 is specifically used to: process the optimized displacement residual of the target measuring point using an autoregressive integrated moving average model to predict the displacement residual of the target measuring point in the future time period. Superimpose the fitted displacement and displacement residual of the target measuring point in the future time period to obtain the displacement of the target measuring point in the future time period. For example, the prediction module 904 is specifically used to implement S1051-S1052 of the above-mentioned method for predicting dam deformation.

[0151] In one implementation, combining Fig.10 ,like Fig.11 As shown, the above-mentioned device for predicting dam deformation further includes a classification module 905. The classification module 905 is used to classify all the measuring points deployed in the dam to obtain multiple measuring areas. For example, the classification module 905 is specifically used to implement S101 of the above-mentioned method for predicting dam deformation.

[0152] Each module of the above-mentioned device for predicting dam deformation can also be used to execute other steps in the above-mentioned method embodiment. All relevant contents involved in the above-mentioned method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0153] The embodiment of the present application also provides an electronic device, including: a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method described in the above embodiment. The processor can implement the above fitting module 501, determination module 502, optimization module 503, prediction module 504 and classification module 505; the above memory can also be used to store monitoring displacements, fitting displacements, displacement residuals and optimized displacement residuals of multiple measuring points.

[0154] An embodiment of the present application further provides a computer-readable storage medium, which includes a computer program. When the computer program runs on a computer, the method described in the above embodiment is executed.

[0155] An embodiment of the present application further provides a computer program product, which includes computer program instructions. When the computer program instructions are run on a computer, the method described in the above embodiment is executed.

[0156] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting dam deformation, characterized in that: include: Based on the monitored displacements of multiple measuring points in a measuring area of ​​the dam in the first historical time period and the monitored displacements of the multiple measuring points in the second historical time period, fitting is performed to obtain the fitted displacements of the multiple measuring points in the second historical time period; wherein the multiple measuring points in the measuring area have the same deformation law; and the first historical time period is before the second historical time period; determining displacement residuals of the plurality of measuring points in the measuring area according to the monitored displacements and the fitted displacements of the plurality of measuring points in the second historical time period; Optimizing the displacement residuals of the multiple measuring points in the measuring area to obtain optimized displacement residuals of the multiple measuring points; The displacements of the multiple measuring points in the future time period are determined according to the fitted displacements of the multiple measuring points in the future time period and the optimized displacement residuals of the multiple measuring points; the fitted displacements of the multiple measuring points in the future time period are obtained by data fitting based on the monitored displacements of the multiple measuring points in the measuring area in the historical time period.

2. The method according to claim 1, characterized in that Taking any measuring point among the multiple measuring points in the measuring area as a target measuring point, and determining the optimized displacement residual of the target measuring point, including: Based on the displacement residual of the target measuring point, decomposing the displacement residual of each measuring point in the multiple measuring points to obtain multiple modal residuals corresponding to the multiple measuring points; wherein each measuring point corresponds to at least one modal residual; Screening out a plurality of related modal residuals from the plurality of modal residuals corresponding to the plurality of measuring points; the correlation coefficients between the displacement residual of the target measuring point and the plurality of related modal residuals are all greater than a preset threshold; From the multiple modal residual groups, a target modal residual group having the strongest correlation with the displacement residual of the target measuring point is screened out; the multiple modal residual groups are obtained by combining the multiple related modal residuals, and each modal residual group includes N related modal residuals; The relevant modal residuals in the target modal residual group are weighted summed to obtain the optimized displacement residual of the target measuring point.

3. The method according to claim 2, characterized in that Based on the displacement residual of the target measuring point, the displacement residual of each measuring point in the multiple measuring points is decomposed to obtain multiple modal residuals corresponding to the multiple measuring points, including: Determining the mode number of the displacement residual according to the center frequency of the displacement residual of the target measuring point; The displacement residual of each measuring point in the plurality of measuring points is decomposed by using a variational mode decomposition method to obtain a plurality of mode residuals corresponding to the plurality of measuring points; wherein the number of mode residuals corresponding to each measuring point is equal to the mode number.

4. The method according to claim 3, characterized in that N is equal to the number of modes.

5. The method according to claim 2, characterized in that A weighted summation is performed on a plurality of related modal residuals in the target modal residual group to obtain an optimized displacement residual of the target measuring point, including: Calculating multiple weighted sum values ​​of the relevant modal residuals in the target modal residual group under multiple groups of residual weight coefficients; wherein one group of residual weight coefficients corresponds to one weighted sum value; Among the multiple weighted sum values, the weighted sum value having the strongest correlation with the displacement residual of the target measuring point is used as the optimized displacement residual of the target measuring point.

6. The method according to claim 1, characterized in that Taking any measuring point among the multiple measuring points in the measuring area as a target measuring point, and determining the displacement of the target measuring point in a future time period, includes: The optimized displacement residual of the target measuring point is processed by using an autoregressive integrated moving average model to predict the displacement residual of the target measuring point in a future time period; The fitted displacement and the displacement residual of the target measuring point in the future time period are superimposed to obtain the displacement of the target measuring point in the future time period.

7. The method according to claim 1, characterized in that The method further comprises: All measuring points deployed in the dam are classified to obtain a plurality of measuring areas.

8. A device for predicting dam deformation, characterized in that: It includes fitting module, determination module, optimization module and prediction module; The fitting module is used to fit the fitting displacements of the multiple measuring points in a measuring area of ​​the dam in the first historical time period based on the monitoring displacements of the multiple measuring points in a measuring area of ​​the dam in the second historical time period; wherein the multiple measuring points in the measuring area have the same deformation law; and the first historical time period is before the second historical time period; The determination module is used to determine the displacement residuals of the plurality of measuring points in the measurement area according to the monitored displacements and the fitted displacements of the plurality of measuring points in the second historical time period; The optimization module is used to optimize the displacement residuals of the multiple measuring points in the measuring area to obtain the optimized displacement residuals of the multiple measuring points; The prediction module is used to determine the displacement of the multiple measuring points in the future time period according to the fitted displacement of the multiple measuring points in the future time period and the optimized displacement residuals of the multiple measuring points; the fitted displacement of the multiple measuring points in the future time period is obtained by data fitting based on the monitored displacement of the multiple measuring points in the measuring area in the historical time period.

9. An electronic device, characterized in that: It comprises a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The method comprises computer program instructions, which, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 7.

Images