Rockfill dam material parameter tracking inversion method and system based on information transfer

Abstract

The invention provides a rock-fill dam material parameter tracking inversion method and system based on information transfer, and the method comprises the steps: extracting the deformation characteristics of a typical measurement point at a current stage in rock-fill dam multi-source monitoring data; comparing the deformation characteristics of the typical measuring points in the current stage with the predicted deformation characteristics of the typical measuring points in the previous stage, and calculating an inversion reliability coefficient of the previous stage; based on the deformation features of the typical measuring points in the current stage and the optimal material parameters and inversion reliability coefficients obtained through inversion in the previous stage, a pre-trained proxy model is used for inversion, rock-fill dam material parameter optimization is carried out through an improved particle swarm algorithm, and the optimal material parameters in the current stage are obtained; and according to the optimal material parameters of the current stage, calculating predicted deformation characteristics of the next stage.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of hydraulic engineering, and particularly relates to a rock-fill dam material parameter tracking inversion method and system based on information transmission. BACKGROUND

[0002] Using a reasonable constitutive model and material parameters for rock-fill dam finite element numerical simulation is one of important ways to evaluate dam deformation state and ensure dam deformation safety. For high rock-fill dams, the life period is long, and the material properties are difficult to remain stable during the period affected by various factors. In order to ensure accurate prediction of high rock-fill dam deformation, continuous tracking inversion update of dam material parameters needs to be carried out according to the continuously accumulated monitoring data. In the current related work, only the inversion times in the dam construction and operation process are increased, but the successive inversions are independent of each other, and the continuity of the dam material property change process is not considered. Due to the randomness of the optimization process, the material parameters obtained by inversion fluctuate greatly in value, that is, the material properties have a large degree of mutation in the life period of the dam, which is obviously inconsistent with the engineering practice. SUMMARY

[0003] In order to overcome the lack of consideration of the continuity of the dam material property change process in the existing rock-fill dam material parameter inversion technology, and the deficiency that the material parameters obtained by inversion fluctuate greatly in value, the application provides a rock-fill dam material parameter tracking inversion method and system based on information transmission. The posterior effect of the material parameters obtained by inversion in the previous stage is used as prior information in the current stage, the material parameters with higher precision are searched on the basis of the inversion parameters in the previous stage combined with the monitoring data in the current stage, and accordingly, the material property change process in the life period of the dam can be obtained more reasonably, and the prediction accuracy can be effectively improved as the dam deformation monitoring data is continuously enriched, thereby effectively improving the accuracy and rationality of the rock-fill dam deformation prediction and dam safety analysis.

[0004] According to one aspect of the application, a rock-fill dam material parameter tracking inversion method based on information transmission is provided, which comprises:

[0005] Extracting the deformation characteristics of a typical measuring point in the current stage from the multi-source monitoring data of the rock-fill dam;

[0006] Comparing the deformation characteristics of the typical measuring point in the current stage with the predicted deformation characteristics in the previous stage to calculate the inversion reliability coefficient of the previous stage;

[0007] Based on the deformation characteristics of the typical measuring point in the current stage, the optimal material parameters obtained by inversion in the previous stage and the inversion reliability coefficient, using a pre-trained surrogate model for inversion, and using an improved particle swarm algorithm for rock-fill dam material parameter optimization, the optimal material parameters in the current stage are obtained;

[0008] According to the optimal material parameters of the current stage, the predicted deformation characteristics of the next stage are calculated.

[0009] As a further technical solution, the acquisition process of the typical measuring point is: the multi-source monitoring data of the rockfill dam is input into the cloud generation model in the form of measuring points, and the cloud generation model performs measuring point quality evaluation based on the numerical characteristics of the set typical measuring points, and the typical measuring points are obtained through screening.

[0010] As a further technical solution, the inversion reliability coefficient of the previous stage is equal in value to the determination coefficient calculated according to the deformation characteristics of the typical measuring point in the current stage and the predicted deformation characteristics in the previous stage; in the initial stage, the determination coefficient is calculated according to the deformation characteristics of the typical measuring point in the initial stage and the deformation characteristics calculated by the three-dimensional finite element model of the rockfill dam according to the initial material parameters, and the inversion reliability coefficient of the initial stage is obtained.

[0011] As a further technical solution, the pre-training process of the surrogate model includes:

[0012] According to the preset material parameter value interval, a plurality of groups of trial material parameters are designed;

[0013] A three-dimensional finite element model of the rockfill dam is constructed, and a finite element seeding calculation is performed based on the plurality of groups of trial material parameters, and the deformation characteristics in the calculation results are extracted;

[0014] The trial parameters are taken as the input of the artificial neural network, and the deformation characteristics in the calculation results are taken as the output, and the artificial neural network is trained, and the trained artificial neural network is output as the surrogate model.

[0015] As a further technical solution, the improved particle swarm optimization algorithm introduces a clustering algorithm into the particle swarm optimization algorithm, clusters the particle swarm, obtains a plurality of particle clusters, and introduces the optimal solution of the clustered particle clusters into the calculation formula of the conventional particle swarm optimization algorithm.

[0016] As a further technical solution, the acquisition steps of the optimal material parameters of the current stage are: setting a plurality of groups of rockfill dam material parameter optimizations, obtaining the mean value and 95% confidence interval of the optimization results, selecting a group of material parameters whose material parameters fall within the 95% confidence interval of the optimization results and whose value of the objective function is the smallest as the optimal material parameters of the current stage.

[0017] As a further technical solution, the mathematical representation of the objective function is:

[0018]

[0019] In the formula, k is the number of inversions, which is used to represent the current stage; j represents the serial number of the material parameter, and p is the total number of the material parameters; x j，krepresents the value of the jth material parameter in the kth inversion process; F k (·) represents the objective function of the current stage inversion of the prior information iterative supervision; f k (·) is the basic objective function of the current stage inversion; g(k) is the prior constraint of the current stage, which is calculated by the deformation characteristics of the typical measuring point in the current stage, the optimal material parameters obtained in the last stage inversion, and the inversion reliability coefficient; w0 represents the weight coefficient.

[0020] According to another aspect of the present specification, a rock-fill dam material parameter tracking inversion system based on information transmission is provided, comprising:

[0021] A deformation feature extraction module is configured to extract the deformation characteristics of the typical measuring point in the current stage from the multi-source monitoring data of the rock-fill dam.

[0022] A prior information acquisition module is configured to compare the deformation characteristics of the typical measuring point in the current stage with the predicted deformation characteristics in the last stage, and calculate the inversion reliability coefficient in the last stage.

[0023] A tracking inversion module is configured to use a pre-trained surrogate model for inversion based on the deformation characteristics of the typical measuring point in the current stage, the optimal material parameters obtained in the last stage inversion, and the inversion reliability coefficient, and perform rock-fill dam material parameter optimization through an improved particle swarm algorithm to obtain the optimal material parameters in the current stage.

[0024] A posterior information calculation module is configured to calculate the predicted deformation characteristics in the next stage according to the optimal material parameters in the current stage.

[0025] According to another aspect of the present specification, an electronic device is provided, comprising a memory and a processor, wherein the memory stores program instructions executed by the processor, and the processor invokes the program instructions to perform a rock-fill dam material parameter tracking inversion method based on information transmission.

[0026] According to another aspect of the present specification, a non-transitory computer readable storage medium is provided, which stores computer instructions for executing a rock-fill dam material parameter tracking inversion method based on information transmission.

[0027] Compared with the prior art, the beneficial effects of the present application are that: by connecting multiple inversion processes in sequence through tracking material parameter inversion results, the whole tracking inversion process is completed, and the development law of material properties in the construction and operation process of rockfill dams, especially high rockfill dams, is revealed to a certain extent; by taking the posterior effect of the inversion parameters obtained in the previous stage as prior information in the current stage, and combining the monitoring data in the current stage to guide the parameter optimization process, the prior information is iteratively transmitted, each inversion is re-optimized based on the previous inversion, as the monitoring data becomes richer, the inversion parameters become more reasonable and reliable, the deformation prediction accuracy of the rockfill dam is gradually improved, and the accuracy and rationality of the deformation prediction and safety analysis of the rockfill dam are effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0029] Figure 1 A flowchart of a rockfill dam material parameter tracking inversion method based on information transmission provided by the embodiment of the present application is shown in the figure.

[0030] Figure 2 An artificial neural network structure used in the embodiment of the present application is shown in the figure.

[0031] Figure 3 A prediction effect verification figure of the agent model obtained by training in the embodiment of the present application is shown in the figure.

[0032] Figure 4 A six-time inversion time node distribution figure used in the embodiment of the present application is shown in the figure.

[0033] Figure 5 An information transmission process schematic diagram used in the embodiment of the present application is shown in the figure.

[0034] Figure 6 A comparison figure of the optimization process in the 6th stage and the conventional method in the embodiment of the present application is shown in the figure.

[0035] Figure 7 An evolution process of the material parameters obtained by the 6th stage inversion of the Jiangpinghe rockfill dam project in the embodiment of the present application is shown in the figure.

[0036] Figure 8 A deformation cloud chart of a typical section of the Jiangpinghe rockfill dam in the full storage period and a panel deflection cloud chart at that time in the embodiment of the present application are shown in the figure.

[0037] Figure 9A typical deformation time history curve diagram of a measurement point in the embodiment of the present application is shown in the figure;

[0038] Figure 10 A structure schematic diagram of a rockfill dam material parameter tracking inversion system based on information transmission provided by the embodiment of the present application is shown in the figure.

[0039] Figure 11 A structure schematic diagram of an electronic device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0040] It should be noted that:

[0041] The terms "include" and "have" and any variations thereof in the specification and claims of the present application and the above-mentioned figures are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to the steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0042] The block diagram shown in the figure is only a functional entity, which does not necessarily correspond to a physically independent entity. That is, the functional entity can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. The flowchart shown in the figure is only an exemplary description, which does not necessarily include all contents and operations / steps, and does not necessarily be executed in the order described. For example, some operations / steps can be further decomposed, and some operations / steps can be combined or partially combined, so the actual execution order can be changed according to the actual situation.

[0043] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described clearly and completely below in combination with the figures in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application. In addition, the technical features in each embodiment or single embodiment provided by the present application are combined with each other at will to form new technical schemes, which are not restricted by the order of steps and / or structure composition mode, but must be based on the implementation by those skilled in the art. When the combination of technical schemes contradicts each other or cannot be implemented, it should be considered that the combination of technical schemes does not exist, and is not within the scope of protection required by the present application.

[0044] As Figure 1As shown, a rock-fill dam material parameter tracking inversion method based on information transmission comprises:

[0045] Step 1, extract the deformation characteristics of the typical measuring point in the current stage in the multi-source monitoring data of the rock-fill dam;

[0046] Step 2, compare the deformation characteristics of the typical measuring point in the current stage and the predicted deformation characteristics in the last stage, and calculate the inversion reliability coefficient of the last stage;

[0047] Step 3, based on the deformation characteristics of the typical measuring point in the current stage and the optimal material parameters and the inversion reliability coefficient obtained in the last stage, use the proxy model inversion, and use the improved particle swarm algorithm to optimize the material parameters of the rock-fill dam, and obtain the optimal material parameters in the current stage;

[0048] Step 4, according to the optimal material parameters in the current stage, calculate the predicted deformation characteristics in the next stage.

[0049] As a specific embodiment, in step 1, the multi-source monitoring data includes the internal settlement monitoring data of the dam obtained by various monitoring instruments arranged in the dam, and the external deformation observation data of the rock-fill dam obtained by InSAR technology, etc., the main research object is the settlement of the rock-fill dam during operation, and the monitoring instrument includes but is not limited to: water pipe type settlement instrument, electromagnetic type settlement instrument, flexible displacement meter, pipeline robot, etc.

[0050] According to the inversion requirement, select appropriate deformation characteristics, and in the conventional case, the settlement increment in the stage covered by the multi-source monitoring data of the rock-fill dam can be selected, in this paper, the deformation characteristics are uniformly expressed as settlement. In order to construct the inversion objective function later, the measured settlement value of each measuring point in the monitoring stage (current stage) in the multi-source monitoring data of the rock-fill dam is selected as the deformation characteristics.

[0051] As a preferred embodiment, step 1 further comprises: pre-processing the monitoring data, using the adaptive wavelet transform filtering method (WT), and adaptively adjusting the filtering parameters according to the characteristics of the monitoring data to realize the noise reduction of the monitoring data.

[0052] Further, in step 1, the acquisition process of the typical measuring point is: input the multi-source monitoring data of the rock-fill dam in the form of measuring point into the cloud generation model, the cloud generation model performs measuring point quality evaluation based on the numerical characteristics of the set typical measuring point, and selects the typical measuring point.

[0053] As a specific embodiment, the cloud generation model generates a large number of "cloud droplets" in the numerical space to express the uncertainty characteristics of the concept by setting the expectation (Ex), entropy (En) and hyper-entropy (He) representing the characteristics of the "high-quality measuring point". The measuring point data is substituted into the cloud generation model to calculate the degree of overlap with the target concept, so as to measure the membership degree of the measuring point, thereby screening out high-quality measuring points meeting the expected distribution.

[0054] Further, in step 2, the inversion reliability coefficient of the previous stage is numerically equivalent to the determination coefficient calculated according to the deformation characteristics of the typical measuring point in the current stage and the predicted deformation characteristics in the previous stage; in the initial stage, the determination coefficient is calculated according to the deformation characteristics of the typical measuring point in the initial stage and the deformation characteristics calculated by the three-dimensional finite element model of the rockfill dam according to the initial material parameters, to obtain the inversion reliability coefficient of the initial stage.

[0055] Further, in step 3, the training process of the surrogate model includes:

[0056] Step 3-1, according to the preset material parameter value range, design several groups of trial material parameters;

[0057] Step 3-2, construct a finite element three-dimensional model of the rockfill dam, and perform finite element seeding calculation based on the several groups of trial material parameters to extract the deformation characteristics in the calculation results;

[0058] Step 3-3, taking the trial parameters as the input of the artificial neural network and the deformation characteristics in the calculation results as the output, training the artificial neural network, and outputting the trained artificial neural network as the surrogate model.

[0059] As a specific embodiment, in step 3-1, the suitable material parameter value range is determined according to the test material parameters of the rockfill dam material obtained by the indoor triaxial test, and the Latin hypercube sampling method is used to obtain several orthogonal material parameter combinations, which are arranged as trial material parameters.

[0060] As a preferred embodiment, in the initial stage, the predicted deformation characteristics of the previous stage are the deformation characteristics calculated by the three-dimensional finite element model of the rockfill dam according to the initial material parameters, the initial material parameters are provided by the trial material parameters, and the deformation characteristics calculated based on the three-dimensional finite element model of the rockfill dam are used as the predicted deformation characteristics of the previous stage in the initial stage, which provides an information transmission path for the evolution process from the test parameters to the inversion parameters in each stage, and realizes reasonable connection of the evolution process from the trial parameters to the latest inversion parameters.

[0061] As a specific embodiment, in step 3-2, a three-dimensional finite element model of the rockfill dam is constructed by using the modeling function of ANSYS software, and the finite element calculation is carried out based on the trial parameters in step 3-1 by using ABAQUS software, the calculation results of each group are sorted out, and the deformation calculation amount of the key time nodes (such as each period of water storage node, acceptance node: before each period of water storage, it needs to be filled to a certain characteristic elevation and safety analysis is carried out for this time node, in addition, the project also has a review and acceptance stage, and an analysis report is provided for the review time node) in the calculation results is extracted.

[0062] As a specific embodiment, a multi-layer feedforward neural network is trained as a surrogate model. The model input is a plurality of material parameters after normalization processing, and the output is the calculated settlement of the typical measuring point of the rockfill dam structure. The network learns the nonlinear mapping relationship between the input and the output through training, replaces the original numerical simulation with high calculation cost, and realizes the rapid approximation of deformation prediction.

[0063] Further, in step 3, the improved particle swarm optimization algorithm obtains a plurality of particle clusters by introducing a clustering algorithm to cluster the particle swarm, and introduces the optimal solution of the clustered particle cluster into the calculation formula of the particle swarm optimization algorithm.

[0064] As a supplementary description, the calculation formula of the conventional particle swarm optimization algorithm is:

[0065]

[0066]

[0067] In the formula, v and x respectively represent the particle velocity and position; ω represents the learning rate; k represents the time; id represents the particle number; d represents the particle global; c1 and c2 are learning factors; r1, and r2 are random numbers; pBest represents the particle individual extreme value (personal best); and gBest represents the global extreme value (global best).

[0068] The improved particle swarm algorithm obtains a plurality of particle clusters by introducing a K-means clustering algorithm to cluster the particle swarm. In addition to retaining the individual extreme value and the global extreme value, the optimal solution (abest) of the clustered particle cluster is also added to the above conventional algorithm, and the calculation formula is as follows:

[0069]

[0070]

[0071] In the formula, c3 is a learning factor, r3 is a random number; and aBest represents the particle cluster extreme value.

[0072] Further, in step 3, the step of obtaining the optimal material parameters of the current stage is: setting multiple groups of rockfill dam material parameter optimization, obtaining the optimization result mean value and 95% confidence interval, selecting the material parameters falling within the 95% confidence interval of the optimization result and having the minimum objective function value as the optimal material parameters of the current stage.

[0073] Further, the objective function of the inversion process is:

[0074]

[0075] In the formula, k is the number of inversions, used to represent the current stage; j represents the material parameter serial number, and p is the total number of material parameters; x j，k represents the value of the jth material parameter in the kth inversion process; F k (·) represents the objective function of the rockfill dam parameter tracking inversion method with prior information iterative supervision; f k (·) is the conventional inversion objective function; g(k) is the prior constraint of the current stage, which is calculated based on the deformation characteristics of the typical monitoring points in the current stage, the optimal material parameters obtained in the previous stage, and the inversion reliability coefficient; w0 represents the weight coefficient.

[0076] As a supplementary explanation, in order to comprehensively consider the monitoring data and the material parameters obtained in the previous stage and their inversion reliability coefficients, the method adjusts the objective function of the conventional inversion as follows:

[0077] The conventional inversion objective function is as follows:

[0078]

[0079] In the formula, represents the conventional inversion objective function; represents the measured deformation characteristics of the ith observation point, represents the simulated deformation characteristics of the observation point calculated by the parameters , ,..., n represents the total number of observation points.

[0080] The objective function of the rockfill dam parameter tracking inversion method with prior information iterative supervision is:

[0081]

[0082] Among them, the weight coefficient w0 is 0.1 optionally;

[0083] The calculation formula of the prior constraint is:

[0084]

[0085] where r is the reliability of the inversion parameter of the previous stage; and respectively represent the current estimated value of the jth material parameter at the kth iteration and the optimal value in the (k-1)th iteration; and respectively represent the upper and lower limits of the value of the jth material parameter; represents the real observed deformation feature; represents the deformation feature calculated using the material parameters in the (k-1)th iteration; ω is a weight coefficient.

[0086] As a specific embodiment, in step 4, the optimal material parameters of the current stage obtained in step 3 are used to regress the finite element calculation to calculate the predicted deformation feature of the next stage.

[0087] As a specific embodiment, a material parameter tracking inversion method based on information transmission provided by the present application is used to track and invert the material parameters of the dam body of the Jiangpinghe face rockfill dam, including the following steps:

[0088] S1, collect the dam body settlement monitoring data of a total of 73 monitoring points on three monitoring sections (R0+050.000, L0+010.000, L0+060.000) of the Jiangpinghe face rockfill dam, and use the adaptive wavelet transform filtering method to clean the data and filter the data noise. Then, based on the cloud generation model, the quality of the monitoring points is evaluated, and 29 high-quality typical monitoring points are selected to extract their deformation features.

[0089] S2, based on the test parameters and parameter sensitivity analysis, determine the material parameters to be inverted and their value range. Generate 500 groups of trial material parameter combination samples by Latin hypercube sampling method, and bring them into finite element calculation to obtain the dam body settlement calculation results, and extract the deformation features as training samples.

[0090] S3, based on the artificial neural network (ANN), the node space-time coordinates and the parameter combination are used as input variables, and the corresponding settlement is used as output to construct a proxy model of the space-time distribution of the settlement of the typical monitoring points of the Jiangpinghe face rockfill dam. Using this model to replace the finite element numerical calculation can effectively save the calculation resources, but the prediction accuracy requirement is higher. Therefore, the model prediction effect needs to be verified. As shown in Figure 2 , 3 indicated, the proxy model used in the present example has high accuracy.

[0091] S4, Figure 4A six-time inversion time node distribution diagram adopted by the embodiment is shown, starting from trial material parameters, bringing into finite element calculation, obtaining the error of the dam body deformation calculation result under the material parameters and the current stage monitoring data, and extracting the deformation characteristic calculation value and the measured value at each typical measuring point to calculate the determination coefficient R 2 as the inversion reliability coefficient. Subsequently, it is ensured that the sequentially inverted parameters are correlated with each other, and the distance between the current stage inversion parameter and the calculation effect (represented by the above R 2 ) of the previous stage inversion parameter (if it is the first stage, it is the test parameter) is determined, and the specific information transmission process is shown in Figure 5 .

[0092] S5, single inversion is carried out multiple times, and the results are sorted to take the mean value and 95% confidence interval. As shown in Figure 6 , it is a comparison diagram of the optimization process in the sixth stage in the embodiment of the application and the conventional method, which compares the inversion parameter results of the direct tracking inversion method and the method of the application. It can be seen that the method of the application can obtain a more reasonable change history of the material properties of the rockfill dam.

[0093] In order to further verify the effect of the method of the application, in the embodiment, the parameter results obtained by single optimization process are also displayed, Figure 7 , it is an evolution process diagram of the material parameters (K) obtained by the sixth stage inversion of the Jiangpinghe rockfill dam project in the embodiment of the application. (The E-B model is a constitutive model used for simulating the stress and strain of the rockfill dam, which is widely used in rockfill dam projects and is typical and experienced. Among them, the parameters inverted include the tangent modulus K and the bulk modulus Kb, which are material parameters that can be more intuitively judged for material strength.) The results show that the method of the application can effectively reduce the randomness of the optimization results.

[0094] In order to further verify the rationality of the parameters obtained by the method of the application, in the embodiment, the final inversion results are taken for a finite element calculation to obtain the deformation of the typical section (L0+010.000) of the Jiangpinghe face rockfill dam and the panel deflection at the full storage period (2022.06), as shown in Figure 8 , Figure 8 (a) and (b) are respectively the dam body settlement cloud chart and the horizontal displacement cloud chart of the typical section of the Jiangpinghe rockfill dam at the full storage period obtained by the embodiment of the application, Figure 8 (c) is the panel deflection cloud chart at that time; and the typical measuring points (measuring point numbers: SG1e-04 (upper) SG1d-05 (lower)) are selected, Figure 9 to show the fitting effect of the time history line of the calculation result and the measured value.

[0095] The implementation basis of each embodiment of the present application is achieved by a system with processor function programmed processing. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present application are packaged into various modules. Based on this actual situation, on the basis of the above embodiments, the embodiment of the present application provides an information transmission-based rockfill dam material parameter tracking inversion system, which is used to execute one of the information transmission-based rockfill dam material parameter tracking inversion methods in the above method embodiments.

[0096] Referring to Figure 10 The system comprises:

[0097] The deformation feature extraction module is configured to extract the deformation feature of the typical measuring point in the current stage in the rockfill dam multi-source monitoring data. The prior information acquisition module is configured to compare the deformation feature of the typical measuring point in the current stage and the predicted deformation feature in the previous stage, and calculate the inversion reliability coefficient in the previous stage. The tracking inversion module is configured to use the pre-trained surrogate model to perform inversion based on the deformation feature of the typical measuring point in the current stage, the optimal material parameter obtained by inversion in the previous stage and the inversion reliability coefficient, and perform rockfill dam material parameter optimization through the improved particle swarm algorithm to obtain the optimal material parameter in the current stage. The posterior information calculation module is configured to calculate the predicted deformation feature in the next stage according to the optimal material parameter in the current stage.

[0098] It should be noted that the system embodiments provided by the present application are used to implement the methods in the above method embodiments, and are also used to implement the methods in other method embodiments provided by the present application. The difference is only that the corresponding function modules are set, and the principle is basically the same as that of the above system embodiments provided by the present application. As long as the person skilled in the art improves the modules in the above system embodiments on the basis of the above system embodiments, refers to the specific technical solutions in other method embodiments, obtains the corresponding technical means by combining technical features, and the technical solutions composed of these technical means, as long as the technical solutions have practicality, the corresponding system class embodiments are obtained by improving the modules in the above system embodiments, which are used to implement the methods in other method class embodiments.

[0099] The method of the embodiment of the present application is implemented by relying on an electronic device, so it is necessary to introduce the related electronic device. For this purpose, the embodiment of the present application provides an electronic device, such as Figure 11As shown, the electronic device includes: at least one processor, a communication interface, at least one memory, and a communication bus, wherein the at least one processor, the communication interface, and the at least one memory communicate with each other via the communication bus. The at least one processor invokes logical instructions stored in the at least one memory to execute all or part of the steps of the methods provided in the foregoing method embodiments.

[0100] Furthermore, when the logical instructions in at least one of the aforementioned memories are implemented as software functional units and sold or used as independent products, they are stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, is embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (a personal computer, server, or network device) to execute all or part of the steps of the methods described in the various method embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks—various media for storing program code.

[0101] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, located in one place, or distributed across multiple network units. The purpose of this embodiment is achieved by selecting some or all of the modules according to actual needs. Those skilled in the art will understand and implement this without any inventive effort.

[0102] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] Based on the same technical concept as the foregoing embodiments, the present invention provides a non-transitory computer-readable storage medium that stores computer instructions that cause the computer to execute a method for tracking and inverting material parameters of a rockfill dam based on information transmission.

[0107] In summary, this invention discloses a method for tracking and inverting material parameters of rockfill dams based on information transfer. The core of this method lies in increasing the inversion frequency based on conventional parameter inversion methods, while employing information transfer constraints to ensure the integrity of the entire lifecycle tracking and inversion process of the rockfill dam. Specifically, the information transfer process in this invention refers to using dam deformation monitoring data acquired at each stage and parameter results acquired in previous stages as prior information, and using the dam deformation calculation effect under the material parameters acquired in the current stage as posterior information, to evaluate the weight of this set of parameters when used as prior information. The two are cross-transferred. The aim is to continuously track and invert the parameter model during the construction and operation of the rockfill dam, and to achieve reasonable transfer of dam deformation monitoring information and acquired rockfill dam material parameter information at each stage throughout the entire process. This enables serial updates of model parameters at each stage, ensuring the integrity of the dam's lifecycle deformation prediction model, while also enhancing the rationality of the dynamic changes in the acquired material parameters and their application value for dam safety assessment.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not 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 or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for tracking and inverting material parameters of rockfill dams based on information transmission, characterized in that, include: Extract deformation characteristics of typical measuring points at the current stage from multi-source monitoring data of rockfill dams; By comparing the deformation characteristics of typical measuring points at the current stage with the predicted deformation characteristics at the previous stage, the inversion reliability coefficient of the previous stage is calculated. Based on the deformation characteristics of typical measurement points at the current stage and the optimal material parameters and inversion reliability coefficients obtained from the previous stage, a pre-trained surrogate model is used for inversion, and the material parameters of the rockfill dam are optimized through an improved particle swarm optimization algorithm to obtain the optimal material parameters at the current stage. Based on the optimal material parameters at the current stage, calculate the predicted deformation characteristics for the next stage.

2. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 1, characterized in that, The process of obtaining the typical measuring points is as follows: the multi-source monitoring data of the rockfill dam is input into the cloud generation model in the form of measuring points. The cloud generation model evaluates the quality of the measuring points based on the numerical characteristics of the set typical measuring points and selects the typical measuring points.

3. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 1, characterized in that, The inversion reliability coefficient of the previous stage is numerically equivalent to the determination coefficient calculated based on the deformation characteristics of typical measuring points in the current stage and the predicted deformation characteristics of the previous stage. In the initial stage, the determination coefficient is calculated based on the deformation characteristics of typical measuring points in the initial stage and the deformation characteristics calculated by the three-dimensional finite element model of the rockfill dam based on the initial material parameters, thus obtaining the inversion reliability coefficient of the initial stage.

4. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 1, characterized in that, The pre-training process of the proxy model includes: Based on the preset range of material parameter values, design several sets of trial material parameters; A finite element three-dimensional model of the rockfill dam was constructed, and finite element seeding calculations were performed based on several sets of trial material parameters to extract deformation features from the calculation results. The trial parameters are used as input to the artificial neural network, and the deformation features in the calculation results are used as output to train the artificial neural network. The trained artificial neural network is then output as a surrogate model.

5. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 1, characterized in that, The improved particle swarm optimization algorithm introduces a clustering algorithm into the particle swarm optimization algorithm to cluster the particle swarm, obtain multiple particle clusters, and incorporate the optimal solution of the clustered particle clusters into the calculation formula of the conventional particle swarm optimization algorithm.

6. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 1, characterized in that, The steps for obtaining the optimal material parameters at the current stage are as follows: set up multiple sets of material parameters for rockfill dams for optimization, obtain the mean of the optimization results and the 95% confidence interval, and select the set of material parameters that falls within the 95% confidence interval of the optimization results and has the smallest value of the objective function as the optimal material parameters at the current stage.

7. The method for tracking and inverting material parameters of rockfill dams based on information transmission as described in claim 6, characterized in that, The objective function is mathematically represented as follows: ； In the formula, k is the inversion number, used to represent the current stage; j represents the material parameter number; p is the total number of material parameters; x j，k F represents the value of the j-th material parameter during the k-th inversion process; k (·) denotes the objective function for the current stage of inversion using prior information iterative supervision; f k (·) represents the basic objective function of the inversion at the current stage; g(k) represents the prior constraint at the current stage, which is calculated from the deformation characteristics of typical measurement points at the current stage and the optimal material parameters and inversion reliability coefficients obtained from the previous stage; w0 represents the weighting coefficient.

8. A system for tracking and inverting material parameters of rockfill dams based on information transmission, characterized in that, include: The deformation feature extraction module is used to extract the deformation features of typical measuring points in the current stage from the multi-source monitoring data of the rockfill dam. The prior information acquisition module is used to compare the deformation characteristics of typical measuring points in the current stage with the predicted deformation characteristics in the previous stage, and to calculate the inversion reliability coefficient of the previous stage. The tracking inversion module is used to perform inversion using a pre-trained surrogate model based on the deformation characteristics of typical measurement points at the current stage and the optimal material parameters and inversion reliability coefficients obtained from the previous stage. It also uses an improved particle swarm optimization algorithm to optimize the material parameters of the rockfill dam and obtain the optimal material parameters at the current stage. The posterior information calculation module is used to calculate the predicted deformation characteristics for the next stage based on the optimal material parameters of the current stage.

9. An electronic device, characterized in that, The method includes a memory and a processor, the memory storing program instructions that are executed by the processor, the processor invoking the program instructions to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions that cause the computer to perform the method described in any one of claims 1 to 7.

Images