Arch dam aging deformation mechanism analysis method

Abstract

The invention discloses an arch dam aging deformation mechanism analysis method, and relates to the field of hydraulic structure safety. The method comprises the steps that based on three-dimensional terrain and geological data, a high-precision arch dam-foundation overall finite element mesh model is established, concrete materials and permeation partitions are divided carefully, and an anti-seepage and drainage system structure is embedded; by means of multi-source monitoring data, real boundary conditions such as water temperature and air temperature and thermodynamic parameters are dynamically recognized and optimized through the intelligent inversion technology; respectively analyzing the aging influence of single factors such as a temperature field and valley amplitude deformation on arch dam deformation by adopting finite element simulation; on the basis of the normalized deformation response data, a Bayesian optimized long-short-term memory neural network and a hierarchical correlation propagation algorithm are adopted, contribution weights of all factors to deformation are calculated, and weight rationality is verified through global sensitivity analysis. According to the method, quantitative separation and evaluation of the main deformation influence effect of the arch dam are realized, and a theoretical basis can be provided for structural health monitoring and safety management.

Description

Technical Field

[0001] This invention relates to the field of hydropower engineering technology, and more specifically to a method for analyzing the time-dependent deformation mechanism of arch dams. Background Technology

[0002] With the large-scale construction of ultra-high arch dam projects in Southwest my country, representative projects such as Xilawa, Goupitan, Xiaowan, Xiluodu, Jinping I, and Dagangshan have been completed and put into operation. During long-term operation, the deformation of the arch dam structure is affected by the coupling of multiple factors such as temperature changes, concrete creep, seepage, and foundation creep, exhibiting obvious time-related characteristics.

[0003] Monitoring data shows significant differences in the time-dependent deformation patterns of different arch dams. For example, the radial displacement of the crown beam of the Ertan, Xiaowan, and Dagangshan arch dams shows a continuous downstream trend, while the Jinping I and Xiluodu projects exhibit an accumulated upstream displacement. Currently, a systematic theoretical framework for the causes of time-dependent deformation in arch dams has not yet been established, making it difficult to quantitatively assess the influence of factors such as temperature load, concrete creep, seepage, foundation creep, and valley deformation.

[0004] Therefore, this invention proposes a method for analyzing the time-dependent deformation mechanism of arch dams based on finite element multi-field coupled simulation. This method combines monitoring data inversion with single-factor decoupling technology to quantitatively analyze the contribution of each factor to the time-dependent deformation and assess their relative weights, thereby accurately identifying the dominant influencing path. The implementation of this invention can provide a scientific basis for the safety operation assessment, long-term health monitoring, and maintenance decision-making of ultra-high arch dams. Summary of the Invention

[0005] This invention proposes an analytical method for the time-dependent deformation mechanism of arch dams, aiming to address the difficulty in quantitatively assessing the time-dependent deformation mechanism of arch dams under the coupled effects of multiple factors in existing technologies. This method constructs a finite element model with multi-field coupling of temperature, seepage, and stress, performs parameter inversion using long-term monitoring data, and quantitatively evaluates the relative weights of key influencing factors such as temperature load, concrete creep, valley width deformation, and reservoir settlement through single-factor decoupling analysis. Ultimately, it can accurately identify the dominant mechanism of time-dependent deformation of arch dams, providing a scientific basis for engineering operation safety assessment and maintenance decisions.

[0006] To achieve the above objectives, this invention provides a method for analyzing the time-dependent deformation mechanism of arch dams, the method comprising:

[0007] Step 1: Establish a finite element mesh model and construct a high-precision three-dimensional overall finite element model of the arch dam-foundation. The model includes a three-dimensional topography of the dam area that reflects natural and artificial slopes, simulates the spatial distribution and geological structure characteristics of the main rock strata within the dam site, establishes a dam body model with zoning characteristics of concrete materials, clarifies the distribution of concrete of different strength grades, sets permeability coefficient zoning in the dam foundation area to reflect the seepage differences of the rock and soil, and incorporates the geometric structure and physical properties of the seepage prevention curtain and drainage system into the model.

[0008] Step 2: Inversion of the true boundary conditions and material parameters of the arch dam. Based on multi-source heterogeneous monitoring data obtained from the long-term operation of the dam, data-driven intelligent inversion technology is used to automatically identify and dynamically update boundary conditions such as upstream water temperature and air temperature, as well as key thermodynamic parameters of the arch dam. This includes: initializing the parameter range based on engineering experience, establishing the mapping relationship between external load and structural response through a finite element model, using a genetic algorithm for multi-objective optimization, with the goal of minimizing the residual between monitoring data and simulation results, dynamically correcting parameters such as thermal conductivity, specific heat capacity, and elastic modulus until the parameters meet the preset convergence criteria, and outputting the optimal boundary conditions and parameter combinations.

[0009] Step 3: Simulate and analyze the influence of temperature field on the time-dependent deformation of arch dam. Based on the constructed finite element mesh model and the real temperature boundary conditions obtained by inversion, establish a finite element simulation model that only considers the effect of temperature load, exclude the influence of loads such as seepage and external forces, and analyze the spatiotemporal evolution of temperature field and its effect on the deformation behavior of arch dam.

[0010] Step 4: Based on the overall finite element model and actual boundary conditions, apply the valley loading method obtained by inversion to the model, analyze the influence and characteristics of valley deformation on the response of the arch dam structure separately, without considering other load factors, and reproduce the spatiotemporal evolution process of valley deformation through numerical simulation.

[0011] Step 5: Perform separate working condition simulation analysis on each single factor, extract displacement response data of key parts of the arch dam, construct a multi-factor-multi-measuring-point response matrix, and normalize the data to eliminate the influence of dimensions.

[0012] Step 6: Construct a multi-factor coupled model based on a Bayesian optimized long short-term memory neural network. The input layer is the normalized single-factor response sequence, and the output layer is the measured total deformation. Calculate the contribution weight of each input factor through a hierarchical correlation propagation algorithm, and verify the rationality of the weights using global sensitivity analysis. Finally, normalize the weights to obtain the influence weights of each single factor on the arch dam deformation.

[0013] In one scheme, during the establishment of the finite element mesh model, the three-dimensional topography of the dam area is obtained by using multi-source data such as remote sensing measurement and geological exploration. Based on the actual rock strata distribution, the physical and mechanical properties of different rock strata are simulated by a partitioning assignment method. At the same time, corresponding material units are constructed for different design strength levels of the dam concrete.

[0014] In one approach, during the parameter inversion step, various types of historical monitoring data, such as temperature, stress, and deformation, are collected and integrated in a unified manner. An intelligent inversion algorithm is used to automatically identify thermodynamic boundary conditions and material parameters. The genetic algorithm adopts a parallel iterative approach and dynamically adjusts parameters using a multi-objective optimization framework.

[0015] In one approach, the temperature field analysis step establishes a simulation model that only considers the effect of temperature load based on the actual temperature boundary conditions obtained by inversion, eliminating the influence of seepage, earthquake and external load, and analyzing the thermal expansion and contraction and thermal stress response process induced by the simulation of the spatiotemporal distribution of the temperature field in the dam body and dam foundation.

[0016] In one approach, during the valley loading condition analysis, a separation method is used to load the valley deformation single factor onto the overall finite element model. The real valley deformation sequence obtained by inversion is used as input to analyze its response characteristics at key locations such as the dam crest, dam abutment, dam heel, and dam foundation interface, and to identify the transmission path and deformation influence area of ​​the valley deformation within the dam body.

[0017] In one approach, the construction of the multi-factor-multi-measuring-point response matrix involves extracting deformation response data of each key measuring point of the arch dam under the action of a single factor, such as temperature, valley width, and seepage, and preprocessing the response data using a normalization method to eliminate dimensional interference between different physical quantities.

[0018] In one approach, during the multi-factor coupled weight analysis process, a long short-term memory neural network with a Bayesian optimization mechanism is used to train the normalized single-factor response data. The network outputs the measured total deformation response, and the contribution of each single factor is quantified using a hierarchical correlation propagation algorithm. Furthermore, the consistency and rationality of the weights of each factor are verified using a global sensitivity analysis method. Finally, the weights of each factor are normalized to obtain the relative contribution ratio.

[0019] In one approach, the single-factor weighted data obtained through the above steps are used to provide a scientific basis for decision-making regarding long-term safety monitoring of arch dams, tracing the causes of defects, and optimizing control measures.

[0020] Beneficial effects of this invention:

[0021] This invention proposes a systematic mechanism analysis method for the time-dependent deformation of arch dams. Compared with existing technologies, it comprehensively considers the influence and weight of multiple factors on time-dependent deformation. At the technical implementation level, based on multi-source monitoring data, a data-driven intelligent inversion method is employed to accurately obtain the dam's air temperature boundary conditions, water temperature boundary conditions, and key mechanical and thermal parameters. Through finite element simulation, the influence of single factors such as temperature field, concrete creep, valley width deformation, and reservoir settlement on the deformation of the arch dam during its water storage operation is systematically analyzed, and the influence weight of each factor is quantified, thus revealing the intrinsic mechanism of arch dam deformation more comprehensively. Attached Figure Description

[0022] Figure 1 This is the overall technical roadmap of the present invention.

[0023] Figure 2 This is a geometric overall model diagram of an embodiment of the present invention.

[0024] Figure 3 This is a finite element network dam model diagram of Embodiment 1 of the present invention.

[0025] Figure 4 This is a schematic diagram of the radial deformation process of dam section #22 in Embodiment 1 of the present invention. Detailed Implementation

[0026] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0027] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. To facilitate understanding, the invention will now be described more fully with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to make the disclosure of the invention more thorough and complete.

[0028] See Figures 1-4 This invention provides a method for analyzing the time-dependent deformation mechanism of arch dams, comprising the following steps:

[0029] S1: Establish the finite element mesh model

[0030] A high-precision three-dimensional finite element model of the arch dam and its foundation should be constructed, and the model should possess the following key technical features: 1) Establish a three-dimensional topographic model of the dam area that includes both natural and artificial slopes to realistically reflect the geomorphological features of the dam site; 2) Based on geological exploration data, simulate the spatial distribution of the main rock strata and key geological structural features within the dam site area; 3) Construct a dam body model with zoning characteristics of concrete materials to clarify the distribution and arrangement of concrete of different strength grades in the dam body; 4) Set permeability coefficient zoning for the dam foundation area to reflect the seepage differences of each layer of rock and soil in the foundation; 5) Incorporate the geometric structure and physical properties of the seepage prevention curtain and drainage system into the model.

[0031] In step S1, based on the actual 3D topographic mapping data of the arch dam project, the 3D topography of the dam area, encompassing both natural and artificial slopes, is reconstructed using Geographic Information System (GIS) and Digital Elevation Model (DEM) technologies. This provides the geomorphic foundation for subsequent finite element model construction. The 3D topography is then divided into zones to ensure the model accurately reflects the complex and varied geomorphic features of the dam site. Building upon this, combined with geological exploration results and borehole profile data, a layered modeling method is employed to precisely map the main rock strata structures and their spatial distribution within the dam site area into the 3D model. Simultaneously, parameters are assigned according to major geological structures (such as faults and fracture zones) to ensure the accurate depiction of the geometric and physical properties of key geological units and weak structural surfaces. For the dam body itself, multi-region solid modeling embeds concrete zones of different strength grades and mix proportions into the overall dam structure, assigning them corresponding physical and mechanical parameters, such as elastic modulus E, Poisson's ratio ν, density ρ, and linear expansion coefficient α, ensuring the heterogeneous distribution of the dam material remains consistent with the actual engineering.

[0032] When modeling the permeability characteristics of the dam foundation area, the foundation soil and rock mass is divided into several permeability coefficient zones based on measured hydrogeological parameters, and each zone is assigned a different permeability coefficient k. i In the finite element model, the seepage behavior of each element can be described by Darcy's law q = -k▽h, where q is the seepage velocity, k is the permeability coefficient, and h is the hydraulic head. Furthermore, to simulate the actual engineering layout of the anti-seepage curtain and drainage system, based on design drawings and construction records, the anti-seepage curtain is incorporated into the model as a physical boundary of a continuous or discrete surface, and is given low permeability parameters and finite thickness. The drainage system is embedded in the foundation and dam body areas as a solid / pipeline, with its geometric structure (such as size and distribution) and material parameters (such as drainage capacity) set to reflect its actual function in the seepage and stress fields. For the generation of the overall finite element mesh, an adaptive refinement strategy is adopted, with mesh refinement processing performed in areas such as the dam body, the interface between the dam foundation and the structure, weak zones, the curtain, and the drainage system. This improves the analytical accuracy of the model for key detailed areas while ensuring the numerical stability and convergence of the overall calculation. The final three-dimensional finite element overall model can be expressed by the following matrix:

[0033] Ku = F

[0034] Where K represents the overall stiffness matrix, incorporating the effects of heterogeneous materials, different infiltration zones, and the curtain / drainage system; u is the nodal displacement vector; and F is the equivalent load vector under the combined effects of external loads and various boundary conditions. This model provides a high-precision, fine-scale mechanical and seepage field foundation for subsequent simulations of various working conditions and analysis of time-dependent deformation mechanisms.

[0035] S2: Inversion of actual boundary conditions and material parameters for arch dams:

[0036] Based on the multi-source heterogeneous monitoring data accumulated from the long-term operation of the dam, data-driven intelligent inversion technology is used to realize the automatic identification and dynamic updating of upstream water temperature boundary, air temperature boundary conditions and key thermodynamic parameters of the dam body. The specific implementation process includes: (1) Parameter initialization: set reasonable range of material parameters such as linear expansion coefficient and elastic modulus according to engineering experience; (2) Model construction: establish the mapping relationship between external load input and structural mechanical response through finite element numerical model; (3) Intelligent inversion: construct a multi-objective optimization framework based on genetic algorithm, take minimizing the residual between monitoring data and simulation results as the optimization objective, and dynamically correct thermal parameters such as thermal conductivity coefficient and specific heat capacity and mechanical parameters such as elastic modulus through iterative calculation; (4) Parameter convergence: when the inversion parameters meet the preset convergence criteria, output the optimal combination of boundary conditions and material parameters.

[0037] In step S2, to obtain the actual boundary conditions and material parameters of the arch dam that conform to the engineering reality, it is first necessary to fully organize and preprocess the multi-source heterogeneous monitoring data accumulated during the long-term service of the dam, including but not limited to the series of internal and external temperatures, air temperature, water temperature, dam stress-strain, concrete and rock displacement, and seepage pressure. Based on historical engineering experience and relevant specifications, a reasonable range of values ​​for material thermodynamic parameters (such as linear expansion coefficient α, thermal conductivity λ, specific heat capacity c, elastic modulus E, etc.) is initially determined, denoted as [p min ,p max ], and generate an initial parameter population based on this.

[0038] Based on this, a finite element simulation platform was invoked, and a three-dimensional overall finite element model established using S1 was applied. Using the parameter vector p as input, preliminary boundary conditions such as upstream water temperature and air temperature were applied to perform a multi-field (thermo-mechanical) coupled numerical simulation. In each iteration step, the finite element model solved for the structural response satisfying the relationship K(p)u=F(t) based on material properties and external loads, where K(p) is the parameter-dependent system stiffness matrix, u is the simulated displacement or temperature field response, and F(t) is the time-varying load vector. The numerical simulation results of the dam's key location responses were then compared with actual monitoring data d. obsPerform the comparison and calculate the residual sequence ε(p) = d. sim (p)-d obs Using the sum of squared residuals as the objective function:

[0039]

[0040] Where N is the total number of observation data, d sim,i and d obs,i These are the simulated value and the measured value of the i-th monitoring point, respectively.

[0041] Subsequently, a genetic algorithm (GA) is introduced, using the parameter vector p as chromosomes. The population is continuously evolved through selection, crossover, and mutation operations, and the fitness of each parameter group is evaluated using the objective function J(p). In each generation, the globally optimal individual with high fitness is selected, and a new generation of parameter combinations is generated through genetic operations. For multi-objective optimization, different types of monitoring responses (such as temperature field response and displacement field response) can be considered simultaneously, and the objective function is extended to weighted residual minimization.

[0042]

[0043] Where M is the number of monitoring data types, w j Let N be the weight of the j-th class of data. j Let be the number of observation points of type j.

[0044] Through continuous iteration, the genetic algorithm automatically adjusts its parameters, causing the objective function value to gradually converge. The criterion can be set as the rate of change of the objective function |J|. (n) -J (n-1) |<ò or parameter vector Euclidean distance‖p (n) -p (n-1) || < δ is less than a predetermined threshold. Once the convergence criterion is met, the current optimal set of parameters p is output. * The combination of these parameters with the corresponding boundary conditions constitutes the true thermodynamic boundary conditions and material property parameters of the arch dam finite element model. This parameter set not only accurately reflects the actual service state of the arch dam structure but also lays a solid foundation for subsequent multi-field coupled simulation and analysis of aging deformation mechanisms.

[0045] S3: Simulation analysis of the influence of temperature field on the time-dependent deformation of arch dams

[0046] Based on the overall finite element mesh model of the arch dam-foundation constructed in step S1 and the actual temperature boundary conditions of the arch dam obtained by inversion in step S2, a finite element simulation analysis model considering only the effect of temperature load is established. In this model, the effects of seepage, external forces and other loads are excluded, and the spatiotemporal evolution of the temperature field is analyzed.

[0047] In step S3, the high-precision three-dimensional integral finite element mesh model of the arch dam-foundation system established in S1 is first used as the computational basis. Simultaneously, temperature load conditions, such as the actual upstream water temperature boundary and air temperature boundary obtained through intelligent inversion in stage S2, are introduced. To focus on analyzing the individual influence of the temperature field on the time-dependent deformation of the arch dam, seepage, external concentrated loads, and other mechanical effects generated during operation are strictly excluded during modeling, retaining only the thermal loads caused by the external ambient temperature and water temperature. In the model settings, temperature is applied as a time-varying boundary condition to the relevant surfaces of the dam-foundation system, and the heat transfer process follows the heat conduction governing equation. Where ρ is the material density, c is the specific heat capacity, λ is the thermal conductivity, T is the temperature field, and t is time. After simulating the evolution of the temperature field over time, the temperature field results are used as input for thermal stress analysis. The temperature stress generated by thermal expansion and contraction is expressed through the thermal expansion relationship ε of the material. T =αΔT is transformed into the strain response inside the dam body, and then the stiffness relationship Ku=F in the finite element model is used. T (where F) T The temperature-dependent deformation at key structural locations is calculated using the equivalent thermal load vector. This analysis reveals the dynamic distribution characteristics of the temperature field in the spatial and temporal dimensions of the dam body and foundation, as well as the induced deformation response patterns, and provides a foundation for subsequent decomposition of multi-factor coupling effects.

[0048] S4: Simulation Analysis of the Influence of Concrete Creep on the Time-Release Deformation of Arch Dams

[0049] Based on the finite element model of the arch dam-foundation established in stage S1 and the actual boundary conditions determined in stage S2, a nonlinear creep constitutive model considering environmental factors, stress levels, and age effects is used for simulation analysis. By applying typical temperature fields, hydrostatic pressures, and construction period loads, coupled time-varying analysis is carried out to simulate the deformation evolution process of the arch dam during long-term service.

[0050] In step S4, based on the finite element mesh model obtained in S1 and the real environmental boundary and physical parameters obtained from the inversion in S2, the time-varying creep effect of concrete is further incorporated into the structural analysis. The concrete creep analysis employs a nonlinear creep constitutive relation that reflects the effects of environmental temperature and humidity, stress level, and age, such as the B3 or differential power function creep model, with creep strain ε... cr (t, t0) can be described as Where J(t,τ) is the creep compliance function, σ(τ) is the stress under load duration, and t0 is the initial loading age. In practical implementation, hydrostatic pressure and structural redistribution loads during construction, in addition to typical temperature fields, are applied to construct a multi-field coupled analysis environment for the interaction of temperature, load, and creep. Using the finite element time-step integration method, the progressive deformation evolution of the dam body from the construction period to long-term service under the influence of time-bound loads and the environment is simulated. The model automatically tracks the nonlinear deformation path of concrete at different ages, under different stress states, and under different environmental conditions (such as temperature and humidity changes), and outputs the creep deformation curves of key parts of the arch dam and their growth trends over time. This simulation not only reflects the contribution of creep to the time-bound deformation of the dam body but also reveals the structural response mechanism under the combined action of concrete and the foundation system, providing a scientific basis for the safety assessment and maintenance management of actual dams during long-term service.

[0051] S5: Simulation analysis of the impact of changes in hydrogeological conditions after water impoundment on the time-dependent deformation of the arch dam, including deformation of the dam foundation, reservoir, and valley floor.

[0052] Based on the finite element model of the arch dam-foundation established in stage S1 and the boundary conditions determined in stage S2, the influence of changes in hydrogeological conditions after impoundment on the time-dependent deformation of the arch dam is analyzed, with a focus on investigating the mechanism of action of factors affecting the settlement of the dam foundation and reservoir and the deformation of the valley floor.

[0053] S501: Valley Deformation Loading Inversion Method

[0054] Regression processing was performed on the monitoring data of valley deformation to extract its deformation trend over time, and corresponding displacement loads were applied to the truncation boundary of the model to simulate the valley contraction effect. Combining the measured deformation information of the dam body and valley, under the actual boundary conditions and loads determined in stage S2, simulation inversion analysis was conducted on the entire process of the arch dam, including construction, water impoundment, and operation. By minimizing the difference between the simulation results and the monitoring data, the optimal displacement loading distribution in the truncation boundary region was obtained.

[0055] S502: Simulation Study on the Effect of Upstream Reservoir Settlement and Deformation on Arch Dam Deformation

[0056] Based on the integrated finite element mesh of the arch dam and foundation constructed in S1 and the actual boundary conditions given in S2, this study integrates the nonlinear rheological characteristics of the dam foundation rock mass structure with the multi-field coupling effect of water-thermal-mechanical processes. Using nonlinear finite element analysis, a comprehensive coupled simulation of seepage, temperature, and stress fields over a large area within the dam site after impoundment is conducted. The study focuses on the combined effects of rock mass seepage evolution, temperature distribution changes caused by thermal effects, effective stress redistribution, and material property degradation on reservoir deformation under multi-field coupling, thereby revealing the influence of upstream reservoir settlement on arch dam deformation.

[0057] S503: Simulation Study on the Influence of Valley Deformation on Arch Dam Deformation:

[0058] Based on the overall finite element model of the arch dam-foundation established in S1 and the actual boundary conditions determined in S2, the valley loading method obtained from the inversion in S501 is applied to the overall model. The effect of valley deformation on the structural response of the arch dam is analyzed separately, without considering the influence of other loads. The spatiotemporal evolution process of valley deformation is reproduced through numerical simulation to explore its transmission mechanism and influence characteristics on the deformation behavior of the arch dam.

[0059] S6: Weighting Analysis of the Influence of Individual Factors on Arch Dam Deformation

[0060] Based on the separate simulation analysis of each single factor, displacement response data of key parts of the arch dam (arch crown, arch shoulder, dam heel, dam foundation interface, etc.) were extracted for the time-dependent deformation of the arch dam upstream and downstream, and a multi-factor-multi-measuring-point response matrix was constructed. The row vectors represent the deformation sequence of all measuring points under a single factor, while the column vectors represent the multi-factor coupled response of a single measuring point. Min-Max normalization is used to eliminate the influence of dimensions.

[0061]

[0062] Example 1:

[0063] A certain arch dam in southwestern my country has a maximum height of 294.5 meters. (For example...) Figure 4 As shown, with the increase in the service time of the dam, each dam section exhibits a trend of gradually increasing radial deformation towards the downstream, which has not yet fully converged. To address this phenomenon, it is necessary to study and analyze the causes of age-related deformation of arch dams, their future evolution patterns, and the impact of age-related deformation on the dam's operational performance.

[0064] like Figures 2-3 As shown, considering the actual structure and material zoning of the dam, a finite element mesh model of the arch dam foundation during the water storage operation period was constructed. The model has a total of 412,674 elements and 449,365 nodes, including the foundation, dam body, transverse joints and construction joints (which have been backfilled during the water storage operation period).

[0065] Based on the monitoring results of upstream thermometers at dam sections 22, 15, and 19 from 2006 to 2024, the monthly average temperature along elevation was statistically analyzed. The reservoir water temperature along depth was obtained by interpolation. In the simulation calculation, the surface temperature boundary between the dam and the air was taken according to the multi-year average air temperature.

[0066] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0067] It should be understood that the above detailed description of the technical solutions of the present invention with reference to preferred embodiments is illustrative and not restrictive. Those skilled in the art can modify the technical solutions described in the embodiments or make equivalent substitutions for some of the technical features based on reading this specification; however, these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the time-dependent deformation mechanism of arch dams, characterized in that: The method includes: Step 1: Establish a finite element mesh model and construct a high-precision three-dimensional overall finite element model of the arch dam-foundation. The model includes a three-dimensional topography of the dam area that reflects natural and artificial slopes, simulates the spatial distribution and geological structure characteristics of the main rock strata within the dam site, establishes a dam body model with zoning characteristics of concrete materials, clarifies the distribution of concrete of different strength grades, sets permeability coefficient zoning in the dam foundation area to reflect the seepage differences of the rock and soil, and incorporates the geometric structure and physical properties of the seepage prevention curtain and drainage system into the model. Step 2: Inversion of the true boundary conditions and material parameters of the arch dam. Based on multi-source heterogeneous monitoring data obtained from the long-term operation of the dam, data-driven intelligent inversion technology is used to automatically identify and dynamically update the upstream water temperature, air temperature boundary conditions and key thermodynamic parameters of the arch dam. This includes: initializing the parameter range according to engineering experience, establishing the mapping relationship between external load and structural response through a finite element model, using a genetic algorithm for multi-objective optimization, with the goal of minimizing the residual between monitoring data and simulation results, dynamically correcting the thermal conductivity, specific heat capacity, and elastic modulus parameters until the parameters meet the preset convergence criteria, and outputting the optimal boundary conditions and parameter combinations. Step 3: Simulate and analyze the influence of the temperature field on the time-dependent deformation of the arch dam. Based on the constructed finite element mesh model and the real temperature boundary conditions obtained by inversion, establish a finite element simulation model that only considers the effect of temperature load, exclude the influence of seepage and external force load, and analyze the spatiotemporal evolution of the temperature field and its effect on the deformation behavior of the arch dam. Step 4: Based on the overall finite element model and actual boundary conditions, apply the valley loading method obtained by inversion to the model, analyze the influence and characteristics of valley deformation on the response of the arch dam structure separately, without considering other load factors, and reproduce the spatiotemporal evolution process of valley deformation through numerical simulation. Step 5: Perform separate working condition simulation analysis on each single factor, extract displacement response data of key parts of the arch dam, construct a multi-factor-multi-measuring-point response matrix, and normalize the data to eliminate the influence of dimensions. Step 6: Construct a multi-factor coupled model based on a Bayesian optimized long short-term memory neural network. The input layer is the normalized single-factor response sequence, and the output layer is the measured total deformation. Calculate the contribution weight of each input factor through a hierarchical correlation propagation algorithm, and verify the rationality of the weights using global sensitivity analysis. Finally, normalize the weights to obtain the influence weights of each single factor on the arch dam deformation.

2. The method for analyzing the time-dependent deformation mechanism of arch dams according to claim 1, characterized in that: In the process of establishing the finite element mesh model, the three-dimensional topography of the dam area is obtained by using multi-source data such as remote sensing measurement and geological exploration. Based on the actual rock strata distribution, the physical and mechanical properties of different rock strata are simulated by using a partitioned assignment method. At the same time, corresponding material units are constructed for different design strength levels of the dam concrete.

3. The method for analyzing the time-dependent deformation mechanism of arch dams according to claim 1, characterized in that: In the parameter inversion step, historical monitoring data of various types, such as temperature, stress, and deformation, are collected and integrated in a unified manner. The intelligent inversion algorithm is used to automatically identify thermodynamic boundary conditions and material parameters. The genetic algorithm adopts a parallel iterative approach and dynamically adjusts parameters using a multi-objective optimization framework.

4. The method for analyzing the time-dependent deformation mechanism of arch dams according to claim 1, characterized in that: The temperature field analysis step establishes a simulation model that only considers the effect of temperature load based on the actual temperature boundary conditions obtained by inversion, eliminating the influence of seepage, earthquake and external load. By simulating the spatiotemporal distribution of the temperature field in the dam body and dam foundation, the thermal expansion and contraction and thermal stress response process induced by it are analyzed.

5. The method for analyzing the time-dependent deformation mechanism of an arch dam according to claim 1, characterized in that: In the valley loading condition analysis process, the separation method is used to load the valley deformation single factor into the overall finite element model. The real valley deformation sequence obtained by inversion is used as input to analyze its response characteristics at key parts of the dam crest, dam shoulder, dam heel and dam foundation interface, and to identify the transmission path and deformation influence area of ​​the valley deformation in the dam body.

6. The method for analyzing the time-dependent deformation mechanism of an arch dam according to claim 1, characterized in that: In the construction of the multi-factor-multi-measuring-point response matrix, for the effects of multiple single factors such as temperature, valley width, and seepage, the deformation response data of each key measuring point of the arch dam under the action of a single factor are extracted, and the response data are preprocessed by normalization method to eliminate the dimensional interference between different physical quantities.

7. The method for analyzing the time-dependent deformation mechanism of arch dams according to claim 1, characterized in that: In the multi-factor coupled weight analysis process, a long short-term memory neural network with a Bayesian optimization mechanism is used to train the normalized single-factor response data. The network outputs the measured total deformation response, and the contribution of each single factor is quantified using a hierarchical correlation propagation algorithm. The consistency and rationality of the weights of each factor are verified by a global sensitivity analysis method. Finally, the weights of each factor are normalized to obtain the relative contribution ratio.

8. The method for analyzing the time-dependent deformation mechanism of an arch dam according to claim 1, characterized in that: The single-factor weighted data obtained by analyzing the above steps provide a scientific basis for long-term safety monitoring of arch dams, tracing the causes of defects, and optimizing control measures.

Images