Digital twinning enabling dam body self-adaptive early warning method and system

Abstract

The invention relates to a digital twinborn enabling dam body self-adaptive early warning method and system, and belongs to the technical field of hydraulic engineering safety monitoring, and the method comprises the following steps: constructing a digital twinborn model of a dam body; performing multi-source data fusion; the operation state of the dam body is simulated in real time, and a simulation result and actual monitoring data are compared and analyzed; establishing a dynamic early warning threshold model; when the monitoring data or the simulation analysis result exceeds an early warning threshold value, early warning information is sent out, and the early warning information is fed back to the digital twinborn model to be updated and optimized; the method has the beneficial effects that the early warning threshold value is automatically adjusted according to the historical operation data, the real-time monitoring data and the analysis simulation result of the dam body, the contribution of the historical operation data, the real-time monitoring data and the analysis simulation result is flexibly set by adjusting the weight and the coefficient of each part, and the early warning accuracy is improved. And the accuracy and the reliability of the early warning threshold are further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water conservancy project safety monitoring, and in particular relates to a digital twin-enabled dam adaptive early warning method and system. Background Art

[0002] Dam safety is crucial in water conservancy projects. Traditional dam safety monitoring and early warning methods often rely on limited sensor data and static threshold judgments. With climate change and the increasing operational life of water conservancy projects, the risks faced by dams are becoming more complex and dynamic, making traditional methods less adaptable to these dynamic conditions. For example, under extreme weather conditions, seepage and stress-strain parameters in dams can fluctuate beyond expectations. Traditional early warning systems are unable to accurately and timely adjust their warning strategies based on actual conditions, significantly compromising the accuracy and timeliness of warnings.

[0003] Digital twin technology, an emerging technology, has been applied in various fields in recent years. However, while some research has attempted to incorporate digital twin technology into dam safety early warning, most efforts have simply constructed digital models to visualize the status of the dam, failing to fully leverage the technology's advantages in data fusion, real-time simulation, and adaptive early warning. Utilizing digital twin technology to build a system capable of real-time sensing of dam status, dynamically adjusting early warning thresholds, and accurately and promptly issuing warnings remains a pressing challenge. Summary of the Invention

[0004] The present invention provides a digital twin-enabled adaptive dam early warning method and system, which is used to solve the technical problems of insufficient accuracy and timeliness of existing dam early warning methods. By establishing a dynamic early warning threshold model, the early warning threshold is automatically adjusted according to the dam's historical operation data, real-time monitoring data and analysis simulation results, and the early warning threshold is flexibly set, so that it can not only detect potential safety problems in a timely manner, but also avoid excessive false alarms due to excessive sensitivity. As the dam's operation time passes and new data continues to accumulate, the model can regularly update parameters according to actual conditions, further improving the accuracy and reliability of the early warning threshold.

[0005] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0006] A digital twin-enabled adaptive early warning method for dam bodies includes the following steps:

[0007] Construct a digital twin model of the dam, integrating its geometry, structural information, material properties, and geological conditions;

[0008] Deploy a variety of sensors to collect data on seepage, deformation, stress and strain of the dam body, as well as meteorological and hydrological external environmental data, and perform multi-source data fusion;

[0009] Based on the digital twin model of the dam and the integrated multi-source data, numerical simulation algorithms are used to simulate the dam's operating status in real time, and the simulation results are compared and analyzed with the actual monitoring data;

[0010] Establish a dynamic warning threshold model to automatically adjust the warning threshold based on the dam's historical operation data, real-time monitoring data, and analysis and simulation results;

[0011] When the monitoring data or simulation analysis results exceed the warning threshold, a warning message is issued and fed back to the digital twin model for updating and optimization.

[0012] Optionally, in the digital twin model of the dam body, the geometric shape includes: spatial coordinate calculation, area and volume calculation; structural information includes: stress analysis, material anisotropy and deformation coordination calculation; material properties include: material constitutive relationship, viscoelastic material and material damage model; geological conditions include: seepage analysis and geomechanics model.

[0013] Optionally, multi-source data fusion is provided The data collected by different types of sensors are recorded as , the corresponding weight is ,and , the fused data Calculated by the following formula:

[0014] ;

[0015] For seepage data , deformation data , stress-strain data , meteorological data and hydrological data , if the weights of seepage data, deformation data, stress-strain data, meteorological data and hydrological data are 、 、 、 and , the fused data is:

[0016] .

[0017] Optional, numerical simulation algorithm is, data comparison and analysis:

[0018] The root mean square error (RMSE) can comprehensively reflect the overall deviation between the simulated value and the actual monitored value. The formula is:

[0019] ;

[0020] in, is the number of data points, Indicates the analog values, Indicates the Actual observations, This is more noticeable for larger errors because the errors are first squared, summed, and then squared.

[0021] Optional, specific calculation steps:

[0022] Calculate the difference between the simulated value and the actual observed value of each data point, that is, ;

[0023] Squaring the difference, we get , which is convenient for eliminating the influence of the positive and negative signs of the differences, while amplifying the influence of large differences and highlighting the large deviations between the simulation results and the actual observations;

[0024] The sum of the squared differences of all data points is: ;

[0025] Then divide the sum by the number of data points , and the mean square error is: ;

[0026] Taking the square root of the mean square error gives the root mean square error .

[0027] Optionally, a dynamic warning threshold model is established as follows:

[0028] Historical operation data part: calculate the mean value of the indicator in the historical operation data and standard deviation ;

[0029] Assume the weight of historical data is , then the contribution of historical operation data to the warning threshold is: ,in, It is a coefficient determined based on the statistical characteristics of historical data and engineering experience, and is used to adjust the degree of leniency of early warning;

[0030] Real-time monitoring data part:

[0031] Calculate the mean of the current real-time monitoring data and standard deviation ;

[0032] Assume the weight of real-time data is , then the contribution of real-time monitoring data to the warning threshold , is similar The coefficient is determined according to the characteristics and importance of real-time data;

[0033] Analysis of simulation results:

[0034] The theoretical expected value of this indicator under the current working conditions is obtained from the analysis and simulation and the possible range of fluctuations ;

[0035] Assume the weight of the analysis simulation results is , then analyze the contribution of simulation results to the warning threshold , It is a coefficient determined based on the reliability of the simulation and engineering requirements;

[0036] Comprehensive warning threshold calculation:

[0037] Final dynamic warning threshold The weighted sum of these three parts is: ;

[0038] Among them, the weight 、 、 satisfy .

[0039] Optionally, the steps for issuing a warning message are:

[0040] Data comparison and judgment: Real-time comparison of monitoring data, simulation analysis results and dynamic warning thresholds;

[0041] Determine the warning level: Determine the warning level according to the degree of exceeding the threshold and the importance of relevant indicators in accordance with predetermined rules;

[0042] Generate warning content: The warning information should contain detailed content;

[0043] Send early warning notifications: Send the generated early warning information to the system in a timely manner through multiple channels.

[0044] Optionally, feedback to the digital twin model for updating and optimization is:

[0045] Data transmission and reception: Detailed monitoring data and simulation analysis results that trigger early warnings are transmitted to the database of the digital twin model;

[0046] Model analysis and diagnosis: After the digital twin model receives data, it conducts detailed analysis of data that exceeds the warning threshold;

[0047] Model parameter adjustment: Adjust and update the relevant parameters of the digital twin model based on the results of analysis and diagnosis;

[0048] Model verification and evaluation: After completing parameter adjustments, use the updated data to verify and evaluate the digital twin model;

[0049] Visualization and reporting: The updated and optimized digital twin model is presented in a visual manner.

[0050] A digital twin-enabled adaptive dam early warning system, including:

[0051] Data analysis module, used to simulate the dam's operating status in real time, and compare and analyze simulation results with actual monitoring data;

[0052] The second model building module is used to automatically adjust the warning threshold through the historical operation data of the dam, real-time monitoring data and analysis simulation results;

[0053] Early warning module, used to issue early warning information;

[0054] The data analysis module is connected to the second model construction module, and the second model construction module is connected to the early warning module.

[0055] Beneficial effects of the present invention:

[0056] The present invention establishes a dynamic early warning threshold model to automatically adjust the early warning threshold according to the historical operation data of the dam body, real-time monitoring data and analysis simulation results. The contributions of the three parts of historical operation data, real-time monitoring data and analysis simulation results are adjusted by adjusting the weights and coefficients of each part to flexibly set the early warning threshold, so that it can not only detect potential safety problems in a timely manner, but also avoid excessive false alarms due to excessive sensitivity. As the dam body operates for a long time and new data is continuously accumulated, the model can regularly update parameters according to actual conditions to further improve the accuracy and reliability of the early warning threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 Schematic diagram of the system structure of the present invention;

[0059] Figure 2 It is the workflow diagram of the present invention. DETAILED DESCRIPTION

[0060] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0061] Example 1;

[0062] like Figure 1 As shown, this embodiment provides a digital twin-enabled dam adaptive early warning system, including:

[0063] The first model building module is used to build a digital twin model of the dam body;

[0064] Multi-source data fusion module, used for multi-source data fusion;

[0065] Data analysis module, used to simulate the dam's operating status in real time, and compare and analyze simulation results with actual monitoring data;

[0066] The second model building module is used to automatically adjust the warning threshold through the historical operation data of the dam, real-time monitoring data and analysis simulation results;

[0067] Early warning module, used to issue early warning information;

[0068] The first model construction module is connected to the multi-source data fusion module, the multi-source data fusion module is connected to the data analysis module, the data analysis module is connected to the second model construction module, and the second model construction module is connected to the early warning module.

[0069] The digital twin model of the dam body is constructed through the first model construction module. After the multi-source data fusion module fuses the multi-source data, the data analysis module simulates the operating status of the dam body in real time. The second model construction module automatically adjusts the warning threshold, and the warning module finally issues a warning message.

[0070] Example 2;

[0071] Based on Example 1, Figure 2 As shown, this embodiment provides a digital twin-enabled dam adaptive early warning method, comprising the following steps:

[0072] Construct a digital twin model of the dam, integrating its geometry, structural information, material properties, and geological conditions;

[0073] Deploy a variety of sensors to collect data on seepage, deformation, stress and strain of the dam body, as well as meteorological and hydrological external environmental data, and perform multi-source data fusion;

[0074] Based on the digital twin model of the dam and the integrated multi-source data, numerical simulation algorithms are used to simulate the dam's operating status in real time, and the simulation results are compared and analyzed with the actual monitoring data;

[0075] Establish a dynamic warning threshold model to automatically adjust the warning threshold based on the dam's historical operation data, real-time monitoring data, and analysis and simulation results;

[0076] When the monitoring data or simulation analysis results exceed the warning threshold, a warning message is issued and fed back to the digital twin model for updating and optimization.

[0077] Example 3;

[0078] Based on Example 2, the digital twin model of the dam body is:

[0079] (1) Geometric shape:

[0080] Spatial coordinates: any point on the surface or inside the dam The coordinates must satisfy the geometric constraint equations of the dam body. For example, for a simple rectangular dam body, the constraint equation is , , ,in, As the dam chief, is the dam width, is the dam height. For more complex shapes, such as arch dams, parametric equations are needed to represent them. For example, the horizontal cross-section equation of an elliptical arch dam can be expressed as ,in, 、 are the semi-major and semi-minor axes of the ellipse, respectively, and 、 It's about Dam Gao function.

[0081] Area and volume calculations:

[0082] Triangular surface element area: For the triangular surface element on the dam surface, if the coordinates of the three vertices are known , , , then its area ,in, , .

[0083] Complex shape volume: For the calculation of the volume of a dam with a complex shape, a numerical integration method is used. For example, the dam is discretized into multiple thin layers along a certain direction (which can be the height direction). The volume of each thin layer is approximately the average cross-sectional area of ​​the layer multiplied by the thickness. Then, the volumes of all thin layers are summed, i.e. ,in, For the The average cross-sectional area of ​​the layer, For the The thickness of the layer.

[0084] (2) Structural information:

[0085] Stress analysis: Three-dimensional stress state: The stress state of a point can be expressed by the stress tensor express, , corresponding to 、 、 direction. In the Cartesian coordinate system, the stress equilibrium equation is: , , ,in, 、 、 is the amount of physical force per unit volume.

[0086] Considering material anisotropy: For anisotropic materials, the stress-strain relationship needs to be expressed in a complex matrix form, such as: ,in, is the stress vector, is the strain vector, is the stiffness matrix of the material, and the elements of the stiffness matrix depend on the anisotropic properties of the material.

[0087] Deformation coordination equation: In the case of small deformation, the relationship between strain and displacement is , , , , , ,in, 、 and They are 、 and The displacement component in the direction, is the shear strain.

[0088] (3)Material properties:

[0089] Material constitutive relations: Elastoplastic materials: Incremental constitutive relations, such as the Prandtl-Reuss equation, ,in, is the strain increment, is the stress increment, is the elastic modulus, is Poisson's ratio, is the Kronecker symbol, is the plastic multiplier, is the yield function.

[0090] Viscoelastic materials: Commonly used models include the Kelvin-Voigt model, and the constitutive relationship is: ,in, is stress, For strain, is the elastic modulus, is the viscosity coefficient.

[0091] Material damage model: For example, damage model based on continuum damage mechanics, introducing damage variables To describe the degree of damage to the material, the stress-strain relationship can be expressed as , damage variable The damage evolution equation is usually determined according to the specific damage mechanism. For example, the damage evolution equation based on the energy release rate is ,in, are parameters related to material properties, is the equivalent strain.

[0092] (4) Geological conditions:

[0093] Seepage analysis: Three-dimensional seepage equation: In anisotropic media, seepage satisfies the extended form of Darcy's law. , , ,in, 、 、 is the seepage velocity component, are the elements of the permeability tensor, is the water head. At the same time, the continuity equation is satisfied , substituting Darcy's law into the continuity equation, we can get the basic seepage equation ;

[0094] Consider unsaturated seepage: For unsaturated soil, the seepage coefficient is saturation The function of , commonly used models include Van Genuchten model, ,in, is the saturated permeability coefficient, 、 Parameters related to soil properties.

[0095] Specifically, the geomechanical model:

[0096] Mohr-Coulomb strength criterion: used to describe the failure conditions of soil or rock. ,in, is the shear strength, For cohesion, is the normal stress, is the internal friction angle. Under three-dimensional stress state, it can be expressed as: ,in, 、 、 The principal stress.

[0097] Duncan-Zhang model: a commonly used soil constitutive model, whose tangent elastic modulus is and tangent Poisson's ratio The expression is: , ,in, 、 、 、 、 and are model parameters, is atmospheric pressure.

[0098] Example 4;

[0099] Based on Example 2, multi-source data fusion is used to fuse dam monitoring data and external environment data:

[0100] With The data collected by different types of sensors are recorded as , the corresponding weight is ,and , the fused data Calculated by the following formula:

[0101] ;

[0102] For seepage data , deformation data , stress-strain data , meteorological data and hydrological data , if the weights of seepage data, deformation data, stress-strain data, meteorological data and hydrological data are 、 、 、 and , the fused data is:

[0103] ;

[0104] Weighting allows different types of data to be assigned different levels of importance for describing the dam's condition and analyzing its safety. For example, dam deformation data may be directly related to the stability of the dam structure, making it relatively important and therefore warranting a higher weight. Meanwhile, minor factors in meteorological data have a relatively minor impact on the dam's condition and can be assigned lower weights. By properly assigning weights, the role of key data can be highlighted, ensuring that the fusion results focus more on information that is crucial to dam safety.

[0105] Example 5;

[0106] Based on Example 2, the specific numerical simulation algorithm is:

[0107] Data comparison and analysis:

[0108] The root mean square error (RMSE) can comprehensively reflect the overall deviation between the simulated value and the actual monitored value. The formula is:

[0109] ;

[0110] in, is the number of data points, Indicates the analog values, Indicates the Actual observations, This is more noticeable for larger errors because the errors are first squared, summed, and then squared.

[0111] Calculation steps:

[0112] First, the difference between the simulated value and the actual observed value of each data point is calculated, that is, .

[0113] Then, squaring these differences yields , which makes it easy to eliminate the influence of the positive and negative signs of the differences, while amplifying the influence of larger differences and highlighting the larger deviations between the simulation results and the actual observations.

[0114] Next, the squared differences of all data points are summed, i.e. .

[0115] Then divide the sum by the number of data points , and get the mean square error, that is .

[0116] Finally, take the square root of the mean squared error to get the root mean squared error ,The square root is taken to restore the dimension of the error to the same dimension as the original data, which facilitates intuitive understanding and comparison.

[0117] The characteristics are, Giving greater weight to larger errors can more sensitively reflect the larger deviations between the simulated values ​​and the actual observed values. Because the errors are squared before other operations are performed, even if only a few data points have large errors, it will cause Significantly increased.

[0118] In the dam operation state simulation, The accuracy and reliability of digital twin models can be evaluated by comparing different models or different parameter settings. , select the model and parameters that can most accurately simulate the actual operating status of the dam. In addition, It can also be used to monitor the deviation between the simulation results and the actual monitoring data during the operation of the dam, and to promptly detect possible abnormal conditions. A sudden increase means that there is a new problem with the dam or the model needs to be recalibrated.

[0119] Example 6;

[0120] Based on Example 2, the dynamic warning threshold model is established as follows:

[0121] Historical operation data section:

[0122] First calculate the mean of the indicator in the historical running data and standard deviation .

[0123] Assume the weight of historical data is , then the contribution of historical operation data to the warning threshold is: ,in, It is a coefficient determined based on the statistical characteristics of historical data and engineering experience, and is used to adjust the degree of leniency of early warning.

[0124] mean It reflects the average level of key monitoring indicators during the long-term operation of the dam body in the past, and represents the typical value of this indicator under normal working conditions. It measures the degree of dispersion of historical data around the mean and reflects the fluctuation of the indicator. A larger standard deviation means that the indicator has a larger range of variation in history, and there may be some special working conditions or abnormal situations.

[0125] Weight Indicates the relative importance of historical operating data in determining the warning threshold. If the historical data is rich and reliable and can well reflect the long-term operating characteristics of the dam, then Can take larger value. It is determined based on the statistical characteristics of historical data and engineering experience, and is used to adjust the degree of leniency of early warning. The larger the value, the higher the warning threshold and the lower the tolerance for abnormal situations, and vice versa. For example, if the dam has experienced some situations in the past that did not lead to damage but were close to dangerous conditions, then the value can be appropriately increased. , so that timely warnings can be issued when similar situations occur again.

[0126] Real-time monitoring data part:

[0127] Calculate the mean of the current real-time monitoring data and standard deviation (It can be data within the most recent window).

[0128] Assume the weight of real-time data is , then the contribution of real-time monitoring data to the warning threshold , is similar The coefficient is determined according to the characteristics and importance of real-time data.

[0129] mean It is the average value of the current real-time monitoring data within a certain time window, which reflects the current instantaneous status of the dam. It reflects the fluctuation of real-time data and can capture the instantaneous changes in the dam state in a timely manner. Different from the standard deviation of historical data, the standard deviation of real-time data focuses more on reflecting the stability and change trend of the dam in the current short term.

[0130] Weight The real-time monitoring data determines the impact of the early warning threshold. Since the real-time monitoring data can directly reflect the actual operation status of the dam, it is crucial for timely detection of potential dangers. Will take a moderate or large value. The role and similar, It is determined based on the characteristics and importance of real-time data. For example, if the real-time monitoring system has high accuracy and strong data reliability, then it can be adjusted appropriately. , so that the warning threshold can more sensitively reflect the changes in real-time data.

[0131] Analysis of simulation results:

[0132] The theoretical expected value of this indicator under the current working conditions is obtained from the analysis and simulation and the possible range of fluctuations .

[0133] Assume the weight of the analysis simulation results is , then analyze the contribution of simulation results to the warning threshold , It is a coefficient determined based on the reliability of the simulation and engineering requirements.

[0134] Theoretical expected value It is a theoretical calculation value of key monitoring indicators obtained by analyzing and simulating the dam body under currently known working conditions (such as water level, temperature and load). It is based on the structural model of the dam body, material properties and various physical and mechanical principles, and is a prediction of the operating state of the dam body under ideal conditions. It takes into account the uncertainty in the simulation process and the possible deviation between the actual working conditions and the theoretical assumptions, and indicates the range in which the indicators may deviate from the theoretical expected values.

[0135] Weight This reflects the importance of the analysis and simulation results in the entire process of determining the warning threshold. If the model based on the analysis and simulation is accurate, the parameters are reliable, and it can well simulate the actual behavior of the dam, then You can take an appropriate value. Determine based on the reliability of the simulation and engineering requirements. If the uncertainty of the simulation results is large, in order to ensure the safety of the early warning, you can appropriately increase , making the warning threshold more conservative; on the contrary, if the simulation results are very reliable, the .

[0136] Comprehensive warning threshold calculation:

[0137] Final dynamic warning threshold The weighted sum of these three parts is: ;

[0138] Among them, the weight 、 、 satisfy ,Their values ​​need to be determined comprehensively based on the reliability of historical data, the accuracy of real-time monitoring, and the credibility of analytical simulation.

[0139] In practice, the parameters in the above formulas should be appropriately adjusted and calibrated based on the specific conditions of the dam, the characteristics of the monitoring data, and engineering experience to ensure that the warning thresholds accurately reflect the dam's actual operating status and safety risks. Furthermore, the model should be regularly evaluated and updated to adapt to changes in dam operating conditions.

[0140] Final dynamic warning threshold This is achieved by weighting and summing the contributions of historical operating data, real-time monitoring data, and analytical simulation results. This weighted approach comprehensively considers the dam's past experience, current real-time status, and theoretical simulation analysis, providing a more comprehensive and accurate reflection of the dam's actual operating conditions and potential risks. By adjusting the weights and coefficients of each component, early warning thresholds can be flexibly set based on the specific project's characteristics and requirements, ensuring timely detection of potential safety issues while minimizing false alarms due to oversensitivity. Furthermore, as the dam's operating time progresses and new data accumulates, the model can regularly update its parameters based on actual conditions, further improving the accuracy and reliability of the early warning thresholds.

[0141] Example 7;

[0142] Based on Example 2, the steps for issuing warning information are:

[0143] Data comparison and judgment: Real-time comparison of monitoring data, simulation analysis results and dynamic warning thresholds. Once it is found that the data of a key monitoring indicator exceeds the warning threshold, whether it is the mean of the real-time monitoring data or the , standard deviation , or the theoretical expected value in the simulation analysis results or fluctuation range If relevant data exceeds the threshold, the early warning process will be triggered.

[0144] Determine the warning level: Based on the degree of threshold exceedance and the importance of related indicators, the warning level is determined according to pre-defined rules. For example, a slight threshold exceedance can be set as a yellow warning, a more serious exceedance as an orange warning, and a severe exceedance as a red warning. The warning level can be determined based on multiple factors, such as the percentage of the threshold exceedance and the duration of the continuous exceedance.

[0145] Generate warning content: Warning information should include detailed information, such as the specific indicator that exceeded the threshold, the current monitored or simulated data value, the warning level, the occurrence time, and the possible impact. For example, "The dam displacement indicator exceeded the warning threshold. The current real-time displacement average is X mm, exceeding the yellow warning threshold by Y mm. The warning level is yellow. The occurrence time is [specific time], which may affect the stability of the dam structure."

[0146] Send warning notifications: Generate warning information and send it to relevant personnel and systems in a timely manner through multiple channels. Project managers and technicians can be notified via SMS, email, or instant messaging. Warning information is also transmitted to the monitoring center's display system so that relevant personnel can obtain the information in a timely manner and take appropriate measures.

[0147] Feedback to the digital twin model for updates and optimization:

[0148] Data transmission and reception: Detailed monitoring data and simulation analysis results that trigger early warnings are transmitted to the digital twin model's database. The digital twin model should have corresponding interfaces and data reception mechanisms to accurately receive and store this data, ensuring its integrity and accuracy.

[0149] Model Analysis and Diagnosis: After receiving data, the digital twin model first conducts a detailed analysis of data that exceeds the warning threshold. Combining the dam's structural model with historical data, the model determines the possible cause of the data anomaly, including new damage to the dam structure, sudden changes in external loads, or errors in the monitoring system or simulation model itself.

[0150] Model parameter adjustment: Based on the results of the analysis and diagnosis, the relevant parameters of the digital twin model are adjusted and updated. If the data anomaly is caused by changes in the actual characteristics of the dam body, for example, the elastic modulus of the dam material changes due to long-term operation, then the material parameters in the model need to be adjusted accordingly. If it is found that certain assumptions or parameter settings of the simulation model are unreasonable, resulting in a significant deviation between the simulation results and the actual situation, the relevant model parameters are optimized, such as adjusting the boundary conditions and load distribution.

[0151] Model Validation and Evaluation: After parameter adjustments are completed, the digital twin model is validated and evaluated using the updated data. By comparing the model simulation results with actual monitoring data, the model is checked to see if it can more accurately reflect the actual operating status of the dam. If the model passes validation, it indicates that the updates and optimizations are effective. If not, further analysis is required until the model can accurately simulate the dam's behavior.

[0152] Visualization and Reporting: The updated and optimized digital twin model is presented to relevant personnel in a visual manner, including the structural status of the dam and the changing trends of key indicators. A detailed report is also generated, documenting the reasons for the model update, adjusted parameters, and verification results, providing a basis for subsequent decision-making and management.

[0153] This embodiment can issue early warning information in a timely manner when monitoring data or simulation analysis results exceed the early warning threshold, and feed the information back to the digital twin model for updating and optimization, thereby improving the monitoring and management level of the dam's safety status and promptly discovering and handling potential safety hazards.

[0154] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope of the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A digital twin-enabled dam adaptive early warning method, characterized in that: The steps include: Construct a digital twin model of the dam, integrating its geometry, structural information, material properties, and geological conditions; Deploy a variety of sensors to collect data on seepage, deformation, stress and strain of the dam body, as well as meteorological and hydrological external environmental data, and perform multi-source data fusion; Based on the digital twin model of the dam and the integrated multi-source data, numerical simulation algorithms are used to simulate the dam's operating status in real time, and the simulation results are compared and analyzed with the actual monitoring data; Establish a dynamic warning threshold model to automatically adjust the warning threshold based on the dam's historical operation data, real-time monitoring data, and analysis and simulation results; When the monitoring data or simulation analysis results exceed the warning threshold, a warning message is issued and fed back to the digital twin model for updating and optimization.

2. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: In the digital twin model of the dam body, the geometric shape includes: spatial coordinate calculation, area and volume calculation; the structural information includes: stress analysis, material anisotropy and deformation coordination calculation; the material properties include: material constitutive relationship, viscoelastic material and material damage model; the geological conditions include: seepage analysis and geomechanical model.

3. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: The multi-source data fusion is as follows: The data collected by different types of sensors are recorded as , the corresponding weight is ,and , the fused data Calculated by the following formula: ； For seepage data , deformation data , stress-strain data , meteorological data and hydrological data , if the weights of seepage data, deformation data, stress-strain data, meteorological data and hydrological data are 、 、 、 and , the fused data is: 。 4. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: The numerical simulation algorithm is data comparison and analysis: The root mean square error (RMSE) can comprehensively reflect the overall deviation between the simulated value and the actual monitored value. The formula is: ； in, is the number of data points, Indicates the analog values, Indicates the Actual observations, This is more noticeable for larger errors because the errors are first squared, summed, and then squared.

5. The digital twin-enabled dam adaptive early warning method according to claim 4 is characterized in that: Specific calculation steps: Calculate the difference between the simulated value and the actual observed value of each data point, that is, ; Squaring the difference, we get , which is convenient for eliminating the influence of the positive and negative signs of the differences, while amplifying the influence of large differences and highlighting the large deviations between the simulation results and the actual observations; The sum of the squared differences of all data points is: ; Then divide the sum by the number of data points , and the mean square error is: ; Taking the square root of the mean square error gives the root mean square error .

6. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: The establishment of the dynamic warning threshold model is as follows: Historical operation data part: calculate the mean value of the indicator in the historical operation data and standard deviation ; Assume the weight of historical data is , then the contribution of historical operation data to the warning threshold is: ,in, It is a coefficient determined based on the statistical characteristics of historical data and engineering experience, and is used to adjust the degree of leniency of early warning; Real-time monitoring data part: Calculate the mean of the current real-time monitoring data and standard deviation ; Assume the weight of real-time data is , then the contribution of real-time monitoring data to the warning threshold , is similar The coefficient is determined according to the characteristics and importance of real-time data; Analysis of simulation results: The theoretical expected value of this indicator under the current working conditions is obtained from the analysis and simulation and the possible range of fluctuations ; Assume the weight of the analysis simulation results is , then analyze the contribution of simulation results to the warning threshold , It is a coefficient determined based on the reliability of the simulation and engineering requirements; Comprehensive warning threshold calculation: Final dynamic warning threshold The weighted sum of these three parts is: ; Among them, the weight 、 、 satisfy .

7. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: The steps of issuing the warning information are: Data comparison and judgment: Real-time comparison of monitoring data, simulation analysis results and dynamic warning thresholds; Determine the warning level: Determine the warning level according to the degree of exceeding the threshold and the importance of relevant indicators in accordance with predetermined rules; Generate warning content: The warning information should contain detailed content; Send early warning notifications: Send the generated early warning information to the system in a timely manner through multiple channels.

8. The digital twin-enabled dam adaptive early warning method according to claim 1 is characterized in that: The feedback to the digital twin model for updating and optimization is: Data transmission and reception: Detailed monitoring data and simulation analysis results that trigger early warnings are transmitted to the database of the digital twin model; Model analysis and diagnosis: After the digital twin model receives data, it conducts detailed analysis of data that exceeds the warning threshold; Model parameter adjustment: Adjust and update the relevant parameters of the digital twin model based on the results of analysis and diagnosis; Model verification and evaluation: After completing parameter adjustments, use the updated data to verify and evaluate the digital twin model; Visualization and reporting: The updated and optimized digital twin model is presented in a visual manner.

9. A digital twin-enabled dam adaptive early warning system, used to execute the digital twin-enabled dam adaptive early warning method according to any one of claims 1 to 8, characterized in that: include: Data analysis module, used to simulate the dam's operating status in real time, and compare and analyze simulation results with actual monitoring data; The second model building module is used to automatically adjust the warning threshold through the historical operation data of the dam, real-time monitoring data and analysis simulation results; Early warning module, used to issue early warning information; The data analysis module is connected to the second model construction module, and the second model construction module is connected to the early warning module.

Images