Dam safety monitoring method based on digital twinning

Abstract

The invention belongs to the technical field of dam safety early warning, provides a dam safety monitoring method based on digital twinning, and aims at solving the problems that an existing monitoring method is weak in prediction capability, lags in early warning and the like. The method comprises the following steps: deploying multiple types of sensors to collect multi-dimensional physical state data; establishing a digital twinborn model containing a multi-physics field coupling simulation sub-model and data driving correction based on the BIM, the Internet of Things and finite elements; inputting a historical data driving model to predict an output state trend sequence and a state data value at a certain moment; performing difference analysis to construct an error vector, and calculating and correcting a deviation index; and calculating a structure health score, and if a threshold value is exceeded, early warning. The method has the advantages of improving prediction capability, enhancing adaptability, realizing comprehensive evaluation and providing systematic advanced decision basis for safety management.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of dam safety early warning, and particularly relates to a dam safety monitoring method based on digital twinning. BACKGROUND

[0002] A dam is a key water conservancy infrastructure, and its long-term operation state is directly related to the safety of life and property in the downstream area. The existing dam safety monitoring method mainly relies on real-time collection of physical parameters such as displacement, stress and water level, and abnormality is judged based on static threshold, which has problems such as weak prediction ability, lagging early warning and poor coupling with physical structure, and is difficult to adapt to the demand of dam structure state evolution monitoring under complex environment. In recent years, digital twinning technology has been gradually applied in the fields of industry and engineering. By constructing a virtual model corresponding to the actual structure, the visualization, predictability and feedback control of the system operation state are realized. The introduction of digital twinning into dam safety monitoring can integrate finite element simulation and real-time data driving, establish a prediction model for structure evolution, and reflect the structure health level through the difference between the model output and the actual monitoring data. However, the existing method still lacks systematization in multi-physical field coupling simulation, adaptive parameter correction and deviation index quantitative analysis, and it is difficult to form a unified health evaluation standard. SUMMARY

[0003] The application proposes a dam safety monitoring method based on digital twinning to solve the technical problems in the above background art.

[0004] In order to achieve the above purpose, the technical scheme adopted by the application comprises the following steps:

[0005] S1, deploying multiple types of sensors in the dam structure body and its surrounding environment, the sensors including displacement sensors, stress sensors and water level monitors, for collecting multi-dimensional physical state data of the dam during operation;

[0006] S2, establishing a digital twinning model based on BIM, Internet of Things and finite element, the digital twinning model including a dam multi-physical field coupling simulation submodel and data driven correction;

[0007] The dam multi-physical field coupling simulation submodel is used to reflect the dynamic evolution process of each monitoring index over time under normal state;

[0008] The data driven correction is used to adaptively fit and dynamically correct the parameters of the simulation submodel in combination with historical data;

[0009] S3, obtaining historical monitoring data and inputting the digital twinning model to drive the digital twinning model to output, the output including a first output and a second output;

[0010] The first output is a predicted future preset period state trend sequence, and the fluctuation intensity factor and the trend derivative factor in the trend sequence are extracted;

[0011] The second output is a predicted state data value at a certain time in a future preset period;

[0012] S4, difference analysis is performed on the predicted state data value of the second output and the historical stable state data, and a multi-dimensional error mapping vector is constructed;

[0013] S5, based on the error mapping vector, displacement deviation index, stress deviation index, and water level deviation index are calculated respectively, and the fluctuation intensity factor and the trend derivative factor of the first output are used to correct the displacement deviation index, the stress deviation index, and the water level deviation index to obtain corrected displacement deviation correction index, stress deviation correction index, and water level deviation correction index;

[0014] S6, based on the displacement deviation correction index, the stress deviation correction index, and the water level deviation correction index, a structure health score is calculated, and the structure health score is compared with a preset safety threshold, if the score is higher than the threshold, it is determined that the current dam has a structure abnormal risk, and corresponding warning information is output.

[0015] Preferably, the construction process of the dam multi-physical field coupling simulation sub-model in step S2 includes the following steps:

[0016] A finite element discrete model reflecting the dam structure entity is established, the dam structure is spatially meshed, and material parameters and boundary conditions are assigned to each unit;

[0017] A coupled multi-physical field solving module is constructed, and the response coupling between multiple fields is realized by setting multi-field interaction boundary conditions and loading paths;

[0018] Set the historical typical working condition as the simulation input condition, execute the initial simulation, output the theoretical response value of each monitoring point, and compare it with the measured historical data to extract the residual characteristics;

[0019] The extracted residual characteristics are used to fit and adjust the material constitutive parameters and boundary coupling coefficients in the model, and the initialization adjustment of the simulation sub-model is completed.

[0020] Preferably, the data-driven correction step of adaptively fitting and dynamically correcting the parameters of the simulation sub-model in combination with the historical data includes:

[0021] Obtain the measured state data of the dam in multiple historical operation periods, the data including stress, displacement, and seepage pressure time series data, and perform data normalization;

[0022] Residual sequences between historical measured data and corresponding predicted outputs of the simulation sub-model are calculated respectively, and residual fluctuation curves are constructed based on sliding time windows;

[0023] For the monitoring area in the residual fluctuation curve with a drift trend greater than a set threshold, a residual feedback function is constructed to map the residual mean, change rate and local fluctuation rate to a parameter correction factor;

[0024] The parameter correction factor is applied to the corresponding physical unit parameters in the simulation sub-model to complete parameter fine-tuning.

[0025] As a preferred embodiment, the residual feedback function is implemented as: wherein is a weight factor, is a reference value obtained by fitting historical data, respectively represent the residual mean, residual change rate and residual standard deviation.

[0026] As a preferred embodiment, the implementation of the step S3 of driving the digital twin model to generate a predicted output, which includes a first output and a second output, includes the following steps:

[0027] The historical monitoring data is input into the digital twin model to drive multi-physical field response calculation in simulation solution to obtain the current structure state;

[0028] Based on the current structure state, a sliding prediction window is used to generate a multi-time prediction state sequence in a future preset period, and the prediction state sequence is represented as: wherein represents the physical index state value at the t+i moment of prediction, and T is the length of the prediction period;

[0029] The prediction state sequence is taken as the first output, and the fluctuation intensity factor and the trend derivative factor are extracted therefrom;

[0030] The trend derivative factor is calculated as: wherein represents the trend derivative factor, and j represents the jth physical index;

[0031] The fluctuation intensity factor is calculated as: wherein, represents the fluctuation intensity factor, represents the predicted value of the jth physical index at the time t+i moment;

[0032] Finally, a preset target time point t+k is selected in the prediction state sequence, and the corresponding prediction vector x(t+k) is extracted as the second output, which is used for multi-dimensional difference comparison with the historical stable state vector.

[0033] As preferred, the step S4 of constructing the multi-dimensional error mapping vector is achieved by subtracting the multi-dimensional state data vector of the predicted future time from the second output from the reference value vector of the historical stable running state to obtain the error vector ; the error vector comprises displacement deviation, stress deviation, and water level deviation.

[0034] As preferred, the step S5 of obtaining the displacement deviation index, stress deviation index, and water level deviation index based on the error mapping vector comprises:

[0035] The index of the difference between the current predicted displacement in the error mapping vector and the historical stable displacement reference value is defined as the displacement deviation index, i.e. ; wherein, is the displacement deviation index, ;

[0036] The stress safety threshold value is set, and the stress deviation index is calculated, wherein, is the stress deviation index, ;

[0037] The water level deviation index is calculated, wherein, is the water level deviation index, wherein, represents the standard reference water level.

[0038] As preferred, the implementation of the step of correcting the displacement deviation index, stress deviation index, and water level deviation index by using the fluctuation intensity factor and trend derivative factor of the first output to obtain the corrected displacement deviation correction index, stress deviation correction index, and water level deviation correction index is achieved by calculating: wherein, wherein is represented as avoiding the divisor being 0, is represented as the jth type of deviation correction index, including , represents the corrected intermediate term.

[0039] As preferred, the calculation method of the structure health score in the step S6 is the normalized value of the product of the displacement deviation correction index, stress deviation correction index, and water level deviation correction index.

[0040] Compared with the prior art, the application has the advantages and positive effects that:

[0041] 1. Based on BIM and finite element, a multi-physical field coupling simulation submodel is constructed, the dynamic interaction of stress and seepage is integrated, the dynamic evolution law of monitoring indicators is accurately simulated through historical working condition simulation and residual correction parameters, the limitations of traditional single threshold judgment are broken through, and the prediction ability of structure state trend is improved.

[0042] 2. A residual feedback function is constructed through multi-period measured data, the residual characteristics are mapped to parameter correction factors, the material constitutive and boundary coefficients are dynamically fine-tuned, the parameter drift caused by environmental disturbance is automatically compensated, the long-term prediction error accumulation problem of static model is solved, and the adaptability under complex working conditions is enhanced.

[0043] 3. A multi-dimensional error vector is constructed, the deviation index is corrected by combining the fluctuation intensity and trend derivative factor, and the health state is comprehensively evaluated by normalized score. When the score exceeds the threshold, the early warning information containing abnormal type is output in real time, the limitations of independent parameter analysis are broken through, the change from lag response to trend prediction is realized, and the system and advanced decision basis for safety management is provided. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0045] Figure 1 It is a whole structure flow chart of a kind of dam safety monitoring method based on digital twin. DETAILED DESCRIPTION

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0047] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, however, the present application can also be implemented in other ways different from those described herein, therefore, the present application is not limited to the specific embodiments disclosed in the following description.

[0048] In the embodiment, the conventional dam safety monitoring method mainly relies on real-time collection of physical parameters such as displacement, stress and water level, and abnormality judgment based on static threshold. However, this method has significant defects: first, the prediction ability is weak, and only simple comparison can be made on the current data, and the future evolution trend of the dam structure state cannot be effectively predicted; second, the early warning is lagging, and the alarm is often given after the abnormal state has formed a certain scale, missing the best processing opportunity. Therefore, the present application proposes a dam safety monitoring method based on digital twinning, and the specific implementation process is as shown in Figure 1

[0049] First, in order to comprehensively and real-time obtain the physical state information of the dam in the running process, and provide a solid data foundation for the construction and analysis of the subsequent digital twinning model, a scheme of deploying multiple types of sensors in the dam structure body and its surrounding environment is adopted. Specifically, multiple types of sensors are deployed in the dam structure body and its surrounding environment, including displacement sensors, stress sensors, water level monitors, for collecting multi-dimensional physical state data of the dam in the running process. Through the cooperative deployment of multiple types of sensors, multi-dimensional data acquisition of the dam running state is realized, which can reflect the physical state of the dam under different working conditions in real time and comprehensively.

[0050] Then, in order to accurately simulate the dynamic evolution process of each monitoring index of the dam with time under normal state, and truly reflect the response characteristics of the dam structure under the action of multi-physical field coupling, a technical scheme based on BIM, Internet of Things and finite element is adopted to construct a digital twinning model, which includes a dam multi-physical field coupling simulation sub-model and a data-driven correction. The dam multi-physical field coupling simulation sub-model is used to reflect the dynamic evolution process of each monitoring index with time under normal state, and the data-driven correction is used to adaptively fit and dynamically correct the parameters of the simulation sub-model combined with historical data.

[0051] ​The construction process of the dam multi-physical field coupling simulation submodel includes the following steps: a finite element discrete model reflecting the dam structure entity is established, the dam structure is spatially meshed, and material parameters and boundary conditions are assigned to each unit; a coupled multi-physical field solving module is constructed, the mutual interaction boundary conditions and loading paths of multiple fields are set to realize the response coupling between multiple fields; historical typical working conditions are set as simulation input conditions, initial simulation is performed, the theoretical response values of each monitoring point are output, and are compared with the measured historical data to extract residual characteristics; the extracted residual characteristics are used to fit and adjust the material constitutive parameters and boundary coupling coefficients in the model, and the initialization adjustment of the simulation submodel is completed. Specifically, first, the entity geometric structure is established based on the dam design drawings, construction records and BIM model, and the finite element modeling method is used for spatial discrete processing of the dam structure. Specifically, the overall structure is divided into multiple calculation units according to the stress characteristics of the dam structure, and adaptive unit types are selected for meshing. Then, the material constitutive attribute parameters are assigned to each finite element unit, and the boundary conditions reflecting the boundary stress state are set, including foundation constraint boundary, water body action pressure and seepage pressure boundary, etc. On the basis of establishing the structure finite element grid, the multi-physical field solving including stress field and seepage field is constructed. By defining the interaction relationship between fields, the loading path and coupling formula between different physical fields are established. At the same time, the explicit joint solving strategy is adopted, and the influence of each field variable is iterated synchronously in the solving process, and the time stepping mechanism is introduced to simulate the dynamic response of the structure in the long-term operation process. Then, the representative historical typical working conditions are selected as the simulation input conditions. These working conditions can include the running water level in a certain period, the dam load change, etc. These time series data are input as boundary loading to drive the constructed multi-physical field simulation model to perform initial simulation. During the simulation process, the system will output the theoretical response values of all monitoring points at each time node, including displacement, stress, pore water pressure, etc. In order to facilitate subsequent deviation analysis, these theoretical prediction values are aligned with the historical measured data point by point, the difference values are compared, and the residual sequence is recorded. Finally, the residual sequence between the foregoing simulation output and the measured value is taken as the input to adaptively adjust the model parameters. First, the mean, change rate and local fluctuation rate of the residual of each monitoring point are extracted to construct a residual characteristic matrix. Then, the material constitutive parameters and boundary coupling coefficients are fitted and optimized using the characteristic matrix, which is mapped to the corresponding parameter correction factor through the residual feedback function, and the correction method can adopt exponential decay, normal modulation or weighted average method. After adjustment, the simulation is performed again, and if the residual is within the set convergence threshold, it is considered that the model initialization adjustment is completed.

[0052] After the sub-model is constructed, the parameters of the simulation sub-model are adaptively fitted and dynamically corrected by data-driven correction combined with historical data. Specifically, first, data sequences covering typical hydrology, meteorology and operation conditions in multiple historical operation periods of the dam are selected. Specifically, continuous time series data of stress, displacement and seepage pressure are extracted according to sensor categories to form three types of one-dimensional multi-period measured data sets. In order to reduce the dimensional difference between different physical quantities and interfere with the subsequent residual calculation, the historical maximum and minimum values of each type of measured data are normalized by Min-Max normalization processing, so that the value range is uniformly mapped to the [0, 1] interval. After completing the data normalization, the measured data of each type and the theoretical output value of the corresponding time point of the digital twin simulation model are differentiated point by point to obtain residual sequences of the same dimension. In order to more comprehensively reflect the dynamic fluctuation characteristics of the residual with time, the residual sequence is divided into multiple equal-length subsequences using a sliding window mechanism, and the residual mean, standard deviation and derivative trend in each window are calculated to construct a residual fluctuation curve reflecting the deviation trend in the time scale. For the monitoring area where the deviation trend in the residual fluctuation curve is significant, that is, the mean value of the continuous multiple windows deviates more than the set threshold, a residual feedback function is established to map the residual mean, change rate and local fluctuation rate to a model correction factor. The specific implementation is: wherein is the model correction factor, is the residual mean, change rate and local fluctuation rate, respectively, is the weight coefficient. Finally, the correction factor output by the residual feedback function is applied to the constitutive parameters and boundary conditions of the corresponding physical units in the simulation sub-model.

[0053] Next, in order to obtain prediction information of the future operation state of the dam and provide forward-looking data support for safety assessment and early warning, the historical monitoring data is input into the digital twin model to drive the model to output the prediction. The historical monitoring data is input into the digital twin model to drive the digital twin model to output the prediction, and the output includes a first output and a second output. The first output is to generate a future preset period state trend sequence, and the second output is to generate a state data value at a certain time in the future preset period. Specifically, the historical monitoring data is input into the digital twin model to drive the multi-physical field response calculation in the simulation solution to obtain the current structure state;

[0054] Based on the current structure state, a sliding prediction window is used to generate a multi-time prediction state sequence in the future preset period, and the prediction state sequence is represented as: wherein represents a predicted physical index state value at the t+i moment, T is the length of the prediction period; the predicted state sequence is taken as a first output, from which a fluctuation intensity factor and a trend derivative factor are extracted; the trend derivative factor is calculated in the following manner: wherein represents the trend derivative factor, j represents the jth physical index; the fluctuation intensity factor is calculated in the following manner: wherein represents the fluctuation intensity factor, represents a predicted value of the jth physical index at the time t+i; finally, a preset target time point t+k is selected in the predicted state sequence, and a corresponding prediction vector x(t+k) is extracted as a second output, which is used for multi-dimensional difference comparison with a historical stable state vector. Through the above prediction output scheme, dynamic prediction of the future state of the dam and multi-dimensional feature extraction are realized. The first output provides a state trend sequence and feature factors in a future preset period, which can comprehensively reflect the evolution trend and stability of the dam state; the second output precisely predicts a specific time point, which is convenient for comparison and analysis with the historical stable state, provides rich information for subsequent difference analysis and early warning, and significantly improves the prediction ability and foresight of the monitoring system.

[0055] Then, in order to quantitatively analyze the difference between the future predicted state of the dam and the historical stable state, and reveal the abnormality degree of the dam structure state, the predicted state data value of the second output is subjected to difference analysis with the historical stable state data to construct a multi-dimensional error mapping vector. Based on the error mapping vector, a displacement deviation index, a stress deviation index and a water level deviation index are respectively calculated, and the fluctuation intensity factor and the trend derivative factor of the first output are used to correct the displacement deviation index, the stress deviation index and the water level deviation index to obtain a corrected displacement deviation correction index, a stress deviation correction index and a water level deviation correction index. Specifically, first, the multi-dimensional state data vector of the future moment predicted by the second output is subjected to difference calculation with a reference value vector of the historical stable operation state to obtain an error vector ; the error vector includes displacement deviation, stress deviation and water level deviation. The index of the difference between the current predicted displacement in the error mapping vector and the historical stable displacement reference value is defined as the displacement deviation index, i.e. ; wherein, is the displacement deviation index, ; a stress safety threshold is set, and a stress deviation index is calculated, wherein, is the stress deviation index, ; a water level deviation index is calculated, wherein, a water level deviation index, wherein, represents a standard reference water level; finally, the fluctuation intensity factor and the trend derivative factor of the first output are utilized to correct the displacement deviation index, the stress deviation index and the water level deviation index to obtain a corrected displacement deviation correction index, a stress deviation correction index and a water level deviation correction index, and the implementation calculation manner is: wherein, wherein is represented as avoiding the divisor being 0, is represented as the j-th type of deviation correction index, including , represents a corrected intermediate term. Through the correction of the deviation index, the dynamic characteristics of the trend of the dam state are fully considered. The fluctuation intensity factor reflects the stability of the state, and the trend derivative factor embodies the rate of change of the state, and the introduction of the two makes the deviation index not only reflect the current deviation degree, but also dynamically adjust the change of the future state trend.

[0056] Finally, in order to comprehensively evaluate the health state of the dam structure and realize the timely early warning of the structural abnormal risk, a structure health score is calculated based on the displacement deviation correction index, the stress deviation correction index and the water level deviation correction index, and the structure health score is compared with a preset safety threshold value, if the score is higher than the threshold value, it is determined that the current dam has a structural abnormal risk, and corresponding early warning information is output. Specifically, the calculation manner of the structure health score is the normalized value of the product of the displacement deviation correction index, the stress deviation correction index and the water level deviation correction index. The normalization processing can unify the score range to the [0, 1] interval, which is convenient for setting the safety threshold value and comparing the health states in different periods. When the calculated structure health score is higher than the preset safety threshold value, it is determined that the current dam has a structural abnormal risk, and corresponding early warning information is output. The early warning information can include detailed contents such as abnormal type, abnormal degree, occurrence time and position, which provides a decision basis for the dam management department, so as to timely take corresponding maintenance measures to eliminate safety hazards. Through the calculation of the structure health score and the establishment of the early warning mechanism, the quantitative comprehensive evaluation of the health state of the dam structure is realized. The method overcomes the limitations of the existing method which relies on a single index static threshold judgment, can comprehensively consider the deviation of multiple physical indexes and their dynamic trend, form a unified health evaluation standard, and improve the scientificity and accuracy of abnormal judgment.

[0057] The above merely describes preferred embodiments of the present application, but is not intended to limit the present application to other forms, any person skilled in the art can make changes or modifications to the above disclosed technical contents into equivalent embodiments with equivalent changes, and apply to other fields, but any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the technical solution content of the present application still belongs to the protection scope of the present application technical solution.

Claims

1. A dam safety monitoring method based on digital twins, characterized in that, Includes the following steps: S1. Deploy multiple types of sensors in the dam structure and its surrounding environment. The sensors include displacement sensors, stress sensors, and water level monitors, which are used to collect multi-dimensional physical state data of the dam during operation. S2. Establish a digital twin model based on BIM, IoT and finite element method. The digital twin model includes a multi-physics coupled simulation sub-model of the dam and data-driven correction. Among them, the multi-physics field coupled simulation sub-model of the dam is used to reflect the dynamic evolution of each monitoring index over time under normal conditions; Data-driven correction is used to adaptively fit and dynamically correct the parameters of the simulation sub-model by combining historical data; S3. Acquire historical monitoring data and input it into the digital twin model to drive the digital twin model to make prediction outputs, the outputs including a first output and a second output; The first output is to predict and generate a future preset cycle state trend sequence, and extract the volatility intensity factor and trend derivative factor from the trend sequence; The second output is to predict and generate the state data value at a certain moment within a preset future period; S4. Perform a difference analysis between the predicted state data value of the second output and the historical stable state data, and construct a multi-dimensional error mapping vector; S5. Based on the error mapping vector, calculate the displacement deviation index, stress deviation index and water level deviation index respectively. Use the first output fluctuation intensity factor and trend derivative factor to correct the displacement deviation index, stress deviation index and water level deviation index to obtain the corrected displacement deviation correction index, stress deviation correction index and water level deviation correction index. S6. Calculate the structural health score based on the displacement deviation correction index, stress deviation correction index and water level deviation correction index, and compare the structural health score with the preset safety threshold. If the score is higher than the threshold, it is determined that there is a structural anomaly risk in the dam and the corresponding early warning information is output.

2. The dam safety monitoring method based on digital twins according to claim 1, characterized in that, The construction process of the multiphysics coupled simulation sub-model of the dam in step S2 includes the following steps: A finite element discrete model reflecting the dam structure was established, the dam structure was spatially meshed, and material parameters and boundary conditions were assigned to each element. A coupled multiphysics solution module is constructed, and the response coupling between multiple fields is achieved by setting the boundary conditions and loading paths for multi-field interactions; Set historical typical working conditions as simulation input conditions, execute the initial simulation, output the theoretical response values ​​of each monitoring point, and compare them with the measured historical data to extract residual features; The extracted residual features are used to fit and adjust the material constitutive parameters and boundary coupling coefficients in the model, thus completing the initialization and adjustment of the simulation sub-model.

3. The dam safety monitoring method based on digital twins according to claim 2, characterized in that, The data-driven correction step, which combines historical data to adaptively fit and dynamically correct the parameters of the simulation sub-model, includes: The measured status data of the dam during multiple historical operating cycles were obtained, including stress, displacement, and seepage pressure time series data, and the data were normalized. The residual sequences between the historical measured data and the corresponding predicted outputs of the simulation sub-models are calculated respectively, and the residual fluctuation curves are constructed based on the sliding time window; For the monitoring area in the residual fluctuation curve where the offset trend is greater than the set threshold, a residual feedback function is constructed to map the residual mean, rate of change and local volatility into parameter correction factors; The parameter correction factor is applied to the corresponding physical unit parameters in the simulation sub-model to complete the parameter fine-tuning.

4. The dam safety monitoring method based on digital twins according to claim 3, characterized in that, The residual feedback function is implemented as follows: ,in As a weighting factor, The baseline reference value is obtained by fitting historical data. These represent the residual mean, the residual rate of change, and the residual standard deviation, respectively.

5. A dam safety monitoring method based on digital twins according to claim 1, characterized in that, The implementation of driving the digital twin model to make predictions in step S3, including the first output and the second output, includes the following steps: Historical monitoring data is input into the digital twin model to drive multiphysics response calculations in the simulation solution, thereby obtaining the current structural state; Based on the current structural state, a multi-time predicted state sequence within a preset future period is generated using a sliding prediction window. The predicted state sequence is represented as follows: ,in This represents the predicted physical index state value at time t+i, where T is the prediction period length. The predicted state sequence is used as the first output, from which the volatility intensity factor and trend derivative factor are extracted. The trend derivative factor is calculated as follows: ,in represents the trend derivative factor, and j represents the j-th physical index; The fluctuation intensity factor is calculated as follows: ,in, Indicates volatility intensity factor. This represents the predicted value of the j-th physical index at time t+i. Finally, a preset target time point t+k is selected from the predicted state sequence, and the corresponding predicted vector x(t+k) is extracted as the second output, which is used to perform multi-dimensional difference comparison with the historical stable state vector.

6. The dam safety monitoring method based on digital twin according to claim 1, characterized in that, The construction of the multidimensional error mapping vector in step S4 is achieved by using the multidimensional state data vector of the future time predicted by the second output. Reference value vector under historical stable operating conditions By subtracting each dimension, the error vector is obtained. The error vector includes displacement deviation, stress deviation, and water level deviation.

7. The dam safety monitoring method based on digital twin according to claim 1, characterized in that, Step S5, which involves obtaining the displacement deviation index, stress deviation index, and water level deviation index based on the error mapping vector, includes the following: The exponent of the difference between the current predicted displacement and the historical stable displacement reference value in the error mapping vector is defined as the displacement deviation index. ;in, The displacement deviation index. ; Set stress safety threshold Calculate the stress deviation index ,in, The stress deviation index, ; Calculate the water level deviation index ,in, This refers to the water level deviation index. ,in, This represents the standard reference water level.

8. A dam safety monitoring method based on digital twins according to claim 7, characterized in that, The calculation method for correcting the displacement deviation index, stress deviation index, and water level deviation index using the first output fluctuation intensity factor and trend derivative factor is as follows: ,in, ,in This is to avoid division by zero. Represented as the j-th type deviation correction index, including , This represents the intermediate term used for correction.

9. A dam safety monitoring method based on digital twins according to claim 1, characterized in that, The structural health score in step S6 is calculated as the normalized value of the product of the displacement deviation correction index, the stress deviation correction index, and the water level deviation correction index.

Images