CN-121980862-A - Digital twin water conservancy-oriented multi-source monitoring data fusion processing system

CN121980862ACN 121980862 ACN121980862 ACN 121980862ACN-121980862-A

Abstract

The invention relates to the technical field of simulation and emulation, and discloses a multisource monitoring data fusion processing system for digital twin water conservancy; the dam dynamic simulation system comprises a region identification module, a model conversion module, a digital twin module and a simulation module, wherein the region identification module is used for establishing a three-dimensional model and identifying a key region from the three-dimensional model, the model conversion module is used for establishing a physical model, the digital twin module is used for establishing a digital twin model of a dam, the simulation module is used for simulating a real-time compression value and a future compression value, and the simulation system can perform multi-dimensional local simulation on the aspects of structural morphology, seepage materials and dynamic loads, so that the real physical interaction condition inside the dam is comprehensively reflected, the problems of limitation and inefficiency caused by the simulation of the whole structure and single dimension are avoided, the physical model is ensured to have dynamic and accurate time-varying characteristics and higher dynamic response capacity, and further the dynamic change of the dam under the complex load can be truly simulated.

Inventors

SONG LI
CAI YUNZHONG
WANG HAIYAN
WANG JUN
GAO ZHENG
LI CHAN
JIANG HAONAN
SHEN JING
ZHANG YONG

Assignees

宿迁市宿城区水利工程建设服务中心
江苏省水利科学研究院

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (10)

1. Digital twin water conservancy-oriented multi-source monitoring data fusion processing system is applied to water conservancy monitoring platform, and is characterized in that the system comprises: the regional identification module is used for acquiring basic parameters of the dam, wherein the basic parameters comprise geometric parameters and non-geometric parameters, establishing a three-dimensional model corresponding to the basic parameters through a BIM technology, and identifying a key region from the three-dimensional model; the model conversion module is used for constructing a single field model of the key region, wherein the single field model comprises a structural field model, a seepage field model and a dynamic field model, determining the association relation between the two single field models, and coupling the single field models of the key region according to the association relation to promote the three-dimensional model to be converted into a physical model; The digital twin module is used for determining monitoring moments of interval distribution, collecting multi-source monitoring data of the dam at the monitoring moments, wherein the multi-source monitoring data comprise environmental meteorological data, hydrological water flow data, structural response data and geological structure data, and carrying out fusion simulation on the multi-source monitoring data and the physical model to construct a digital twin model of the dam; The simulation module is used for simulating the real-time compression value of the digital twin model through the multi-source monitoring data at a single monitoring moment, simulating the future compression value of the digital twin model through the multi-source monitoring data at a plurality of monitoring moments, and making continuous safety information, current safety information or dangerous and out-of-control information.
2. The multi-source monitoring data fusion processing system for digital twin water conservancy according to claim 1, wherein when a three-dimensional model is established, geometric parameters and non-geometric parameters of a dam are queried through a design drawing and a database, a basic model with a contour framework and an attribute framework is established through a BIM technology, contour units distributed in a strip mode are established on the contour framework, attribute units distributed in a ring mode are established on the attribute framework, and the geometric parameters and the non-geometric parameters are respectively led into the contour units and the attribute units to promote the basic model to be converted into the three-dimensional model.
3. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system as defined in claim 2, wherein the key region identification method comprises the following steps: importing the three-dimensional model into finite element simulation software, and dividing the three-dimensional model into grids by taking a preset unit length as the side length of the grid to generate a grid model; Marking grids covered by the dam body, the dam foundation and the gallery in the grid model respectively, marking the grids as the dam body grids, the dam foundation grids and the gallery grids, and marking the rest grids as grids to be tested; Drawing lines along boundary positions of the dam body and the dam foundation, the dam foundation and the gallery and the boundary positions of the dam body and the gallery respectively, generating a first boundary line, a second boundary line and a third boundary line, measuring distance values from the to-be-tested grid to the first boundary line, the second boundary line and the third boundary line one by one, and marking the to-be-tested grid with the distance value smaller than a calibrated distance threshold value as a key grid; Summarizing the key grids with the distribution states of continuous distribution into a grid set, drawing a closed curve along the outer edges of the key grids in the same grid set, and marking the areas in the closed curve as key areas to obtain C key areas.
4. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system as defined in claim 3, wherein the construction method of the structural field model is as follows: The method comprises the steps of determining a model and parameters, wherein a concrete dam is analyzed by adopting a nonlinear elastic model, a rock-based dam is analyzed by adopting an elastoplastic model, and a creep model or a shrinkage model is adopted to set time-varying characteristics so as to generate a structural sub-model; The boundary condition and load are determined by fixing the bottom of the dam foundation, applying normal constraint to the left bank and the right bank of the dam for fixing, and setting the dead weight, hydrostatic pressure, lifting pressure, sediment pressure and temperature load of the dam; solving and setting, namely combining the Newton-Lawson iteration method with a solver, and setting a convergence criterion of the solver; Time-course analysis, namely dispersing time into a plurality of incremental steps, and updating material properties in each incremental step to promote the conversion of the structural sub-model into a structural field model.
5. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system as defined in claim 4, wherein the seepage field model is constructed by the following steps: Determining a seepage area and boundary conditions, namely marking areas where a dam body, a dam foundation and a downstream area of the dam are located as seepage areas, and configuring a known water head boundary, a known flow boundary, a watertight boundary and a free seepage boundary on the seepage areas to generate a seepage sub-model; determining a permeability coefficient, namely inquiring the permeability coefficient of a dam body, simulating the permeability coefficient of the dam foundation through a pressurized water experiment, and setting a saturated permeability coefficient, a porosity, a non-saturated parameter and a shape parameter of a seepage sub-model; Determining initial conditions, namely in a stable seepage field, taking a designed water level as a boundary, and solving a steady seepage equation to obtain the initial seepage field; Discrete and solving, namely using a finite element method or a finite difference method to discrete a seepage control equation, and adopting an iteration method to solve a seepage submodel; Calculating the osmotic force, namely calculating the osmotic force through an osmotic force calculation formula, and fusing the osmotic force with a seepage sub-model to promote the seepage sub-model to be converted into a seepage field model.
6. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system as defined in claim 5, wherein the power field model is constructed by the following steps: the natural frequency and the natural vibration mode of the dam are detected through a vibration sensor, and a dynamic sub-model matched with the natural frequency and the natural vibration mode is configured; Determining the load and the damping by adopting a seismic acceleration time course to determine the seismic load and adopting Rayleigh damping as a damping model; dynamic time interval analysis, namely analyzing dynamic load by adopting a direct integration method, analyzing wave propagation by adopting explicit integration, and analyzing structural vibration by adopting implicit integration; And establishing a secondary model comprising a dam foundation and a partial foundation, simulating an infinite foundation of the secondary model by adopting a reflection-free boundary or a viscoelastic boundary, and fusing the secondary model with a dynamic sub-model to construct an dynamic field model.
7. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system according to claim 6, wherein the association relationship comprises a master-slave relationship, a causal relationship and a parallel relationship; the method comprises the steps of determining the association relation between a structure field model and a seepage field model as a causal relation, determining the association relation between the structure field model and a power field model as a parallel relation, and determining the association relation between the power field model and the seepage field model as a master-slave relation.
8. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system as defined in claim 7, wherein the conversion method of the physical model is as follows: A1, constructing a basic scene with a closed contour line in a three-dimensional space, and setting three annular coupling bits distributed at equal angles in the basic scene; A2, respectively importing a structural field model, a seepage field model and a dynamic field model in the key area into three coupling positions of a coupling scene, and constructing an out-coupling link between two adjacent coupling positions; A3, remarking a causal relation on a coupling link between the structure field model and the seepage field model, remarking a parallel relation on a coupling link between the structure field model and the power field model, remarking a master-slave relation on a coupling link between the power field model and the seepage field model, and respectively generating a causal link, a parallel link and a master-slave link; A4, repeatedly executing the steps A2-A3 until the causal link, the parallel link and the master-slave link are generated in all the C key areas, and then, promoting the three-dimensional model to be converted into a physical model.
9. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system of claim 8, wherein the method for determining the monitoring time is as follows: taking the last time the data updating of the database is completed as a starting point, taking the current time as an ending point, and inquiring D dam logs from the starting point to the ending point; calculating the maximum duration and the minimum duration of a standard range change phenomenon of dam pressure in D dam logs one by taking a preset dam pressure change range as a standard range, and obtaining D duration peaks and D duration valleys; d time length peaks and corresponding D time length valleys are subjected to difference to obtain D time length differences, the maximum value and the minimum value of the time length differences are removed, the rest D-2 time length differences are accumulated and averaged, and the interval time length is calculated; And taking the current moment as a first monitoring moment, taking the interval duration as a standard, pushing the mark forwards along the time line, and determining E monitoring moments distributed at intervals.
10. The digital twin water conservancy-oriented multi-source monitoring data fusion processing system of claim 9, wherein the method for constructing the digital twin model is as follows: Time alignment is carried out on environmental meteorological data, hydrological water flow data, structural response data and geological structure data at the same monitoring moment; Combining the material elastic modulus, the permeability coefficient and the thermal expansion coefficient of the physical model into a parameter set, configuring a triggering state on the parameter set, and initializing the triggering state into an adjustable state; The multi-source monitoring data is combined with the physical model by utilizing a data assimilation technology, the triggering state is controlled to be switched into the closing state, and the time stamp of the multi-source monitoring data is marked, so that the physical model is converted into a digital twin model.

Description

Digital twin water conservancy-oriented multi-source monitoring data fusion processing system Technical Field The invention relates to the technical field of simulation, in particular to a multi-source monitoring data fusion processing system for digital twin water conservancy. Background The digital twin technology is used as a core means for realizing dynamic interaction of physical entities and virtual models, is gradually applied to safety monitoring of hydraulic engineering, can realize simulation and state evaluation of multiple physical processes such as structural behaviors, seepage states, dynamic responses and the like of a dam by constructing a digital twin model corresponding to an entity dam, and provides important support for hydraulic engineering safety early warning and operation and maintenance decision. The invention discloses a dam self-adaptive early warning method and system based on the patent application with the publication number of CN120671230A, comprising the steps of constructing a digital twin model of the dam, deploying a plurality of sensors to collect seepage, deformation and stress strain data of the dam, collecting weather and hydrologic external environment data, carrying out multi-source data fusion, simulating the running state of the dam in real time by using a numerical simulation algorithm based on the digital twin model of the dam and the fused multi-source data, comparing and analyzing the simulation result with actual monitoring data, establishing a dynamic early warning threshold model, automatically adjusting the early warning threshold according to historical running data, the real-time monitoring data and the analysis result of the dam, and when the monitoring data or the simulation result exceeds the early warning threshold, feeding back the early warning information to the digital twin model for updating and optimizing by sending out early warning information; When the conventional multisource monitoring data fusion processing system builds a virtual model of a dam, a single-dimension physical field is generally adopted for modeling and simulation, and the multidimensional physical fields such as a structure, seepage, power and the like of the dam cannot be efficiently coupled, so that the built virtual model has the problem of limitation when simulating complex real working conditions, the phenomenon of overlarge deviation of a simulation result is easy to occur, dynamic change of the dam under complex load cannot be truly simulated, and the accuracy of the simulation result of the dam simulation is reduced. In view of the above, the present invention proposes a multi-source monitoring data fusion processing system for digital twin water conservancy to solve the above problems. Disclosure of Invention In order to overcome the defects in the prior art and achieve the purposes, the invention provides the technical scheme that a digital twin water conservancy-oriented multi-source monitoring data fusion processing system is applied to a water conservancy monitoring platform and comprises the following components: the regional identification module is used for acquiring basic parameters of the dam, wherein the basic parameters comprise geometric parameters and non-geometric parameters, establishing a three-dimensional model corresponding to the basic parameters through a BIM technology, and identifying a key region from the three-dimensional model; the model conversion module is used for constructing a single field model of the key region, wherein the single field model comprises a structural field model, a seepage field model and a dynamic field model, determining the association relation between the two single field models, and coupling the single field models of the key region according to the association relation to promote the three-dimensional model to be converted into a physical model; The digital twin module is used for determining monitoring moments of interval distribution, collecting multi-source monitoring data of the dam at the monitoring moments, wherein the multi-source monitoring data comprise environmental meteorological data, hydrological water flow data, structural response data and geological structure data, and carrying out fusion simulation on the multi-source monitoring data and the physical model to construct a digital twin model of the dam; The simulation module is used for simulating the real-time compression value of the digital twin model through the multi-source monitoring data at a single monitoring moment, simulating the future compression value of the digital twin model through the multi-source monitoring data at a plurality of monitoring moments, and making continuous safety information, current safety information or dangerous and out-of-control information. Further, when the three-dimensional model is built, the geometric parameters and the non-geometric parameters of the dam are queried through a design drawing and a database,