An intelligent monitoring and operation platform and method for a covering tank based on digital twinning

Abstract

The application discloses a kind of based on digital twinning's covering soil tank intelligent monitoring, operation and maintenance platform and method, belong to digital twinning technical field.The digital twinning platform is as core, integrates physical geometry simulation module, algorithm prediction module, sensing communication module, virtual visualization module, and each functional module carries out data interaction and collaborative work by digital twinning platform.Firstly, physical geometry simulation module is based on the actual structure of covering soil tank and obtains the three-dimensional geometric model of multiple working conditions mechanical response data.Secondly, the collaborative prediction of external load distribution, structural response characteristics and fatigue damage is realized by algorithm prediction module.Thirdly, sensing communication module real-time acquisition and processing sensor data, drive model dynamic update.Finally, virtual visualization module renders the state of multiple physical fields of covering soil tank and damage evolution trend through multi-source data fusion technology.The application significantly improves the comprehensive monitoring capability and failure warning accuracy of covering soil tank, supports predictive maintenance decision, and reduces operation and maintenance cost.

Description

Technical Field

[0001] The present invention belongs to the field of digital twin technology and relates to an intelligent monitoring, operation and maintenance platform and method for earth-covering tanks based on digital twins. Background Art

[0002] Earth-covered tanks are a key component of modern industrial storage and transportation systems, responsible for the safe storage of flammable, explosive, toxic, and hazardous chemicals. Their large capacity, excellent sealing, and strong corrosion resistance make them irreplaceable in the energy and chemical industries. However, due to the complex and harsh operating environment of earth-covered tanks, they face the constant threat of internal and external corrosion and structural aging. Leakage or explosion could cause major safety incidents and environmental pollution, necessitating the establishment of an intelligent monitoring system.

[0003] A Chinese invention patent (publication number CN118643457B) provides a parallel operation and maintenance method, equipment, and media for multi-source data on earth-covered storage tanks. Based on the Apache Spark framework, this system performs real-time parallel computations on heterogeneous data such as tank pressure, temperature, soil parameters, and air composition. It analyzes variable correlations using correlation coefficients and automatically triggers the flame arrester and inert gas release mechanism when an anomaly is detected. A Chinese invention patent (publication number CN119337112A) provides a corrosion monitoring method for earth-covered storage tanks. By constructing a prediction model that integrates a convolutional neural network with an LSTM, it combines environmental data, internal tank sensor data, and acoustic emission signals to intelligently predict corrosion rates and visualize them using digital twin technology. While existing technical solutions can provide relatively accurate real-time monitoring of specific indicators of earth-covered tanks, they lack comprehensive monitoring of the overall structural condition of the tanks, and are unable to visually display the tank's status and predict damage. These shortcomings severely limit the monitoring system's early warning accuracy and risk prevention capabilities.

[0004] Digital twin technology builds a high-fidelity virtual model to achieve two-way transmission and fusion of simulation data and measured data, thereby establishing a precise mapping relationship between physical entities and virtual space. By simulating key indicators such as the operating status, operating parameters, and mechanical properties of physical objects in the digital twin, real-time analysis and monitoring of system operating data can be performed to effectively identify and avoid potential risks. By integrating sensor measured data and numerical simulation results, a digital twin platform is built to present the multi-dimensional mechanical state characteristics of the earth cover tank in real time, enabling real-time monitoring of the underground storage tank status. At the same time, based on the accumulated measured data, fatigue damage estimation can be performed on the earth cover tank, enabling predictive maintenance, reducing maintenance costs in the actual production environment, and effectively controlling the risk of structural failure. Summary of the Invention

[0005] In view of the problems existing in the prior art, the present application provides an intelligent monitoring and operation platform and method for a covering tank based on digital twinning. The present application realizes real-time bidirectional mapping of a physical entity and a virtual model by building a digital twinning platform for the covering tank, integrating multi-source sensing data and finite element simulation analysis. The present application realizes real-time synchronous mapping of multi-dimensional mechanical states of the covering tank, dynamic failure risk early warning and visual interactive feedback on the digital twinning platform by collecting and transmitting multi-source sensing data in real time and constructing a prediction model based on a neural network to realize efficient prediction.

[0006] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0007] An intelligent monitoring and operation platform for a covering tank based on digital twinning, wherein the intelligent monitoring and operation platform for the covering tank takes a digital twinning platform as a core and integrates four functional modules, including a physical geometric simulation module, an algorithm prediction module, a sensing communication module and a virtual visualization module; the functional modules perform data interaction and collaborative work through the digital twinning platform, in particular:

[0008] The physical geometric simulation module is composed of a covering tank physical entity, a three-dimensional geometric model established based on structure parameters of the covering tank physical entity and sensors deployed on the covering tank physical entity, and is used to acquire real state information of the covering tank in real time; a high-fidelity three-dimensional geometric model is constructed through the structure parameters of the covering tank physical entity, and mechanical response data under multiple working conditions are acquired through finite element simulation analysis.

[0009] The algorithm prediction module runs on the digital twinning platform, and a multi-parameter correlation prediction model group is constructed based on the mechanical response data under multiple working conditions generated by the physical geometric simulation module, and mechanical performance information of the covering tank is outputted for rendering. The multi-parameter correlation prediction model group includes a load inversion prediction model, a mechanical response prediction model group and a dynamic damage evolution prediction model.

[0010] The sensing communication module is connected with the sensors deployed on the covering tank physical entity, and the sensing communication module acquires data collected by the sensors (the collected data includes key data such as structure strain of the covering tank) in real time, obtains monitoring data after data preprocessing (the preprocessing includes denoising processing and feature extraction), and transmits the monitoring data to the digital twinning platform through a low-delay and efficient data communication protocol. After receiving the monitoring data, the digital twinning platform drives the algorithm prediction module to predict the external load state of the covering tank in real time, and realizes dynamic calculation of the mechanical state of the covering tank structure.

[0011] The virtual visualization module relies on the digital twinning platform, and through computer graphics technology, the multi-dimensional mechanical state of the covering tank is presented in real time in a three-dimensional dynamic form on the digital twinning platform.

[0012] A digital twin-based intelligent monitoring and operation and maintenance method for earth-covering tanks is implemented based on the above-mentioned intelligent monitoring and operation and maintenance platform for earth-covering tanks, and includes the following steps:

[0013] Step 1: Build a high-fidelity physical geometry simulation module. Build a high-fidelity 3D geometry model based on the physical structure of the earth-covered tank, and obtain mechanical response data under multiple working conditions through finite element simulation analysis. Specifically:

[0014] Step 1.1: Based on the physical structural characteristics of the earth-covered tank, use 3D modeling software to establish a 3D geometric model of the earth-covered tank. By adjusting parameters and optimizing the design, the 3D geometric model can accurately reflect the actual structure of the earth-covered tank.

[0015] Step 1.2: Simplify the 3D geometric model established in step 1.1, and use finite element pre-processing software to divide the simplified model into regional grids to obtain a grid model. The number and spatial coordinates of each grid node are exported to a local data file.

[0016] Step 1.3: Calculate multidimensional load parameters based on the external loads and environmental factors experienced by the earth-covered tank in actual use. Use these as input variables and generate a characteristic sample set within the parameter domain using a spatial sampling method. Perform a static analysis of the earth-covered tank based on this characteristic sample set to identify the key factors affecting its structural performance.

[0017] Step 1.4: Based on the static analysis results obtained in step 1.3, select key measuring points on the physical entity of the earth-covering tank as sensor layout points.

[0018] Step 1.5: Using the finite element analysis method, apply the multidimensional load parameters in the characteristic sample set obtained in step 1.3 to the grid model divided in step 1.2, and output the mechanical response data of the entire earth-covered tank and the key measurement points selected in step 1.4 under multiple working conditions.

[0019] Step 1.6: Deploy sensors at the sensor locations determined in Step 1.4 and build a sensor communication module. The sensors collect data in real time and acquire physical signals at key locations on the earth cover tank.

[0020] Step 2: The algorithm prediction module builds a multi-parameter correlation prediction model group based on the mechanical response data under multiple working conditions provided by the physical geometry simulation module. Specifically:

[0021] Step 2.1: Use the mechanical response data of the key measuring points under multiple working conditions obtained in step 1.5 as input variables and the multidimensional load parameters in the characteristic sample set obtained in step 1.3 as output responses to construct a load inversion prediction model to achieve multidimensional dynamic feedback of the external load state of the earth-covering tank.

[0022] Step 2.2: Use the multidimensional load parameters in the characteristic sample set obtained in step 1.3 as input and the mechanical response of the earth-covered tank (including stress, deformation, displacement, etc.) as output to construct a mechanical response prediction model group to achieve collaborative prediction of structural response under multiple load conditions.

[0023] Step 2.3: Based on the mechanical response prediction model group obtained in step 2.2, a dynamic damage evolution prediction model is constructed. Specifically:

[0024] Step 2.3.1: Configure a double-buffer queue for each grid node of the grid model in step 1.2 to store the real-time stress data of each node output by the mechanical response prediction model group described in step 2.2 in time series, and realize the parallel execution of real-time data acquisition and damage analysis through the asynchronous processing mechanism.

[0025] Step 2.3.2: Traverse the cached data in step 2.3.1, reconstruct the load sequence, and generate the cyclic load spectrum.

[0026] Step 2.3.3: Based on the cyclic load spectrum reconstructed in step 2.3.2, identify the complete stress cycle and calculate the stress amplitude of each cycle and mean stress .

[0027] The stress amplitude The calculation formula is as follows:

[0028]

[0029] The mean stress The calculation formula is as follows:

[0030]

[0031] in, Indicates the maximum stress value in the current cycle; Indicates the minimum stress value in the current cycle;

[0032] Further calculation of the equivalent alternating stress amplitude , in order to eliminate the influence of average stress on fatigue life. The equivalent alternating stress amplitude The calculation formula is as follows:

[0033]

[0034] in, is the tensile strength of the material.

[0035] Step 2.3.4: Based on the equivalent alternating stress amplitude obtained in step 2.3.3 , combined with the material's stress-life (SN) curve, calculate the material's lifespan at a specific stress amplitude ( ) under the condition of fatigue failure, the number of cycles N is as follows:

[0036]

[0037] Where C and k are material-related constants. C represents the proportionality coefficient of the material's fatigue performance, and k represents the attenuation rate of the material's fatigue strength with stress amplitude.

[0038] Step 2.3.5: Based on Miner's linear cumulative damage theory, establish a fatigue damage assessment model under multiple stress levels and calculate the cumulative damage value D as follows:

[0039]

[0040] Where m represents the total number of divided stress levels; Indicates the actual number of cycles under the stress level i; N i It represents the fatigue life determined based on the SN curve at this stress level.

[0041] When the cumulative damage value D>1, it is judged that the earth-covered tank has reached the critical state of fatigue failure.

[0042] Step 3: The sensor communication module captures and processes the data collected by the sensor in step 1 in real time, and transmits the processed real-time data to the digital twin platform. Specifically:

[0043] Step 3.1: Build a digital twin platform.

[0044] Step 3.2: The sensors deployed on the surface of the soil-covered tank capture the physical signal changes, and the raw sensor data is converted into structured transmission data packets through signal analysis.

[0045] Step 3.3: Establish a real-time data channel between the digital twin platform and the sensor communication module to achieve multi-channel data stream synchronization and data packet timing alignment and integrity verification.

[0046] Step 3.4: The data obtained in step 3.3 are calibrated with an adaptive benchmark, and the raw sensor data are dynamically compensated for deviations to eliminate the system deviation introduced by environmental noise and generate standardized data.

[0047] Step 3.5: Transmit the standardized data obtained in step 3.4 to the digital twin platform and use it as input parameters to drive the load inversion prediction model in step 2.1. Calculate the multidimensional dynamic load distribution characteristics of the earth-covered tank in real time and output the multidimensional dynamic load.

[0048] Step 4: The virtual visualization module uses computer graphics technology to achieve three-dimensional dynamic visualization of the multi-dimensional mechanical state of the earth-covering tank on the digital twin platform. Specifically:

[0049] Step 4.1: Based on the finite element mesh data file of the earth-covering tank obtained in step 1.2, a high-fidelity mesh model of the earth-covering tank is constructed in the digital twin platform through parametric modeling and visual mapping mechanism.

[0050] Step 4.2: Use the multidimensional dynamic loads output by the load inversion prediction model in Step 3.5 as input parameters to drive the mechanical response prediction model group described in Step 2.2. This calculation results in multidimensional mechanical parameter values ​​(including stress, deformation, etc.) at each grid node of the high-fidelity mesh model of the earth-covered tank described in Step 4.1. The stress data at each grid node will serve as the key input for Step 4.3.

[0051] Step 4.3: Based on the stress at each grid node outputted in step 4.2, the dynamic damage evolution prediction model described in step 2.3 is driven to calculate the cumulative damage value of each grid node.

[0052] Step 4.4: On the digital twin platform, implement three-dimensional visualization modeling of the multi-dimensional mechanical state of the earth-covered tank using computer graphics technology. The multi-dimensional mechanical parameter values ​​obtained in Step 4.2 are mapped to the vertex attributes of the three-dimensional model. A high-fidelity cloud map is generated using a color-coding algorithm to achieve three-dimensional dynamic visualization of the multi-dimensional mechanical state. The high-fidelity cloud map includes, but is not limited to, a fatigue damage distribution cloud map, a stress distribution cloud map, and a deformation distribution cloud map. The fatigue damage distribution cloud map is generated based on the cumulative damage values ​​of each mesh node obtained in Step 4.3.

[0053] The beneficial effects of the present invention are:

[0054] (1) The present invention constructs an efficient load prediction and mechanical analysis system based on a small number of sensors. Through the data collected by a small number of sensors, the external load distribution of the earth-covering tank can be predicted with high precision. The prediction results are further used to perform real-time prediction and analysis of the mechanical state of the earth-covering tank, which significantly reduces the hardware deployment cost while ensuring the monitoring accuracy.

[0055] (2) This invention achieves real-time calculation and visualization of the cumulative damage of the earth-covered tank structure, and accurately reflects the health status of the earth-covered tank by generating a high-precision fatigue damage distribution cloud map. By dynamically monitoring fatigue damage, the operation and maintenance model is transformed from passive maintenance to predictive maintenance, thereby significantly extending the service life of the equipment and effectively reducing unplanned downtime and maintenance costs caused by sudden structural failure.

[0056] (3) Based on digital twin technology, the present invention builds a digital twin platform. This platform drives a multi-parameter correlation prediction model group by integrating the mechanical response simulation results under multiple working conditions with real-time sensor data. With the help of computer graphics technology, it accurately presents the real-time mechanical state changes of the earth-covering tank under different working conditions, and realizes three-dimensional dynamic visualization of the multi-dimensional mechanical state of the earth-covering tank, providing important technical support for the intelligent operation and maintenance of the equipment.

[0057] In summary, the present invention can significantly improve the comprehensive monitoring capability and failure warning accuracy of the earth covering tank, support predictive maintenance decisions, and reduce operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a system framework diagram of the present invention.

[0059] Figure 2 Flow chart of the method of the present invention.

[0060] Figure 3 It is a schematic diagram of the three-dimensional geometric model of the earth-covering tank of the present invention.

[0061] In the figure: 1 tank body, 2 connecting pipes, 3 connecting flange, 4 load-bearing support, 5 bottom plate. DETAILED DESCRIPTION

[0062] In order to further describe the technical solution of the present invention in detail, the present invention will be explained in conjunction with the accompanying drawings and specific embodiments. The accompanying drawings and specific embodiments are for illustrative purposes only and do not limit the present invention.

[0063] This embodiment provides a soil-covering tank intelligent monitoring, operation and maintenance platform and method based on digital twin. The soil-covering tank intelligent monitoring, operation and maintenance platform is based on the digital twin platform and integrates the following functional modules: physical geometry simulation module, algorithm prediction module, sensor communication module, virtual visualization module; each functional module exchanges data and works collaboratively through the digital twin platform. The system framework diagram of the soil-covering tank intelligent monitoring, operation and maintenance platform is as follows: Figure 1 As shown, specifically:

[0064] The physical geometry simulation module consists of a physical entity of the earth-covering tank, a three-dimensional geometric model established based on the structural parameters of the physical entity of the earth-covering tank, and a fiber grating sensor deployed on the object entity. It is used to obtain the real state information of the earth-covering tank in real time; a high-fidelity three-dimensional geometric model is constructed based on the structural parameters of the physical entity of the earth-covering tank, and mechanical response data under multiple working conditions is obtained through finite element simulation analysis.

[0065] The algorithm prediction module runs on the digital twin platform, and the algorithm prediction module constructs a multi-parameter correlation prediction model group based on the mechanical response data generated by the physical geometric simulation module under multiple working conditions, and outputs the earth covering tank mechanical performance information that can be rendered. The multi-parameter correlation prediction model group includes a load inversion prediction model, a mechanical response prediction model group, and a dynamic damage evolution prediction model.

[0066] The sensing communication module is connected with the fiber grating sensor deployed on the physical entity of the earth covering tank. The sensing communication module obtains the data collected by the fiber grating sensor in real time (the collected data includes key data such as the strain of the earth covering tank structure), obtains the monitoring data after preprocessing the data (the preprocessing includes denoising processing and feature extraction), and transmits the monitoring data to the digital twin platform through a low-delay and efficient data communication protocol. After receiving the monitoring data, the digital twin platform drives the algorithm prediction module to predict the external load state of the earth covering tank in real time, and realizes dynamic calculation of the mechanical state of the earth covering tank structure.

[0067] The virtual visualization module relies on the digital twin platform and uses computer graphics technology to realize real-time visualization of the multi-dimensional mechanical state of the earth covering tank in a three-dimensional dynamic form on the digital twin platform.

[0068] An intelligent monitoring and operation method for an earth covering tank based on digital twinning, as shown in the flowchart Figure 2 The intelligent monitoring and operation method for the earth covering tank is realized based on the above-mentioned intelligent monitoring and operation platform, and specifically includes the following steps:

[0069] Step 1: Construct a high-fidelity physical geometric simulation module. Based on the physical entity structure of the earth covering tank, a high-fidelity three-dimensional geometric model is established Figure 3 , and the mechanical response data under multiple working conditions is obtained through finite element simulation analysis. The physical entity of the earth covering tank is composed of a tank body 1, a connecting pipeline 2, a pipe flange 3, a load-bearing support 4, and a bottom plate 5, and specifically:

[0070] Step 1.1: Based on the physical entity structure characteristics of the earth covering tank, a three-dimensional geometric model of the earth covering tank is established using a three-dimensional modeling software, and the three-dimensional geometric model is adjusted and optimized to accurately reflect the true structure of the earth covering tank.

[0071] Step 1.2: The three-dimensional geometric model established in step 1.1 is simplified, and the simplified model is divided into regions by a finite element pre-processing software to obtain a grid model, and the numbers and spatial coordinates of the grid nodes are exported to a local data file.

[0072] Step 1.3: Obtain multi-dimensional load parameters as input variables by combining the external loads and environmental factors that the soil covering tank is subjected to in practical applications. Generate a feature sample set in the parameter domain through spatial sampling methods. Perform statics analysis on the soil covering tank based on the feature sample set to identify key factors affecting the structural performance of the soil covering tank.

[0073] Step 1.4: The statics analysis results obtained in step 1.3 show that stress concentration in the soil covering tank structure mainly occurs in the connecting pipe 2 and the pipe flange 3 area, followed by the tank body 1, and the stress level of the load-bearing support 4 and the bottom plate 5 is relatively low, having less impact on the overall mechanical performance. Based on the stress distribution characteristics, select key measurement points on the connecting pipe 2 as fiber Bragg grating sensor placement points. These placement points are located in high stress gradient areas and have high representativeness and monitoring value.

[0074] Step 1.5: Use finite element analysis methods to apply the multi-dimensional load parameters in the feature sample set obtained in step 1.3 to the mesh model divided in step 1.2, and output the mechanical response data of the soil covering tank as a whole and the key measurement points selected in step 1.4 under multiple working conditions.

[0075] Step 1.6: Deploy fiber Bragg grating sensors at the placement points determined in step 1.4, and construct a sensing communication module. Real-time data acquisition is performed through the fiber Bragg grating sensors to obtain physical signals at key positions of the soil covering tank.

[0076] Step 2: The algorithm prediction module constructs a group of multi-parameter correlation prediction models based on the mechanical response data under multiple working conditions provided by the physical geometry simulation module. Specifically:

[0077] Step 2.1: Use the mechanical response data of the key measurement points under multiple working conditions as input variables, and use the multi-dimensional load parameters in the feature sample set obtained in step 1.3 as output responses to construct a load inversion prediction model, achieving multi-dimensional dynamic feedback of the external load state of the soil covering tank.

[0078] Step 2.2: Use the multi-dimensional load parameters in the feature sample set obtained in step 1.3 as input, and use the mechanical response quantities (including stress, deformation, displacement, etc.) of the soil covering tank as output to construct a group of mechanical response prediction models, achieving collaborative prediction of structural responses under multiple working conditions.

[0079] Step 2.3: Based on the group of mechanical response prediction models obtained in step 2.2, construct a dynamic damage evolution prediction model. Specifically:

[0080] Step 2.3.1: Configure a double-buffer queue for each grid node of the grid model in step 1.2 to store the real-time stress data of each node output by the mechanical response prediction model group described in step 2.2 in time series, and realize the parallel execution of real-time data acquisition and damage analysis through the asynchronous processing mechanism.

[0081] Step 2.3.2: Traverse the cached data in step 2.3.1, reconstruct the load sequence, and generate the cyclic load spectrum.

[0082] Step 2.3.3: Based on the cyclic load spectrum reconstructed in step 2.3.2, identify the complete stress cycle and calculate the stress amplitude of each cycle and mean stress . Further calculate the equivalent alternating stress amplitude , in order to eliminate the influence of mean stress on fatigue life.

[0083] Step 2.3.4: Based on the equivalent alternating stress amplitude obtained in step 2.3.3 , combined with the material's stress-life (SN) curve, calculate the material's lifespan at a specific stress amplitude ( ) under the condition of fatigue failure cycle number N;

[0084] Step 2.3.5: Based on Miner's linear cumulative damage theory, a fatigue damage assessment model under multiple stress levels is established to calculate the cumulative damage value D. When the cumulative damage value D>1, it is determined that the earth-covered tank has reached the critical state of fatigue failure.

[0085] Step 3: The sensor communication module captures and processes the data collected by the fiber Bragg grating sensor in step 1 in real time, and transmits the processed real-time data to the digital twin platform. Specifically:

[0086] Step 3.1: Build a digital twin platform.

[0087] Step 3.2: The FBG wavelength offset is monitored by a fiber Bragg grating sensor deployed on the surface of the soil-covered tank. The fiber Bragg grating demodulator converts the wavelength signal into a hexadecimal data packet, which is encapsulated and transmitted in hexadecimal format through the communication protocol.

[0088] Step 3.3: Establish a real-time communication link between the digital twin platform and the fiber Bragg grating interrogator (FBG) in Step 3.2, using an asynchronous, non-blocking I / O architecture to receive high-speed data streams. The system continuously monitors the hexadecimal data packets sent by the interrogator, uses a sliding window algorithm to locate the start identifier of valid data frames, and parses the data packets layer by layer according to a predefined frame structure, achieving timing alignment and integrity verification for multi-channel data.

[0089] Step 3.4: Convert the hexadecimal-coded wavelength raw data obtained in step 3.3 to single-precision floating-point values ​​and perform a reference wavelength calibration. The reference wavelength calibration method involves continuously collecting the first n data packets during system initialization to form a sliding window. The arithmetic mean of the wavelength data within this window is calculated as the initial reference wavelength to eliminate system deviations introduced by environmental noise.

[0090] Reference wavelength The calculation formula is as follows:

[0091]

[0092] in, represents the original wavelength value collected by the fiber Bragg grating sensor in the i-th data packet;

[0093] Step 3.5: Calculate the current wavelength in real time With reference wavelength The wavelength variation is converted into micro-strain value through the micro-strain conversion model, combined with the grating sensitivity coefficient K, and the strain data matrix is ​​output.

[0094] Real-time offset The calculation formula is as follows:

[0095]

[0096] The microstrain conversion is shown as follows:

[0097]

[0098] in, Indicates the micro-strain value of the key measuring point of the soil covering tank;

[0099] Step 3.6: The real-time microstrain measurement values ​​obtained in step 3.5 are transmitted to the digital twin platform and used as input parameters to drive the load inversion prediction model described in step 2.1. The multidimensional dynamic load distribution characteristics of the earth-covered tank are solved in real time, and the multidimensional dynamic load is output.

[0100] Step 4: The virtual visualization module uses computer graphics technology to achieve three-dimensional dynamic visualization of the multi-dimensional mechanical state of the earth-covering tank on the digital twin platform. Specifically:

[0101] Step 4.1: Based on the finite element mesh data file of the earth-covering tank obtained in step 1.2, a high-fidelity mesh model of the earth-covering tank is constructed in the digital twin platform through parametric modeling and visual mapping mechanism.

[0102] Step 4.2: Use the multidimensional dynamic loads output by the load inversion prediction model in Step 3.6 as input parameters to drive the mechanical response prediction model group described in Step 2.2. This calculation results in multidimensional mechanical parameter values ​​(including stress, deformation, etc.) at each grid node of the high-fidelity mesh model of the earth-covered tank described in Step 4.1. The stress data at each grid node will serve as the key input for Step 4.3.

[0103] Step 4.3: Based on the stress at each grid node outputted in step 4.2, the dynamic damage evolution prediction model described in step 2.3 is driven to calculate the cumulative damage value of each grid node.

[0104] Step 4.4: On the digital twin platform, implement three-dimensional visualization modeling of the multi-dimensional mechanical state of the earth-covered tank using computer graphics technology. The multi-dimensional mechanical parameter values ​​obtained in Step 4.2 are mapped to the vertex attributes of the three-dimensional model. A high-fidelity cloud map is generated using a color-coding algorithm to achieve three-dimensional dynamic visualization of the multi-dimensional mechanical state. The high-fidelity cloud map includes, but is not limited to, a fatigue damage distribution cloud map, a stress distribution cloud map, and a deformation distribution cloud map. The fatigue damage distribution cloud map is generated based on the cumulative damage values ​​of each mesh node obtained in Step 4.3.

[0105] The above-described embodiments merely express the implementation methods of the present invention, but should not be understood as limiting the scope of the present invention. It should be pointed out that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, which all fall within the scope of protection of the present invention.

Claims

1. An intelligent monitoring and operation and maintenance platform for earth-covering tanks based on digital twins, characterized in that: The intelligent monitoring and operation and maintenance platform for earth-covering tanks is centered on the digital twin platform and integrates four functional modules, including: physical geometry simulation module, algorithm prediction module, sensor communication module, and virtual visualization module. Each functional module exchanges data and works collaboratively through the digital twin platform. The intelligent monitoring and operation and maintenance platform for earth-covering tanks is specifically: The physical geometry simulation module consists of a physical entity of the earth-covering tank, a three-dimensional geometric model established based on the structural parameters of the physical entity of the earth-covering tank, and sensors deployed on the physical entity of the earth-covering tank. It is used to obtain the real state information of the earth-covering tank in real time. A high-fidelity three-dimensional geometric model is constructed based on the structural parameters of the physical entity of the earth-covering tank, and mechanical response data under multiple working conditions are obtained through finite element simulation analysis. The algorithm prediction module runs on the digital twin platform. The algorithm prediction module builds a multi-parameter correlation prediction model group based on the mechanical response data under multiple working conditions generated by the physical geometry simulation module, and outputs the mechanical performance information of the earth-covering tank that can be rendered; The sensor communication module is connected to the sensor deployed on the physical entity of the earth-covering tank. The sensor communication module obtains the data collected by the sensor in real time, obtains monitoring data after preprocessing the data, and transmits it to the digital twin platform. After receiving the monitoring data, the digital twin platform drives the algorithm prediction module to predict the external load state of the earth-covering tank in real time, thereby realizing dynamic measurement of the structural mechanical state of the earth-covering tank; The virtual visualization module relies on the digital twin platform and uses computer graphics technology to visualize the multi-dimensional mechanical state of the earth-covering tank in a three-dimensional dynamic form on the digital twin platform in real time. In the algorithm prediction module, the multi-parameter correlation prediction model group includes a load inversion prediction model, a mechanical response prediction model group and a dynamic damage evolution prediction model.

2. A digital twin-based intelligent monitoring and operation and maintenance method for earth-covering tanks, characterized in that: The intelligent monitoring and operation and maintenance platform for soil covering tanks according to claim 1 is implemented, including the following steps: Step 1: Construct a high-fidelity physical geometry simulation module. Build a high-fidelity 3D geometric model based on the physical structure of the earth-covered tank, and obtain mechanical response data under multiple working conditions through finite element simulation analysis. Step 2: The algorithm prediction module constructs a multi-parameter correlation prediction model group based on the mechanical response data under multiple working conditions provided by the physical geometry simulation module; Step 3: The sensor communication module captures and processes the data collected by the sensor in step 1 in real time, and transmits the processed real-time data to the digital twin platform; Step 4: The virtual visualization module uses computer graphics technology to realize three-dimensional dynamic visualization of the multi-dimensional mechanical state of the earth-covering tank on the digital twin platform.

3. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 2 is characterized in that: The step 1 is specifically as follows: Step 1.1: Based on the physical structural characteristics of the earth-covered tank, a 3D geometric model of the earth-covered tank is created using 3D modeling software. Parameters are adjusted and the design is optimized to ensure that the 3D geometric model accurately reflects the actual structure of the earth-covered tank. Step 1.2: Simplify the 3D geometric model created in step 1.1, and use finite element pre-processing software to perform regional meshing on the simplified model to obtain a mesh model. The number and spatial coordinates of each mesh node are then exported to a local data file. Step 1.3: Combine the external loads and environmental factors that the earth-covered tank is subjected to in actual application to obtain multidimensional load parameters, use them as input variables, and generate a feature sample set within the parameter domain through spatial sampling methods; Static analysis of the earth-covered tank is performed based on the characteristic sample set to identify the key factors affecting the structural performance of the earth-covered tank; Step 1.4: Based on the static analysis results obtained in step 1.3, select key measuring points on the physical entity of the earth-covered tank as sensor layout points; Step 1.5: Using the finite element analysis method, apply the multidimensional load parameters in the characteristic sample set obtained in step 1.3 to the mesh model divided in step 1.2, and output the mechanical response data of the entire earth-covered tank and the key measurement points selected in step 1.4 under multiple working conditions; Step 1.6: Deploy sensors at the sensor deployment points determined in step 1.4 and build the sensor communication module; Data is collected in real time through sensors to obtain physical signals at key positions of the covering tank.

4. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 3 is characterized in that: The step 2 is specifically as follows: Step 2.1: Use the mechanical response data of key measuring points under multiple working conditions obtained in step 1.5 as input variables and the multi-dimensional load parameters in the characteristic sample set obtained in step 1.3 as output responses to construct a load inversion prediction model to achieve multi-dimensional dynamic feedback of the external load state of the earth-covering tank; Step 2.2: Use the multi-dimensional load parameters in the characteristic sample set obtained in step 1.3 as input and the mechanical response of the earth-covered tank as output to construct a mechanical response prediction model group to achieve collaborative prediction of structural response under multiple load conditions; Step 2.3: Based on the mechanical response prediction model group obtained in step 2.2, a dynamic damage evolution prediction model is constructed.

5. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 4 is characterized in that: The step 2.3 is specifically as follows: Step 2.3.1: Configure a double-buffer queue for each grid node of the grid model in step 1.2 to store the real-time stress data of each node output by the mechanical response prediction model group in step 2.2 in a time series manner. This allows for parallel execution of real-time data acquisition and damage analysis through an asynchronous processing mechanism. Step 2.3.2: Traverse the cached data in step 2.3.1, reconstruct the load sequence, and generate the cyclic load spectrum; Step 2.3.3: Based on the cyclic load spectrum reconstructed in step 2.3.2, identify the complete stress cycle and calculate the stress amplitude of each cycle and mean stress ; Further calculate the equivalent alternating stress amplitude , as shown below: ； in, is the tensile strength of the material; Step 2.3.4: Based on the equivalent alternating stress amplitude obtained in step 2.3.3 , combined with the material's stress-life curve, calculate the number of cycles N for fatigue failure of the material under a specific stress amplitude, which is in the following form: ； Among them, C and k are material-related constants; C represents the proportional coefficient of material fatigue performance, and k represents the attenuation rate of material fatigue strength with stress amplitude; Step 2.3.5: Based on Miner's linear cumulative damage theory, establish a fatigue damage assessment model under multiple stress levels and calculate the cumulative damage value D as follows: ； Where m represents the total number of divided stress levels; Indicates the actual number of cycles under the stress level i; N i Indicates the fatigue life determined based on the SN curve at this stress level; When the cumulative damage value D>1, it is judged that the earth-covered tank has reached the critical state of fatigue failure.

6. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 5 is characterized in that: In step 2.3.3: The stress amplitude The calculation formula is: ； The mean stress The calculation formula is: ； in, Indicates the maximum stress value in the current cycle; Indicates the minimum stress value in the current cycle.

7. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 5 is characterized in that: The step 3 is specifically as follows: Step 3.1: Build a digital twin platform; Step 3.2: Sensors deployed on the surface of the earth-covered tank capture physical signal changes, and the raw sensor data is converted into structured transmission data packets through signal analysis; Step 3.3: Establish a real-time data channel between the digital twin platform and the sensor communication module to achieve multi-channel data stream synchronization and data packet timing alignment and integrity verification; Step 3.4: Adaptively calibrate the data obtained in step 3.3 and dynamically compensate for the original sensor data to eliminate the system deviation introduced by environmental noise and generate standardized data. Step 3.5: Transmit the standardized data obtained in step 3.4 to the digital twin platform and use it as input parameters to drive the load inversion prediction model in step 2.

1. Calculate the multidimensional dynamic load distribution characteristics of the earth-covered tank in real time and output the multidimensional dynamic load.

8. The method for intelligent monitoring and operation and maintenance of earth-covering tanks based on digital twins according to claim 7 is characterized in that: The step 4 is specifically as follows: Step 4.1: Based on the finite element mesh data file of the earth-covering tank obtained in step 1.2, a high-fidelity mesh model of the earth-covering tank is constructed in the digital twin platform through parametric modeling and visual mapping mechanism; Step 4.2: Use the multidimensional dynamic load output by the load inversion prediction model in step 3.5 as an input parameter to drive the mechanical response prediction model group described in step 2.2 to calculate the multidimensional mechanical parameter values ​​at each grid node of the high-fidelity grid model of the earth-covered tank described in step 4.

1. The stress data of each grid node will serve as the key input for step 4.

3. Step 4.3: Based on the stress at each grid node outputted in step 4.2, the dynamic damage evolution prediction model described in step 2.3 is driven to calculate the cumulative damage value of each grid node; Step 4.4: In the digital twin platform, use computer graphics technology to achieve three-dimensional visualization modeling of the multi-dimensional mechanical state of the earth-covered tank; map the multi-dimensional mechanical parameter values ​​obtained in step 4.2 to the vertex attributes of the three-dimensional model, and combine the color coding algorithm to generate a high-fidelity cloud map to achieve three-dimensional dynamic visualization of the multi-dimensional mechanical state.

Images