Large aqueduct intelligent temperature control system and method based on digital twinning technology

Abstract

The application provides a large aqueduct intelligent temperature control system and method based on digital twinning technology, relates to the technical field of aqueduct temperature monitoring, and through the construction of a three-dimensional digital model of the aqueduct and the correction of the geometric parameters of the three-dimensional digital model through the three-dimensional point cloud data of the aqueduct, the simulation result is obtained through the thermodynamic simulation calculation based on the three-dimensional digital model, the simulation result of the aqueduct concrete temperature field is compared with the collected temperature of the aqueduct, the temperature deviation is obtained, the parameters of the thermodynamic simulation calculation are corrected according to the temperature deviation, so that the simulation result is updated, the accuracy of the simulation result is improved, the water flow of the corresponding water pipe is controlled according to the simulation result after the correction of the geometric parameters of the three-dimensional digital model and the correction of the parameters of the thermodynamic simulation calculation, and the cooling is realized, the problem that the prior art cannot be targeted to the local hydration heat and the targeted cooling is solved, and the application is suitable for the intelligent temperature control of the hydration heat process of a large aqueduct.

Description

Technical Field

[0001] This invention relates to the field of aqueduct temperature monitoring technology, and in particular to a large-scale intelligent temperature control system and method for aqueducts based on digital twin technology. Background Technology

[0002] Large aqueducts refer to aqueducts where, without temperature control, harmful cracks can develop due to temperature changes and shrinkage caused by the heat of hydration. During construction and curing, the heat of hydration is difficult to control. The hydration of cement releases heat, and the large-volume concrete structure has a relatively thick cross-section and a small surface area coefficient, causing the heat generated by the cement to accumulate inside the structure and not easily dissipate. This prevents the heat of hydration inside the concrete from dissipating in time, leading to an increased temperature difference between the inside and outside, and ultimately, cracks. This negatively impacts the aqueduct's impermeability, durability, and safety.

[0003] Traditional cooling methods primarily involve pre-embedding cooling pipes inside the structure and then circulating cold water through these pipes to lower the internal temperature of large-volume concrete. While this method can reduce the internal temperature of large-volume concrete, it has certain drawbacks. The cooling pipe material has relatively low strength and a smooth outer surface, leaving channels in the concrete structure after it has hardened. This can lead to stress concentration during later stress processes, reducing the structure's safety and reliability. Furthermore, it cannot achieve automated control of the cooling system, lacks effective solutions for addressing excessive localized heat of hydration within the concrete itself, and fails to develop targeted cooling measures, resulting in poor cooling performance and some waste. Summary of the Invention

[0004] The technical problem solved by this invention: This invention provides a large-scale aqueduct intelligent temperature control system and method based on digital twin technology, which solves the problem that existing technologies cannot specifically cool down areas with excessive heat of hydration.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is: a large-scale aqueduct intelligent temperature control system based on digital twin technology, wherein the aqueduct is equipped with prestressed pipes, and water pipes are installed in the prestressed pipes; the system includes:

[0006] The information acquisition subsystem is used to collect the temperature of the aqueduct.

[0007] A digital twin model construction subsystem is used to construct a three-dimensional digital model of the aqueduct based on construction drawings;

[0008] The simulation calculation subsystem is used to perform thermodynamic simulation calculations based on a three-dimensional digital model and obtain simulation results, including simulation results of the temperature field of the aqueduct concrete.

[0009] The feedback correction subsystem is used to correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct, and to compare the simulation results of the temperature field of the aqueduct concrete with the collected temperature of the aqueduct to obtain the temperature deviation, and to correct the parameters calculated by thermodynamic simulation based on the temperature deviation.

[0010] The temperature control subsystem controls the water flow rate of the corresponding water pipes to cool the water based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation.

[0011] Furthermore, the information acquisition subsystem also includes an information processing subsystem, which is used to denoise and fill in missing values ​​on the data acquired by the information acquisition subsystem;

[0012] Noise reduction includes the following steps:

[0013] S01. Normalize the historical data. The normalization formula is: ,in, It is historical data. It is the minimum value in the historical data collection. It is the maximum value in the historical data collection. This represents the normalized historical data.

[0014] S02. Construct a supervised learning dataset, which includes an input sequence and a target output. The input sequence is noisy data of a preset length from historical data collection, and the target output is denoised data at the next time step.

[0015] S03. Construct a first long short-term memory network model, which includes a forget gate, an input gate, a state update gate, an output gate, and a fully connected layer.

[0016] S04. Using mean squared error as the loss function, the Adam adaptive learning rate optimizer is used to minimize the loss function to supervise the learning of the input sequence and target output in the dataset. The parameters of the first long short-term memory network are iteratively updated through the backpropagation algorithm and optimizer until the loss function converges, and a trained long short-term memory network model for denoising is obtained.

[0017] S05. Use the trained long short-term memory network model to denoise the data collected by the information collection subsystem to obtain denoised normalized data, and then obtain denoised data through inverse normalization.

[0018] Missing value completion includes the following steps:

[0019] S001. In the original time series data, mark missing values ​​and normalize them. Use the sliding window method to divide the normalized time series data into multiple subsequences.

[0020] S002, Construct a second long short-term memory network model;

[0021] S003. Using mean squared error as the loss function, the weights and biases of the second long short-term memory network model are updated through the backpropagation algorithm to minimize the loss function and obtain the long short-term memory network model for missing value completion.

[0022] S004. For each missing value, using the data from the preceding and following time steps as input, the missing value is predicted through the Long Short-Term Memory Network Model for missing value completion until all missing values ​​are filled, thus obtaining the completed normalized time series data.

[0023] S005. Perform inverse normalization on the completed normalized time series data to obtain the completed original time series data.

[0024] Further thermodynamic simulation calculations include the following steps:

[0025] S11. Construct a finite element model in Midas FEA / NX based on the aforementioned three-dimensional digital model;

[0026] S12. Apply boundary conditions, including constraint boundaries and convection boundaries;

[0027] S13. Using the concrete pouring temperature as the initial temperature field, set the heat source function as follows: ,in, Indicates time, This indicates the adiabatic temperature rise of the concrete at time t. This indicates the maximum adiabatic temperature rise of concrete. Represents the natural constant. Indicates the heat transfer coefficient. Indicates the start-up time of the heat source. , This indicates the amount of gel material used per cubic meter. This represents the heat of hydration per kilogram of cement. This indicates the adjustment coefficient for the heat of hydration of the admixture. This indicates the specific heat capacity of concrete. Indicates the mass density of concrete;

[0028] S14. Define the calculation formula for heat exchange in cooling water pipes as follows: ,in, Indicates temperature. Indicates time, Represents the standard heat conduction term. Indicates the thermal conductivity of concrete. Represents three-dimensional spatial coordinates. This indicates the equivalent negative heat source term for water pipe cooling. Indicates the initial temperature of the concrete. Indicates water temperature. , Indicates time, , , This represents the heat transfer coefficient of the concrete-water pipe contact surface. Indicates the spacing between water pipes. Indicates the outer diameter of the water pipe. Indicates the inner diameter of the water pipe. Indicates the thermal conductivity of concrete. Indicates the thermal conductivity of the water pipe. This indicates the heat of hydration of concrete. , This represents the heat release coefficient of concrete. The coefficient represents the intensity of the influence of the heat of hydration on the temperature field per unit time. Indicates temperature Regarding time The first-order partial derivative;

[0029] S15. Define the heat dissipation formula for concrete surface as follows: ,in, This indicates the amount of heat dissipated from the concrete surface. This represents the heat transfer coefficient of the concrete surface. Represents the surface area of ​​concrete. Indicates the surface temperature of the concrete. Indicates ambient temperature. , Indicates wind speed;

[0030] S16. Set the calculation time step and output the simulation results of the temperature field of the aqueduct concrete.

[0031] Furthermore, in S15, if there is an insulation layer on the concrete surface, the convection coefficient is adjusted, and the adjusted convection coefficient is: , ,in, Indicates the first The thickness of the insulation layer, Indicates the first The thermal conductivity of the insulation layer.

[0032] Furthermore, the simulation results also include simulation results of the temperature stress field of the aqueduct concrete, and the thermodynamic simulation calculations also include:

[0033] S17. Define the temperature-dependent elastic modulus and coefficient of thermal expansion, perform temperature-stress coupling analysis, and obtain the simulation results of the temperature stress field of the aqueduct concrete. The temperature stress is the stress caused by temperature change and temperature difference between the inside and outside of the concrete. Analyze the simulation results of the temperature stress field to obtain the location of potential cracks.

[0034] Furthermore, the geometric parameters of the 3D digital model are corrected using the 3D point cloud data of the aqueduct, including the following steps:

[0035] S21. Set up fixed targets or control points on the aqueduct and obtain the three-dimensional point cloud data of the aqueduct through a three-dimensional laser scanning device.

[0036] S22. Using fixed targets or control points, the acquired three-dimensional point cloud data of the aqueduct is transformed and registered to the unified coordinate system of the three-dimensional digital model through coordinate transformation.

[0037] S23. Calculate the positional deviation between points in the point cloud and corresponding points in the 3D digital model;

[0038] S24. Correct the geometric parameters of the three-dimensional digital model based on the positional deviation.

[0039] Furthermore, in S22, the ICP algorithm is used to find the rotation matrix and translation vector so that after the three-dimensional point cloud data of the aqueduct is transformed to the unified coordinate system of the three-dimensional digital model by the rotation matrix and translation vector, the overall positional deviation between the points in the three-dimensional point cloud data and the corresponding points in the three-dimensional digital model is minimized.

[0040] Furthermore, the parameters calculated by thermodynamic simulation are corrected based on the temperature deviation, including the following steps:

[0041] S31. When the temperature deviation exceeds the preset threshold, adjust the thermal conductivity or specific heat capacity of the concrete, re-perform the thermodynamic simulation calculation, obtain the new concrete temperature field simulation result, and then compare it with the collected temperature of the aqueduct to obtain the new temperature deviation. If the new temperature deviation is greater than the temperature deviation, adjust the thermal conductivity or specific heat capacity of the concrete in the opposite direction.

[0042] S32. Repeat S31 until the temperature deviation does not exceed the preset threshold or the maximum number of adjustments is reached to obtain the latest simulation results.

[0043] Furthermore, based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation, the water flow rate of the corresponding water pipe is controlled for cooling. This includes: analyzing the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation to obtain the local temperature and local temperature difference. If the local temperature exceeds the first threshold, the water flow rate in the corresponding water pipe is increased. If the local temperature difference is greater than the second threshold, the water flow rate in the corresponding water pipe with the larger local temperature difference is increased.

[0044] This invention also provides a method for intelligent temperature control of large aqueducts based on digital twin technology, applied to the aforementioned intelligent temperature control system for large aqueducts based on digital twin technology. The method includes the following steps:

[0045] S101. Construct a three-dimensional digital model of the aqueduct based on the construction drawings, and correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct.

[0046] S102. Perform thermodynamic simulation calculations based on a three-dimensional digital model to obtain simulation results, including simulation results of the temperature field of the aqueduct concrete.

[0047] S103. Collect the temperature of the aqueduct, compare the simulation results of the temperature field of the aqueduct concrete with the collected temperature of the aqueduct, obtain the temperature deviation, correct the parameters of the thermodynamic simulation calculation based on the temperature deviation, and obtain the simulation results after correcting the parameters of the thermodynamic simulation calculation.

[0048] S104. Based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation, control the water flow rate of the corresponding water pipe to cool down.

[0049] The beneficial effects of this invention are as follows: This invention provides a large-scale intelligent temperature control system and method for aqueducts based on digital twin technology. It constructs a three-dimensional digital model of the aqueduct and corrects the geometric parameters of the three-dimensional digital model using three-dimensional point cloud data of the aqueduct. Thermodynamic simulation calculations are performed based on the three-dimensional digital model to obtain simulation results. The simulation results of the aqueduct concrete temperature field are compared with the collected aqueduct temperature to obtain the temperature deviation. The parameters of the thermodynamic simulation calculations are corrected based on the temperature deviation, thereby updating the simulation results and improving their accuracy. Based on the corrected geometric parameters of the three-dimensional digital model and the corrected thermodynamic simulation calculation parameters, the water flow rate of the corresponding water pipes is controlled to achieve localized targeted cooling, solving the problem that existing technologies cannot target localized excessive heat of hydration for targeted cooling. Attached Figure Description

[0050] Figure 1This is a schematic diagram of a large-scale intelligent temperature control system for aqueducts based on digital twin technology, provided by the present invention.

[0051] Figure 2 This is a schematic diagram of the aqueduct entity in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention;

[0052] Figure 3 This is a schematic diagram of an aqueduct with water pipes arranged in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention.

[0053] Among them, 1 represents the inlet pipe, 2 represents the U-shaped water pipe, 3 represents the outlet pipe, 4 represents the outlet pipe connection pipe, 5 represents the prestressed pipe, 6 represents the water tank, 7 represents the water pump, and 8 represents the T-shaped pipe.

[0054] Figure 4 This is a schematic diagram of the noise reduction process in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention;

[0055] Figure 5 This is a schematic diagram of the missing value completion process in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention.

[0056] Figure 6 This is a schematic diagram of the process for correcting the geometric parameters of a three-dimensional digital model in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention.

[0057] Figure 7 This is a schematic diagram of the parameter flow for corrected thermodynamic simulation calculation in a large-scale aqueduct intelligent temperature control system based on digital twin technology provided by the present invention.

[0058] Figure 8 This is a schematic diagram of a method for intelligent temperature control of large aqueducts based on digital twin technology provided by the present invention. Detailed Implementation

[0059] This invention addresses the problem of existing technologies failing to effectively cool areas with excessive localized heat of hydration. It provides a large-scale intelligent temperature control system for aqueducts based on digital twin technology. The aqueduct is equipped with prestressed pipes, and water pipes are installed within these prestressed pipes. The system is as follows: Figure 1As shown, the system includes an information acquisition subsystem, a digital twin model construction subsystem, a simulation calculation subsystem, a feedback correction subsystem, and a temperature control subsystem. The information acquisition subsystem is used to acquire the temperature of the aqueduct. The digital twin model construction subsystem is used to construct a three-dimensional digital model of the aqueduct based on construction drawings. The simulation calculation subsystem is used to perform thermodynamic simulation calculations based on the three-dimensional digital model to obtain simulation results, including the simulation results of the aqueduct concrete temperature field. The feedback correction subsystem is used to correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct, and to compare the simulation results of the aqueduct concrete temperature field with the acquired aqueduct temperature to obtain the temperature deviation, and to correct the parameters of the thermodynamic simulation calculation based on the temperature deviation. The temperature control subsystem controls the water flow of the corresponding water pipes to cool the aqueduct based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation.

[0060] Specifically, the aqueduct entity such as Figure 2 As shown, an aqueduct with water pipes is arranged as follows: Figure 3 As shown, the water pump 7 draws cooling water from the water tank 6 and sends it into the inlet pipe 1. The cooling water in the inlet pipe 1 passes through multiple U-shaped water pipes 2 to cool the inside of the aqueduct before reaching the outlet pipe 3. Multiple T-shaped pipes 8 are installed on the outlet pipe 3. Spray devices are installed at the two outlet ends of the T-shaped pipes 8. Each U-shaped water pipe 2 is equipped with an intelligent water valve to control the water flow in the U-shaped water pipe. Intelligent water valves are also installed at both ends of each T-shaped pipe 8 to control the amount of water sprayed by the spray device, thereby cooling the surface of the aqueduct.

[0061] Specifically, the information acquisition subsystem includes multiple temperature sensors arranged inside the aqueduct to detect the temperature of the aqueduct. To ensure the accuracy of the temperature data, the information acquisition subsystem also includes an information processing subsystem, which is used to denoise and fill in missing values ​​on the data acquired by the information acquisition subsystem.

[0062] Noise reduction, such as Figure 4 As shown, it includes the following steps:

[0063] S01. Normalize the historical data. The normalization formula is: ,in, It is historical data. It is the minimum value in the historical data collection. It is the maximum value in the historical data collection. This represents the normalized historical data.

[0064] S02. Construct a supervised learning dataset, which includes an input sequence and a target output. The input sequence is noisy data of a preset length from historical data collection, and the target output is denoised data at the next time step.

[0065] S03. Construct a first long short-term memory (LSTM) network model. This model includes a forget gate, an input gate, a state update gate, an output gate, and a fully connected layer. The forget gate determines which information to discard, the input gate determines which information to store, the state update gate combines the information from the forget gate and the input gate to update the state, and the output gate determines the output based on the current input and cell state.

[0066] S04. Using mean squared error as the loss function, the Adam adaptive learning rate optimizer is used to minimize the loss function to supervise the learning of the input sequence and target output in the dataset. The parameters of the first long short-term memory network are iteratively updated through the backpropagation algorithm and optimizer until the loss function converges, and a trained long short-term memory network model for denoising is obtained.

[0067] S05. Use the trained long short-term memory network model to denoise the data collected by the information collection subsystem to obtain denoised normalized data, and then obtain denoised data through inverse normalization.

[0068] Missing value completion, such as Figure 5 As shown, it includes the following steps:

[0069] S001. In the original time series data, mark missing values ​​and normalize them. Use the sliding window method to divide the normalized time series data into multiple subsequences.

[0070] S002, Construct a second long short-term memory network model;

[0071] S003. Using mean squared error as the loss function, the weights and biases of the second long short-term memory network model are updated through the backpropagation algorithm to minimize the loss function and obtain the long short-term memory network model for missing value completion.

[0072] S004. For each missing value, using the data from the preceding and following time steps as input, the missing value is predicted through the Long Short-Term Memory Network Model for missing value completion until all missing values ​​are filled, thus obtaining the completed normalized time series data.

[0073] S005. Perform inverse normalization on the completed normalized time series data to obtain the completed original time series data.

[0074] Specifically, thermodynamic simulation calculations include the following steps:

[0075] S11. Construct a finite element model in Midas FEA / NX based on the three-dimensional digital model.

[0076] S12. Apply boundary conditions, which include constraint boundaries and convection boundaries; the constraint boundaries include the steel formwork constraint at the bottom of the aqueduct, and the convection boundaries include the convection between the surface of the concrete and the fluid in contact with it.

[0077] S13. Using the concrete pouring temperature as the initial temperature field, set the heat source function as follows: ,in, Indicates time, This indicates the adiabatic temperature rise of the concrete at time t. This indicates the maximum adiabatic temperature rise of concrete. Represents the natural constant. Indicates the heat transfer coefficient. Indicates the start-up time of the heat source. , This indicates the amount of gel material used per cubic meter. This represents the heat of hydration per kilogram of cement. This indicates the adjustment coefficient for the heat of hydration of the admixture. This indicates the specific heat capacity of concrete. This indicates the mass density of concrete.

[0078] S14. Define the calculation formula for heat exchange in cooling water pipes as follows: ,in, Indicates temperature. Indicates time, Represents the standard heat conduction term. Indicates the thermal conductivity of concrete. Represents three-dimensional spatial coordinates. This indicates the equivalent negative heat source term for water pipe cooling. Indicates the initial temperature of the concrete. Indicates water temperature. , Indicates time, , , This represents the heat transfer coefficient of the concrete-water pipe contact surface. Indicates the spacing between water pipes. Indicates the outer diameter of the water pipe. Indicates the inner diameter of the water pipe. Indicates the thermal conductivity of concrete. Indicates the thermal conductivity of the water pipe. This indicates the heat of hydration of concrete. , This represents the heat release coefficient of concrete. The coefficient represents the intensity of the influence of the heat of hydration on the temperature field per unit time. Indicates temperature Regarding time The first partial derivative of , in physical terms, is the rate of change of the temperature of concrete at a fixed spatial location with time, that is, the amount of temperature change at that point per unit time, and is used to describe the changing characteristics of the temperature field.

[0079] S15. Define the heat dissipation formula for concrete surface as follows: ,in, This indicates the amount of heat dissipated from the concrete surface. This represents the heat transfer coefficient of the concrete surface. Represents the surface area of ​​concrete. Indicates the surface temperature of the concrete. Indicates ambient temperature. , This indicates wind speed; if the concrete surface has an insulation layer, the convection coefficient should be adjusted accordingly. The adjusted convection coefficient is... , ,in, Indicates the first The thickness of the insulation layer, Indicates the first The thermal conductivity of the insulation layer.

[0080] S16. Set the calculation time step and output the simulation results of the temperature field of the aqueduct concrete.

[0081] To better illustrate the impact of temperature on the aqueduct, the simulation results also include simulation results of the temperature stress field of the aqueduct concrete. The thermodynamic simulation calculations further include: S17, defining the temperature-dependent elastic modulus and coefficient of thermal expansion, performing temperature-stress coupling analysis, and obtaining the simulation results of the temperature stress field of the aqueduct concrete. The temperature stress is the stress caused by temperature changes and the temperature difference between the inside and outside of the concrete. The simulation results of the temperature stress field are then analyzed to obtain the locations of potential cracks. In this way, the cooling effect is characterized by the locations of potential cracks and visualized.

[0082] Specifically, the geometric parameters of the 3D digital model are corrected using the 3D point cloud data of the aqueduct, such as... Figure 6 As shown, it includes the following steps:

[0083] S21. Set up fixed targets or control points on the aqueduct and obtain the three-dimensional point cloud data of the aqueduct through a three-dimensional laser scanning device.

[0084] S22. Using fixed targets or control points, the acquired 3D point cloud data of the aqueduct is transformed and registered to the unified coordinate system of the 3D digital model through coordinate transformation. The ICP algorithm is used to find the rotation matrix and translation vector so that after the 3D point cloud data of the aqueduct is transformed to the unified coordinate system of the 3D digital model, the overall positional deviation between the points in the 3D point cloud data and the corresponding points in the 3D digital model is minimized.

[0085] S23. Calculate the positional deviation between points in the point cloud and corresponding points in the 3D digital model. The positional deviation can be displayed intuitively through a color chart.

[0086] S24. Correct the geometric parameters of the three-dimensional digital model based on the positional deviation. For example, if the diameter of a certain section of the trough is generally larger than the diameter of that section of the trough in the three-dimensional digital model, it can be concluded that the positional deviation is not caused by construction error or scanning noise, but by the inaccuracy of the diameter of that section of the trough in the three-dimensional digital model. Correction is then made so that the geometric shape of the three-dimensional digital model is consistent with the actual aqueduct.

[0087] Specifically, the parameters of the thermodynamic simulation calculation are corrected based on the temperature deviation, such as... Figure 7 As shown, it includes the following steps:

[0088] S31. When the temperature deviation exceeds the preset threshold, adjust the thermal conductivity or specific heat capacity of the concrete, re-perform the thermodynamic simulation calculation, obtain the new concrete temperature field simulation results, and then compare them with the collected temperature of the aqueduct to obtain the new temperature deviation. If the new temperature deviation is greater than the temperature deviation, then adjust the thermal conductivity or specific heat capacity of the concrete in the opposite direction.

[0089] S32. Repeat S31 until the temperature deviation does not exceed the preset threshold or the maximum number of adjustments is reached to obtain the latest simulation results. In this way, by collecting the temperature of the aqueduct, the thermal conductivity or specific heat capacity of concrete in the thermodynamic simulation calculation is updated, thereby updating the simulation results and improving their accuracy.

[0090] Specifically, cooling is achieved by controlling the water flow rate in the corresponding water pipes based on the simulation results after correcting the geometric parameters of the 3D digital model and the parameters of the thermodynamic simulation calculations. This includes analyzing the simulation results after correcting the geometric parameters of the 3D digital model and the parameters of the thermodynamic simulation calculations to obtain the local temperature and local temperature difference. If the local temperature exceeds a first threshold, the water flow rate in the water pipe passing through that local area is increased. If the local temperature difference is greater than a second threshold, the water flow rate in the water pipe passing through the area with the higher local temperature difference is increased. This achieves targeted cooling of the local hydration heat. Water in the water pipes can also be drawn out through a spray device to spray and cool areas with higher surface temperatures in the aqueduct.

[0091] This invention also provides a method for intelligent temperature control of large aqueducts based on digital twin technology, applied to the aforementioned intelligent temperature control system for large aqueducts using digital twin technology. The method, as described above... Figure 8 As shown, it includes the following steps:

[0092] S101. Construct a three-dimensional digital model of the aqueduct based on the construction drawings, and correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct.

[0093] S102. Perform thermodynamic simulation calculations based on a three-dimensional digital model to obtain simulation results, including simulation results of the temperature field of the aqueduct concrete.

[0094] S103. Collect the temperature of the aqueduct, compare the simulation results of the temperature field of the aqueduct concrete with the collected temperature of the aqueduct, obtain the temperature deviation, correct the parameters of the thermodynamic simulation calculation based on the temperature deviation, and obtain the simulation results after correcting the parameters of the thermodynamic simulation calculation.

[0095] S104. Based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation, control the water flow rate of the corresponding water pipe to cool down.

[0096] This invention corrects the geometric parameters of a 3D digital model using 3D point cloud data of the aqueduct, obtaining a 3D digital model that is consistent with the actual aqueduct. Based on this 3D digital model, thermodynamic simulation is performed. By collecting the temperature of the aqueduct, the thermal conductivity or specific heat capacity of concrete in the thermodynamic simulation calculation is updated, thereby updating the simulation results and improving their accuracy. The simulation results are analyzed to identify local overheating of hydration heat, and targeted cooling is carried out to avoid cracking as much as possible.

Claims

1. A large-scale aqueduct intelligent temperature control system based on digital twin technology, characterized in that, The aqueduct is equipped with prestressed pipes, and water pipes are installed in the prestressed pipes. The system includes: The information acquisition subsystem is used to collect the temperature of the aqueduct. A digital twin model construction subsystem is used to construct a three-dimensional digital model of the aqueduct based on construction drawings; The simulation calculation subsystem is used to perform thermodynamic simulation calculations based on a three-dimensional digital model and obtain simulation results, including simulation results of the temperature field of the aqueduct concrete. The feedback correction subsystem is used to correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct, and to compare the simulation results of the temperature field of the aqueduct concrete with the collected temperature of the aqueduct to obtain the temperature deviation, and to correct the parameters calculated by thermodynamic simulation based on the temperature deviation. The temperature control subsystem controls the water flow rate of the corresponding water pipes to cool the water based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation.

2. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 1, characterized in that, The information acquisition subsystem also includes an information processing subsystem, which is used to denoise and fill in missing values ​​in the data acquired by the information acquisition subsystem. Noise reduction includes the following steps: S01. Normalize the historical data. The normalization formula is: ,in, It is historical data. It is the minimum value in the historical data collection. It is the maximum value in the historical data collection. This represents the normalized historical data. S02. Construct a supervised learning dataset, which includes an input sequence and a target output. The input sequence is noisy data of a preset length from historical data collection, and the target output is denoised data at the next time step. S03. Construct a first long short-term memory network model, which includes a forget gate, an input gate, a state update gate, an output gate, and a fully connected layer. S04. Using mean squared error as the loss function, the Adam adaptive learning rate optimizer is used to minimize the loss function to supervise the learning of the input sequence and target output in the dataset. The parameters of the first long short-term memory network are iteratively updated through the backpropagation algorithm and optimizer until the loss function converges, and a trained long short-term memory network model for denoising is obtained. S05. Use the trained long short-term memory network model to denoise the data collected by the information collection subsystem to obtain denoised normalized data, and then obtain denoised data through inverse normalization. Missing value completion includes the following steps: S001. In the original time series data, mark missing values ​​and normalize them. Use the sliding window method to divide the normalized time series data into multiple subsequences. S002, Construct a second long short-term memory network model; S003. Using mean squared error as the loss function, the weights and biases of the second long short-term memory network model are updated through the backpropagation algorithm to minimize the loss function and obtain the long short-term memory network model for missing value completion. S004. For each missing value, using the data from the preceding and following time steps as input, the missing value is predicted through the Long Short-Term Memory Network Model for missing value completion until all missing values ​​are filled, thus obtaining the completed normalized time series data. S005. Perform inverse normalization on the completed normalized time series data to obtain the completed original time series data.

3. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 1, characterized in that, Thermodynamic simulation calculations include the following steps: S11. Construct a finite element model in Midas FEA / NX based on the aforementioned three-dimensional digital model; S12. Apply boundary conditions, including constraint boundaries and convection boundaries; S13. Using the concrete pouring temperature as the initial temperature field, set the heat source function as follows: ,in, Indicates time, This indicates the adiabatic temperature rise of the concrete at time t. This indicates the maximum adiabatic temperature rise of concrete. Represents the natural constant. Indicates the heat transfer coefficient. Indicates the start-up time of the heat source. , This indicates the amount of gel material used per cubic meter. This represents the heat of hydration per kilogram of cement. This indicates the adjustment coefficient for the heat of hydration of the admixture. This indicates the specific heat capacity of concrete. Indicates the mass density of concrete; S14. Define the calculation formula for heat exchange in cooling water pipes as follows: ,in, Indicates temperature. Indicates time, Represents the standard heat conduction term. Indicates the thermal conductivity of concrete. Represents three-dimensional spatial coordinates. This indicates the equivalent negative heat source term for water pipe cooling. Indicates the initial temperature of the concrete. Indicates water temperature. , Indicates time, , , This represents the heat transfer coefficient of the concrete-water pipe contact surface. Indicates the spacing between water pipes. Indicates the outer diameter of the water pipe. Indicates the inner diameter of the water pipe. Indicates the thermal conductivity of concrete. Indicates the thermal conductivity of the water pipe. This indicates the heat of hydration of concrete. , This represents the heat release coefficient of concrete. The coefficient represents the intensity of the influence of the heat of hydration on the temperature field per unit time. Indicates temperature Regarding time The first-order partial derivative; S15. Define the heat dissipation formula for concrete surface as follows: ,in, This indicates the amount of heat dissipated from the concrete surface. This represents the heat transfer coefficient of the concrete surface. Represents the surface area of ​​concrete. Indicates the surface temperature of the concrete. Indicates ambient temperature. , Indicates wind speed; S16. Set the calculation time step and output the simulation results of the temperature field of the aqueduct concrete.

4. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 3, characterized in that, In S15, if the concrete surface has an insulation layer, the convection coefficient is adjusted, and the adjusted convection coefficient is: , ,in, Indicates the first The thickness of the insulation layer, Indicates the first The thermal conductivity of the insulation layer.

5. A large-scale aqueduct intelligent temperature control system based on digital twin technology according to claim 4, characterized in that, The simulation results also include simulation results of the temperature stress field of the aqueduct concrete, and the thermodynamic simulation calculations also include: S17. Define the temperature-dependent elastic modulus and coefficient of thermal expansion, perform temperature-stress coupling analysis, and obtain the simulation results of the temperature stress field of the aqueduct concrete. The temperature stress is the stress caused by temperature change and temperature difference between the inside and outside of the concrete. Analyze the simulation results of the temperature stress field to obtain the location of potential cracks.

6. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 1, characterized in that, The geometric parameters of the 3D digital model are corrected using the 3D point cloud data of the aqueduct, including the following steps: S21. Set up fixed targets or control points on the aqueduct and obtain the three-dimensional point cloud data of the aqueduct through a three-dimensional laser scanning device. S22. Using fixed targets or control points, the acquired three-dimensional point cloud data of the aqueduct is transformed and registered to the unified coordinate system of the three-dimensional digital model through coordinate transformation. S23. Calculate the positional deviation between points in the point cloud and corresponding points in the 3D digital model; S24. Correct the geometric parameters of the three-dimensional digital model based on the positional deviation.

7. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 6, characterized in that, In S22, the ICP algorithm is used to find the rotation matrix and translation vector so that after the three-dimensional point cloud data of the aqueduct is transformed to the unified coordinate system of the three-dimensional digital model by the rotation matrix and translation vector, the overall positional deviation between the points in the three-dimensional point cloud data and the corresponding points in the three-dimensional digital model is minimized.

8. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 1, characterized in that, Correcting the parameters of thermodynamic simulation calculations based on temperature deviations includes the following steps: S31. When the temperature deviation exceeds the preset threshold, adjust the thermal conductivity or specific heat capacity of the concrete, re-perform the thermodynamic simulation calculation, obtain the new concrete temperature field simulation result, and then compare it with the collected temperature of the aqueduct to obtain the new temperature deviation. If the new temperature deviation is greater than the temperature deviation, adjust the thermal conductivity or specific heat capacity of the concrete in the opposite direction. S32. Repeat S31 until the temperature deviation does not exceed the preset threshold or the maximum number of adjustments is reached to obtain the latest simulation results.

9. The intelligent temperature control system for a large aqueduct based on digital twin technology according to claim 1, characterized in that, The water flow rate in the corresponding water pipe is controlled to cool the water based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation. This includes: analyzing the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation to obtain the local temperature and local temperature difference; if the local temperature exceeds the first threshold, the water flow rate in the corresponding water pipe is increased; if the local temperature difference is greater than the second threshold, the water flow rate in the corresponding water pipe with the larger local temperature difference is increased.

10. A method for intelligent temperature control of a large aqueduct based on digital twin technology, applied to the intelligent temperature control system for a large aqueduct based on digital twin technology as described in claim 1, characterized in that, The method includes the following steps: S101. Construct a three-dimensional digital model of the aqueduct based on the construction drawings, and correct the geometric parameters of the three-dimensional digital model using the three-dimensional point cloud data of the aqueduct. S102. Perform thermodynamic simulation calculations based on a three-dimensional digital model to obtain simulation results, including simulation results of the temperature field of the aqueduct concrete. S103. Collect the temperature of the aqueduct, compare the simulation results of the temperature field of the aqueduct concrete with the collected temperature of the aqueduct, obtain the temperature deviation, correct the parameters of the thermodynamic simulation calculation based on the temperature deviation, and obtain the simulation results after correcting the parameters of the thermodynamic simulation calculation. S104. Based on the simulation results after correcting the geometric parameters of the three-dimensional digital model and the parameters of the thermodynamic simulation calculation, control the water flow rate of the corresponding water pipe to cool down.

Images