CN-121744954-B - Digital twinning technology-based large-scale aqueduct intelligent temperature control system and method

CN121744954BCN 121744954 BCN121744954 BCN 121744954BCN-121744954-B

Abstract

The invention provides a large-scale aqueduct intelligent temperature control system and method based on a digital twin technology, which relate to the technical field of aqueduct temperature monitoring and realize local targeted cooling by constructing a three-dimensional digital model of an aqueduct and correcting geometric parameters of the three-dimensional digital model through three-dimensional point cloud data of the aqueduct, carrying out thermodynamic simulation calculation based on the three-dimensional digital model to obtain a simulation result, comparing the simulation result of an aqueduct concrete temperature field with the acquired temperature of the aqueduct to obtain a temperature deviation, correcting the thermodynamic simulation calculation parameters according to the temperature deviation, thereby updating the simulation result, improving the accuracy of the simulation result, controlling water flow of a corresponding water pipe according to the corrected geometric parameters of the three-dimensional digital model and the simulation result after correcting the thermodynamic simulation calculation parameters.

Inventors

LIU PINGAN
WANG TIEHU
OU YUPENG
CHEN YOULIN
LI MEIRONG
YU XIAOFEI
WANG YONG
CHEN SHUANG
YUE FEI

Assignees

中国电建集团成都勘测设计研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260228

Claims (10)

1. The utility model provides a large-scale aqueduct intelligence temperature control system based on digital twin technique, its characterized in that, the aqueduct is provided with prestressing force pipeline, be provided with the water pipe in the prestressing force pipeline, the system includes: the information acquisition subsystem is used for acquiring the temperature of the aqueduct; the digital twin model construction subsystem is used for constructing a three-dimensional digital model of the aqueduct based on the construction drawing; the simulation calculation subsystem is used for carrying out thermodynamic simulation calculation based on the three-dimensional digital model to obtain simulation results, wherein the simulation results comprise simulation results of the aqueduct concrete temperature field; The feedback correction subsystem is used for correcting the geometric parameters of the three-dimensional digital model through three-dimensional point cloud data of the aqueduct, comparing the simulation result of the concrete temperature field of the aqueduct with the acquired temperature of the aqueduct, obtaining temperature deviation, and correcting the thermodynamic simulation calculation parameters according to the temperature deviation; And the temperature regulation subsystem controls the water flow of the corresponding water pipe to cool according to the simulation result after correcting the geometric parameters of the three-dimensional digital model and correcting the thermodynamic simulation calculation parameters.
2. The intelligent temperature control system of the large aqueduct based on the digital twin technology as claimed in claim 1, wherein the information acquisition subsystem further comprises an information processing subsystem, and the information processing subsystem is used for denoising and supplementing missing values of data acquired by the information acquisition subsystem; the denoising method comprises the following steps: s01, normalizing the historical acquisition data, wherein the normalization formula is as follows: , wherein, Is the data collected in the history of the process, Is the minimum value in the historical acquisition data, Is the maximum value in the historical acquisition data, Representing normalized historical acquisition data; S02, constructing a supervised learning data set, wherein the supervised learning data set comprises an input sequence and target output, the input sequence is noise-containing data with preset length in the historical acquisition data, and the target output is denoising data at the next moment; S03, constructing a first long-term memory network model, wherein the first long-term memory network model comprises a forgetting door, an input door, a state updating and outputting door and a full-connection layer; s04, adopting a mean square error as a loss function, using an Adam self-adaptive learning rate optimizer to minimize the loss function so as to monitor an input sequence and target output in a learning data set, and iteratively updating parameters of a first long-short-term memory network through a back propagation algorithm and the optimizer until the loss function converges, so as to obtain a trained long-short-term memory network model for denoising; s05, denoising the data acquired by the information acquisition subsystem by using the trained long-period memory network model to obtain denoised normalized data, and performing inverse normalization to obtain denoised data; the deficiency value complement includes the following steps: S001, marking a missing value in original time sequence data, normalizing, and dividing the normalized time sequence data into a plurality of subsequences by using a sliding window method; s002, constructing a second long-term and short-term memory network model; S003, adopting a mean square error as a loss function, updating the weight and bias of the second long-short-term memory network model through a back propagation algorithm, and minimizing the loss function to obtain the long-short-term memory network model for deficiency value complementation; S004, for each missing value, using the data of the previous and subsequent time steps as input, and predicting the missing value through the long-short-period memory network model for the completion of the missing value until all the missing values are filled, so as to obtain the completed normalized time sequence data; s005, carrying out inverse normalization on the complemented normalized time sequence data to obtain complemented original time sequence data.
3. The intelligent temperature control system of the large aqueduct based on the digital twin technology as claimed in claim 1, wherein the thermodynamic simulation calculation comprises the following steps: s11, constructing a finite element model in Midas FEA/NX based on the three-dimensional digital model; S12, applying boundary conditions, wherein the boundary conditions comprise constraint boundaries and convection boundaries; s13, setting a heat source function by taking the concrete molding temperature as an initial temperature field, wherein the heat source function is as follows: , wherein, The time is represented by the time period of the day, The concrete adiabatic temperature rise at the moment t is shown, The maximum adiabatic temperature rise value of the concrete is shown, Represents a natural constant of the natural product, The coefficient of heat release is indicated, Indicating the start-up time of the heat source, , Represents the amount of gel material per cubic meter, Represents the quantity of heat of hydration per kilogram of cement, Represents the hydration heat adjustment coefficient of the admixture, Represents the specific heat capacity of the concrete, Representing the mass density of the concrete; s14, defining a cooling water pipe heat exchange calculation formula as follows: , wherein, The temperature is indicated as a function of the temperature, The time is represented by the time period of the day, Represents a standard thermal conduction term and, The temperature conductivity coefficient of the concrete is shown, Representing the coordinates of a three-dimensional space, Represents the equivalent negative heat source item of water pipe cooling, The initial temperature of the concrete is indicated, The water temperature is indicated and the water temperature, , The time is represented by the time period of the day, , , Indicating the heat release coefficient of the contact surface of the concrete and the water pipe, Indicating the distance between the water pipes, Represents the outer diameter of the water pipe, Indicating the inner diameter of the water pipe, The thermal conductivity coefficient of the concrete is represented, Indicating the coefficient of thermal conductivity of the water pipe, Represents the hydration heat of the concrete, , The heat release coefficient of the concrete is expressed, Representing the coefficient, referring to the influence intensity of hydration heat on the temperature field in unit time, Indicating temperature For time of day Is the first partial derivative of (a); s15, defining a concrete surface heat dissipation formula as follows: , wherein, Indicating the amount of heat dissipated from the surface of the concrete, The heat release coefficient of the concrete surface is shown, Indicating the surface area of the concrete, The surface temperature of the concrete is indicated, Indicating the temperature of the environment and, , Representing wind speed; s16, setting calculation time step and outputting simulation results of the aqueduct concrete temperature field.
4. The intelligent temperature control system of a large-scale aqueduct based on digital twin technology as defined in claim 3, wherein in S15, if the concrete surface has an insulating layer, the convection coefficient is adjusted, and the adjusted convection coefficient is , , wherein, Represent the first The thickness of the heat-insulating layer is equal to that of the heat-insulating layer, Represent the first Thermal conductivity of the thermal insulation layer.
5. The digital twinning technology-based large-scale aqueduct intelligent temperature control system according to claim 4, wherein the simulation result further comprises a aqueduct concrete temperature stress field simulation result, and the thermodynamic simulation calculation further comprises: s17, defining the elastic modulus and the thermal expansion coefficient which are dependent on temperature, carrying out temperature stress coupling analysis to obtain a simulation result of a temperature stress field of the aqueduct concrete, wherein the temperature stress is stress caused by temperature change and internal and external temperature difference of the concrete, and analyzing the simulation result of the temperature stress field to obtain a potential crack position.
6. The intelligent temperature control system of a large-scale aqueduct based on the digital twin technology as defined in claim 1, wherein the geometric parameters of the three-dimensional digital model are modified by three-dimensional point cloud data of the aqueduct, comprising the following steps: S21, arranging fixed targets or control points on the aqueduct, and acquiring three-dimensional point cloud data of the aqueduct through three-dimensional laser scanning equipment; S22, converting and registering the acquired three-dimensional point cloud data of the aqueduct into a unified coordinate system where the three-dimensional digital model is located by using the distributed fixed targets or control points through coordinate transformation; s23, calculating the position deviation between the points in the point cloud and the corresponding points in the three-dimensional digital model; s24, correcting the geometric parameters of the three-dimensional digital model according to the position deviation.
7. The large-scale aqueduct intelligent temperature control system based on the digital twin technology according to claim 6, wherein in S22, a rotation matrix and a translation vector are searched by adopting an ICP algorithm, so that after three-dimensional point cloud data of the aqueduct is transformed to a unified coordinate system where a three-dimensional digital model is located by the rotation matrix and the translation vector, the integral position deviation between points in the three-dimensional point cloud data and corresponding points in the three-dimensional digital model is minimum.
8. The intelligent temperature control system of a large-scale aqueduct based on the digital twin technology as defined in claim 1, wherein the parameters calculated by thermodynamic simulation are corrected according to the temperature deviation, comprising the steps of: S31, when the temperature deviation exceeds a preset threshold, adjusting the heat conductivity coefficient or the specific heat capacity of the concrete, carrying out thermodynamic simulation calculation again to obtain a new simulation result of the concrete temperature field, comparing the simulation result with the temperature of the acquired aqueduct to obtain a new temperature deviation, and if the new temperature deviation is larger than the temperature deviation, reversely adjusting the heat conductivity coefficient or the specific heat capacity of the concrete; S32, repeating the step S31 until the temperature deviation does not exceed a preset threshold value or the maximum adjustment times are reached, and obtaining the latest simulation result.
9. The intelligent temperature control system of the large aqueduct based on the digital twin technology according to claim 1 is characterized in that the water flow of the corresponding water pipe is controlled to be cooled according to simulation results after correcting the geometric parameters of the three-dimensional digital model and correcting the parameters of thermodynamic simulation calculation, the simulation results after correcting the geometric parameters of the three-dimensional digital model and correcting the parameters of thermodynamic simulation calculation are analyzed to obtain local temperature and local temperature difference, if the local temperature exceeds a first threshold value, the water flow in the water pipe which is locally corresponding is increased, and if the local temperature difference is greater than a second threshold value, the water flow in the water pipe which is locally corresponding and is large in temperature in the local temperature difference is increased.
10. The large-scale aqueduct intelligent temperature control method based on the digital twin technology is applied to the large-scale aqueduct intelligent temperature control system based on the digital twin technology as claimed in claim 1, and is characterized in that the method comprises the following steps: S101, constructing a three-dimensional digital model of the aqueduct based on a construction drawing, and correcting geometric parameters of the three-dimensional digital model through three-dimensional point cloud data of the aqueduct; S102, carrying out thermodynamic simulation calculation based on a three-dimensional digital model to obtain a simulation result, wherein the simulation result comprises a simulation result of an aqueduct concrete temperature field; S103, acquiring the temperature of the aqueduct, comparing the simulation result of the aqueduct concrete temperature field with the acquired temperature of the aqueduct to obtain temperature deviation, and correcting the thermodynamic simulation calculation parameters according to the temperature deviation to obtain a simulation result after correcting the thermodynamic simulation calculation parameters; S104, controlling the water flow of the corresponding water pipe to cool according to the simulation result after correcting the geometric parameters of the three-dimensional digital model and correcting the thermodynamic simulation calculation parameters.

Description

Digital twinning technology-based large-scale aqueduct intelligent temperature control system and method Technical Field The invention relates to the technical field of aqueduct temperature monitoring, in particular to an intelligent temperature control system and method for a large-scale aqueduct based on a digital twin technology. Background The large aqueduct refers to an aqueduct which can cause harmful cracks to be generated due to temperature change and shrinkage caused by hydration heat under the condition of not controlling temperature. The aqueduct is difficult to control hydration heat during construction and maintenance, certain heat is released in the cement hydration heat process, the structural section of the large-volume concrete is thicker, the surface coefficient is smaller, and the heat generated by cement is accumulated in the structure and is not easy to dissipate. The hydration heat in the concrete cannot be timely emitted, the larger the inner and outer temperature difference, the more cracks are generated, and the influence on the impermeability, durability and safety of the aqueduct structure is caused. The traditional cooling measures mainly adopt the mode of burying the cooling pipe in advance in the structure, and then introducing cold water into the cooling pipe to reduce the temperature in the mass concrete. Although the mode can reduce the temperature inside the mass concrete, certain defects exist, the strength of the cooling pipe material is relatively low, the outer surface is smooth, a pore canal is left on the concrete structure after the concrete is coagulated and hardened, so that the stress concentration and other problems occur in the later stress process of the structure, the safety and the reliability of the structure are reduced, meanwhile, the automatic control of cooling facilities cannot be realized, the problem of overhigh local hydration heat of the concrete body is solved, no specific cooling measures are formed, the cooling effect is poor, and certain waste is caused. Disclosure of Invention The invention provides a large-scale aqueduct intelligent temperature control system and method based on a digital twin technology, which solve the problem that the prior art cannot aim at the targeted cooling of local overhigh hydration heat. The technical scheme adopted for solving the technical problems is that the large-scale aqueduct intelligent temperature control system based on the digital twin technology is provided with a prestressed pipeline, a water pipe is arranged in the prestressed pipeline, and the system comprises: the information acquisition subsystem is used for acquiring the temperature of the aqueduct; the digital twin model construction subsystem is used for constructing a three-dimensional digital model of the aqueduct based on the construction drawing; the simulation calculation subsystem is used for carrying out thermodynamic simulation calculation based on the three-dimensional digital model to obtain simulation results, wherein the simulation results comprise simulation results of the aqueduct concrete temperature field; The feedback correction subsystem is used for correcting the geometric parameters of the three-dimensional digital model through three-dimensional point cloud data of the aqueduct, comparing the simulation result of the concrete temperature field of the aqueduct with the acquired temperature of the aqueduct, obtaining temperature deviation, and correcting the thermodynamic simulation calculation parameters according to the temperature deviation; And the temperature regulation subsystem controls the water flow of the corresponding water pipe to cool according to the simulation result after correcting the geometric parameters of the three-dimensional digital model and correcting the thermodynamic simulation calculation parameters. Further, the information acquisition subsystem further comprises an information processing subsystem, wherein the information processing subsystem is used for denoising and supplementing missing values of the data acquired by the information acquisition subsystem; the denoising method comprises the following steps: s01, normalizing the historical acquisition data, wherein the normalization formula is as follows: , wherein, Is the data collected in the history of the process,Is the minimum value in the historical acquisition data,Is the maximum value in the historical acquisition data,Representing normalized historical acquisition data; S02, constructing a supervised learning data set, wherein the supervised learning data set comprises an input sequence and target output, the input sequence is noise-containing data with preset length in the historical acquisition data, and the target output is denoising data at the next moment; S03, constructing a first long-term memory network model, wherein the first long-term memory network model comprises a forgetting door, an input door, a state updating and outputt