Concrete pouring and curing control method and system with auxiliary temperature field monitoring

Abstract

The present invention discloses a concrete pouring and curing control method and system with auxiliary temperature field monitoring, which relates to the field of temperature measurement technology. The method includes: obtaining concrete pouring scene characteristics; constructing a concrete temperature prediction model based on a graph neural network; interacting with a user end to receive a number of expected temperature time series information; receiving environmental monitoring information; predicting the initial temperature of the concrete, the cooling water temperature, the cooling water flow rate, and a number of monitored temperature time series information, and generating a number of concrete predicted temperature time series information; if the number of expected temperature time series information is not met, a temperature warning signal is generated and sent to the user end. The technical problem of the difficulty in real-time monitoring of concrete temperature changes and accurate prediction and curing control in the existing concrete pouring and curing control is solved, and the comprehensive perception and intelligent prediction of the concrete temperature field are realized, achieving the technical effect of improving the accuracy, real-time nature of concrete temperature prediction and intelligently controlling concrete pouring and curing.

Description

Technical Field

[0001] The present application relates to the technical field related to temperature measurement, and specifically to a concrete pouring and curing control method and system for auxiliary temperature field monitoring. Background Art

[0002] In modern construction projects, the concrete pouring and curing process has a vital impact on the safety and durability of the structure, especially in complex environments such as extreme climates, high altitudes or frozen soil areas. Concrete temperature control becomes an extremely challenging task. The temperature change of concrete not only affects its internal stress distribution and deformation behavior, but is also directly related to its strength development and durability. Traditional concrete temperature monitoring mainly uses embedded thermometers for temperature monitoring, ignoring the comprehensive influence of various factors such as cement type, aggregate type, water-cement ratio, etc. on concrete temperature changes. The internal temperature control is performed through embedded pipes, but the temperature monitoring accuracy and real-time performance are insufficient, and the temperature control operation is complicated, making it impossible to intelligently perform concrete pouring and curing control.

[0003] Therefore, in the current technologies related to concrete pouring and curing control, there is a technical problem that it is difficult to monitor the temperature changes of concrete in real time and to make accurate predictions and curing control. Summary of the invention

[0004] The present application solves the technical problem of difficulty in real-time monitoring of concrete temperature changes and accurate prediction and curing control in existing concrete pouring and curing control by providing a concrete pouring and curing control method and system with auxiliary temperature field monitoring, realizes comprehensive perception and intelligent prediction of the concrete temperature field, and achieves the technical effect of improving the accuracy, real-time nature of concrete temperature prediction and intelligently controlling concrete pouring and curing.

[0005] The present application provides a concrete pouring and curing control method for auxiliary temperature field monitoring, the method comprising: obtaining concrete pouring scene characteristics, wherein the concrete pouring scene characteristics include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipeline deployment topology; through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipeline deployment topology, a concrete temperature prediction model is constructed based on a graph neural network, wherein the concrete temperature prediction model has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes are The points correspond one-to-one with several pre-buried temperature measuring points of the concrete pouring body equivalent model; the interactive user end receives several expected temperature time series information, wherein the several expected temperature time series information corresponds one-to-one with the several topological nodes; receives environmental monitoring information, wherein the environmental monitoring information includes ambient temperature, solar radiation, wind speed, and ambient humidity; according to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, the initial temperature of the concrete, the cooling water temperature, the cooling water flow, and several monitored temperature time series information are predicted to generate several concrete predicted temperature time series information; when the several concrete predicted temperature time series information does not meet the several expected temperature time series information, a temperature warning signal is generated and sent to the user end.

[0006] In a possible implementation, according to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, the initial temperature of the concrete, the cooling water temperature, the cooling water flow, and several monitored temperature time series information are predicted to generate several concrete predicted temperature time series information, and the following processing is also performed before: when the several monitored temperature time series information do not meet the several expected temperature time series information, a temperature alarm signal is generated and sent to the user end.

[0007] In a possible implementation, a concrete temperature prediction model is constructed based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipeline deployment topology, and the following processing is also performed: a concrete temperature prediction network topology is constructed based on a graph neural network through the concrete pouring body equivalent model, wherein the concrete temperature prediction network topology has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measuring points of the concrete pouring body equivalent model; with the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipeline deployment topology as constraints, the temperature field historical monitoring data is retrieved, wherein any piece of temperature field historical monitoring data includes ambient temperature record information, solar radiation record information, wind speed record information, ambient humidity record information, concrete initial temperature record information, cooling water temperature record information information, cooling water flow recording information and temperature field recording timing information; divide the temperature field recording timing information into two segments for multiple times to construct multiple temperature field front-end recording timing information and multiple temperature field rear-end recording timing information; extract the temperature of the multiple temperature field front-end recording timing information and the multiple temperature field rear-end recording timing information according to the multiple pre-buried temperature measuring points of the concrete pouring body equivalent model to generate multiple temperature field front-end recording timing information and multiple temperature field rear-end recording timing information; store the ambient temperature recording information, the solar radiation recording information , the wind speed record information, the ambient humidity record information, the concrete initial temperature record information, the cooling water temperature record information, and the cooling water flow record information are respectively combined with the plurality of temperature field front-end record time series information to generate a plurality of sets of training input data; the plurality of temperature field rear-end record time series information are regarded as supervision data, the plurality of training input data are regarded as input data, the concrete temperature prediction network topology is trained to generate the concrete temperature prediction model, wherein the input data of the corresponding temperature measurement point during the training process is input into the corresponding topological node.

[0008] In a possible implementation, the plurality of temperature field back-end record time series information is regarded as supervision data, the plurality of sets of training input data is regarded as input data, the concrete temperature prediction network topology is trained, the concrete temperature prediction model is generated, and the following processing is further performed: constructing a subnode training loss function:

[0009]

[0010] Among them, LOSSi Characterizes the loss value of the i-th topological node, T i,j Represents the predicted temperature value of the i-th topological node at the j-th moment, T i,j,0 represents the supervised temperature value of the i-th topological node at the j-th moment, N represents the total number of output moments, and a represents a small constant;

[0011] Construct a global training loss function:

[0012]

[0013] Among them, M represents the total number of topological nodes, Q represents the training cycle of fitting the global loss, and k represents the kth training of any cycle; according to the sub-node training loss function and the global training loss function, the multiple temperature field back-end record time series information is regarded as supervision data, and the multiple groups of training input data are regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated.

[0014] In a possible implementation, according to the sub-node training loss function and the global training loss function, the multiple temperature field back-end record time series information is regarded as supervision data, the multiple groups of training input data are regarded as input data, the concrete temperature prediction network topology is trained to generate the concrete temperature prediction model, and the following processing is performed: configuring the sub-node loss threshold and the global loss threshold; when in several consecutive training cycles, the sub-node training loss function is less than or equal to the sub-node loss threshold, and the proportion of cycles in which the global training loss function is less than or equal to the global loss threshold is greater than or equal to the convergence ratio, the concrete temperature prediction model is output.

[0015] In a possible implementation, when the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information, a temperature warning signal is generated and sent to the user end, and the following processing is also performed: deviation analysis is performed on the plurality of concrete predicted temperature time series information and the plurality of expected temperature time series information to obtain a proportion of the number of deviation temperature measuring points; a distribution coordinate set of deviation temperature measuring points is obtained; a distribution distance mean analysis is performed on the distribution coordinate set of deviation temperature measuring points to obtain a distribution distance coefficient; when the proportion of the number of deviation temperature measuring points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is less than the distribution distance threshold, a first-level temperature warning signal is generated and set as the temperature warning signal; when the proportion of the number of deviation temperature measuring points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is greater than or equal to the distribution distance threshold, a second-level temperature warning signal is generated and set as the temperature warning signal; wherein the warning level of the second-level temperature warning signal is higher than that of the first-level temperature warning signal.

[0016] In a possible implementation, when the plurality of concrete predicted temperature time series information does not meet the plurality of expected temperature time series information, a temperature warning signal is generated and sent to the user end, and the following processing is also performed: when the proportion of the number of deviation temperature measurement points is less than the warning proportion threshold, the temperature warning signal is not generated.

[0017] The present application also provides a concrete pouring and curing control system for auxiliary temperature field monitoring, including: a concrete pouring scene feature acquisition module, the concrete pouring scene feature acquisition module is used to obtain concrete pouring scene features, wherein the concrete pouring scene features include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipeline deployment topology; a concrete temperature prediction model construction module, the concrete temperature prediction model construction module is used to construct a concrete temperature prediction model based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipeline deployment topology, wherein the concrete temperature prediction model has a plurality of topological nodes, the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model. ; an expected temperature time series information receiving module, the expected temperature time series information receiving module is used for interacting with the user end, receiving a plurality of expected temperature time series information, wherein the plurality of expected temperature time series information corresponds one to one with the plurality of topological nodes; an environmental monitoring information receiving module, the environmental monitoring information receiving module is used for receiving environmental monitoring information, wherein the environmental monitoring information includes environmental temperature, solar radiation, wind speed, and environmental humidity; a concrete predicted temperature time series information generating module, the concrete predicted temperature time series information generating module is used for predicting the concrete initial temperature, cooling water temperature, cooling water flow, and a plurality of monitoring temperature time series information according to the concrete temperature prediction model, based on the environmental temperature, the solar radiation, the wind speed, and the environmental humidity, and generating a plurality of concrete predicted temperature time series information; a temperature warning signal generating module, the temperature warning signal generating module is used for generating a temperature warning signal and sending it to the user end when the plurality of concrete predicted temperature time series information does not meet the plurality of expected temperature time series information.

[0018] It is intended to obtain the characteristics of the concrete pouring scene through the concrete pouring and curing control method and system with auxiliary temperature field monitoring proposed in this application; build a concrete temperature prediction model based on graph neural network; interact with the user end to receive several expected temperature time series information; receive environmental monitoring information; predict the initial temperature of the concrete, cooling water temperature, cooling water flow, and several monitored temperature time series information, and generate several concrete predicted temperature time series information; if several expected temperature time series information are not met, generate a temperature warning signal and send it to the user end. It solves the technical problem of the existing concrete pouring and curing control that it is difficult to monitor the temperature change of concrete in real time and make accurate predictions and curing control, realizes the comprehensive perception and intelligent prediction of the concrete temperature field, and achieves the technical effect of improving the accuracy and real-time nature of concrete temperature prediction and intelligently controlling concrete pouring and curing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0020] Figure 1 A schematic diagram of the process flow of the concrete pouring and curing control method with auxiliary temperature field monitoring provided in an embodiment of the present application;

[0021] Figure 2 A schematic diagram of the structure of a concrete pouring and curing control system with auxiliary temperature field monitoring provided in an embodiment of the present application.

[0022] Explanation of the accompanying drawings: concrete pouring scene feature acquisition module 10, concrete temperature prediction model construction module 20, expected temperature time series information receiving module 30, environmental monitoring information receiving module 40, concrete predicted temperature time series information generation module 50, temperature warning signal generation module 60. DETAILED DESCRIPTION

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0024] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0025] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0026] The present application embodiment provides a concrete pouring and curing control method with auxiliary temperature field monitoring, such as Figure 1 As shown, the method includes:

[0027] Step S100, obtaining the concrete pouring scene characteristics, wherein the concrete pouring scene characteristics include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipe deployment topology. The concrete pouring scene characteristics specifically refer to the collection of various conditions and factors that affect the quality, performance and construction efficiency of the concrete pouring body during the concrete pouring process, including cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipe deployment topology. Specifically, cement is the main cementitious material in concrete. According to its use and performance, it can be divided into various types such as general-purpose, special-purpose and special cement, which directly affects the strength, durability and working performance of concrete. Aggregate is in concrete. Granular materials are usually divided into coarse aggregate (such as crushed stone, pebbles) and fine aggregate (such as sand). Their types and properties have a significant impact on the strength, workability, durability, etc. of concrete; cement content refers to the amount of cement in concrete, usually expressed as the weight of cement per cubic meter of concrete. The right amount of cement can improve the strength and durability of concrete, but too much or too little will affect the performance of concrete; aggregate particle size refers to the size of aggregate particles. For example, the particle size of coarse aggregate in cast-in-place pile concrete should be between 15mm and 40mm to ensure the strength and stability of the pile body. Qualitative; water-cement ratio refers to the ratio of water consumption per cubic meter of concrete to the consumption of all cementitious materials (including cement, fly ash, slag powder, etc.). For example, the water-cement ratio of waterproof concrete shall not be greater than 0.50, and should not be greater than 0.45 when there is an aggressive medium; fly ash content refers to the proportion of fly ash added to concrete. Fly ash, as an admixture, can improve the performance of concrete, such as improving the fluidity of concrete and reducing the hydration heat, and is generally not greater than 25%; slag powder content refers to the proportion of slag powder added to concrete. Slag powder can improve the performance of concrete, such as improving The equivalent model of concrete pouring body refers to simplifying the concrete pouring body into a geometric model that is easy to analyze and calculate. By establishing an equivalent model, the mechanical properties, temperature field, etc. of the concrete pouring body can be simulated and analyzed; the cooling pipe deployment topology refers to the layout and connection relationship of the cooling pipes set up to control the internal temperature of the concrete during the large-volume concrete pouring process. Reasonable cooling pipe deployment topology can effectively reduce the internal temperature of the concrete and prevent quality problems such as cracks in the concrete due to excessive temperature.

[0028] Step S200, constructing a concrete temperature prediction model based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology, wherein the concrete temperature prediction model has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measuring points of the concrete pouring body equivalent model. Graph neural network is a neural network that specializes in processing graph structure data and can capture complex relationships and dependencies between nodes. In concrete temperature prediction, the equivalent model of concrete pouring body can be regarded as a graph, where nodes represent pre-embedded temperature measuring points and edges represent heat conduction relationships between nodes. Long short-term memory neural network is a special recurrent neural network that can effectively process long-term dependencies in time series data. In concrete temperature prediction, it is used to capture the changing trend of temperature data over time. Specifically, in the equivalent model of concrete pouring body, several topological nodes are set according to the positions of pre-embedded temperature measuring points. Each topological node corresponds to a long short-term memory neural network unit for prediction. The temperature at the measuring point is measured, and the characteristics such as cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, etc. are used as the input of the long short-term memory neural network unit, which are converted into a format acceptable to the neural network through the embedding layer. A graph structure is constructed according to the geometric and physical properties of the equivalent model of the concrete pouring body. The edge relationship between the nodes is weighted according to heat conduction to reflect the intensity of heat exchange between different nodes. The model is trained using historical temperature data and feature data. The parameters of the long short-term memory network unit and the weights of the graph neural network are optimized through the back propagation algorithm. Finally, a concrete temperature prediction model is constructed, which can accurately predict the temperature change of concrete.

[0029] In a possible implementation, step S200 further includes step S210, wherein a concrete temperature prediction network topology is constructed based on a graph neural network through the concrete pouring body equivalent model, wherein the concrete temperature prediction network topology has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-embedded temperature measuring points of the concrete pouring body equivalent model. Based on the concrete pouring body equivalent model, a graph neural network is used to construct a concrete temperature prediction network topology, because the graph neural network can process graph structure data well, wherein nodes and edges can represent pre-embedded temperature measuring points in the concrete pouring body and their mutual relationships, and specifically, a plurality of topological nodes in the network topology are all long short-term memory neural networks, and these nodes correspond one-to-one to a plurality of pre-embedded temperature measuring points of the concrete pouring body equivalent model.

[0030] Step S200 also includes step S220, which retrieves temperature field historical monitoring data based on the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipeline deployment topology as constraints, wherein any temperature field historical monitoring data includes environmental temperature record information, solar radiation record information, wind speed record information, environmental humidity record information, concrete initial temperature record information, cooling water temperature record information, cooling water flow record information, and temperature field record timing information. Based on the constraints of cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, cooling pipeline deployment topology, etc., the temperature field historical monitoring data is retrieved, and each temperature field historical monitoring data includes environmental temperature record information, solar radiation record information, wind speed record information, environmental humidity record information, concrete initial temperature record information, cooling water temperature record information, cooling water flow record information, and temperature field record timing information.

[0031] Step S200 also includes step S230, performing multiple binary segmentation on the temperature field recording time series information to construct multiple temperature field front segment recording time series information and multiple temperature field rear segment recording time series information. Performing multiple binary segmentation on the temperature field recording time series information, that is, dividing the time series data into multiple front segments and rear segments, increases the amount of training data for model training. It also includes step S240, performing temperature measurement point temperature extraction on the multiple temperature field front segment recording time series information and the multiple temperature field rear segment recording time series information according to the multiple pre-buried temperature measurement points of the concrete pouring body equivalent model, and generating multiple temperature field front segment recording time series information and multiple temperature field rear segment recording time series information. Performing temperature measurement point temperature extraction on the multiple temperature field front segment recording time series information and the multiple temperature field rear segment recording time series information according to the pre-buried temperature measurement points of the concrete pouring body equivalent model, and generating multiple temperature field front segment recording time series information and rear segment recording time series information.

[0032] Step S200 also includes step S250, combining the ambient temperature record information, the solar radiation record information, the wind speed record information, the ambient humidity record information, the concrete initial temperature record information, the cooling water temperature record information, and the cooling water flow record information with the plurality of temperature field front-end record time series information to generate multiple sets of training input data. It also includes step S260, treating the plurality of temperature field rear-end record time series information as supervision data, treating the plurality of training input data as input data, training the concrete temperature prediction network topology, and generating the concrete temperature prediction model, wherein the input data of the corresponding temperature measurement point during the training process is input into the corresponding topological node. The recorded information of ambient temperature, solar radiation, wind speed, ambient humidity, initial temperature of concrete, cooling water temperature and cooling water flow are respectively combined with the time series information of the front-end records of multiple temperature fields to generate multiple sets of training input data. Then the multiple sets of training input data are used as input in the training process, and the time series information of the back-end records of multiple temperature fields is used as supervision data to guide the learning of the network in the training process. During the training process, the input data of the corresponding temperature measuring point is input into the corresponding topological node, which ensures that the network can make independent temperature predictions for each temperature measuring point, and takes into account the mutual influence between different temperature measuring points, and finally generates a concrete temperature prediction model, which can predict the temperature changes of the concrete pouring body at different time points based on given constraints and input data (such as ambient temperature, solar radiation, etc.).

[0033] In a possible implementation, step S260 further includes step S261, constructing a sub-node training loss function:

[0034]

[0035] Among them, LOSS i Characterizes the loss value of the i-th topological node, T i,j Represents the predicted temperature value of the i-th topological node at the j-th moment, T i,j,0 represents the supervised temperature value of the i-th topological node at the j-th moment, N represents the total number of output moments, and a represents a small constant;

[0036] Construct a global training loss function:

[0037]

[0038] Among them, M represents the total number of topological nodes, Q represents the training cycle of fitting the global loss, and k represents the kth training in any cycle.

[0039] Step S260 also includes step S262, according to the sub-node training loss function and the global training loss function, the multiple temperature field back-end recorded time series information is regarded as supervision data, the multiple sets of training input data are regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated. The sub-node training loss function helps to adjust the network parameters more carefully during the training process to improve the overall performance. The global training loss function is the objective function to be minimized by the entire network during the training process. Specifically, the time series information recorded in the back of the temperature field is used as supervision data, and multiple sets of training input data are prepared. The structure of the concrete temperature prediction network is designed, including the number and configuration of input layers, hidden layers, and output layers. The prepared training input data and supervision data are used to train the network by minimizing the global training loss function. During the training process, the network parameters are adjusted according to the feedback of the sub-node loss function. After sufficient training iterations, the network will learn the relationship between the input data and the output data, and generate a concrete temperature prediction model for temperature prediction of new input data.

[0040] In a possible implementation, step S262 further includes step S2621, configuring a sub-node loss threshold and a global loss threshold. For each sub-node in the network, a loss threshold is set according to its characteristics and contribution to the overall prediction result, which is used to measure whether the loss of the sub-node during the training process has reached an acceptably low level; the global loss threshold is configured for the global loss function of the entire network, and is used to evaluate whether the overall performance of the entire network during the training process has reached the expected level.

[0041] Step S262 also includes step S2622, when the sub-node training loss function is less than or equal to the sub-node loss threshold in a number of consecutive training cycles, and the proportion of cycles in which the global training loss function is less than or equal to the global loss threshold is greater than or equal to the convergence ratio, the concrete temperature prediction model is output. At the end of each training cycle, the values ​​of the sub-node training loss function and the global training loss function are calculated, and then a convergence ratio is set, which represents the proportion of training cycles that meet the following two conditions in a number of consecutive training cycles, including that the sub-node training loss function is less than or equal to the sub-node loss threshold and the global training loss function is less than or equal to the global loss threshold. If this ratio is greater than or equal to the convergence ratio, it is considered that the model has converged, that is, the performance of the model has been stabilized during the training process, and the trained concrete temperature prediction model is output after the training is completed.

[0042] Step S300, the interactive user terminal receives a plurality of expected temperature timing information, wherein the plurality of expected temperature timing information corresponds one-to-one with the plurality of topological nodes. The expected temperature timing information refers to the temperature data sequence that the user expects to reach at multiple locations of the concrete within a certain time period in the future. The plurality of expected temperature timing information corresponds one-to-one with the plurality of topological nodes. Each expected temperature timing information contains the expected temperature values ​​at different time points. When a one-to-one correspondence is established between the expected temperature timing information received by the interactive user terminal and the topological node, it means that each topological node (i.e., each pre-buried temperature measuring point) has a clear and corresponding expected temperature timing information to ensure the accuracy of temperature prediction and control. For example, a unique identifier (such as an ID number) is assigned to each topological node in the data preprocessing stage, and when receiving the expected temperature timing information, the information is matched and associated with the corresponding identifier.

[0043] Step S400, receiving environmental monitoring information, wherein the environmental monitoring information includes environmental temperature, solar radiation, wind speed, and environmental humidity. Receiving environmental monitoring information refers to collecting various physical monitoring data related to the environment surrounding the concrete pouring body, which directly affects the heat dissipation conditions, water evaporation rate, and distribution of the internal temperature field of the concrete. Environmental monitoring information includes environmental temperature, solar radiation, wind speed, and environmental humidity. Specifically, the environmental temperature refers to the temperature of the air surrounding the concrete pouring body. A high temperature environment will accelerate the hydration reaction of the concrete, generate more heat, and increase the temperature of the concrete. A low temperature environment will slow down the hydration reaction rate, resulting in too low a temperature on the concrete surface, causing problems such as cold cracking. Solar radiation refers to the energy transmitted to the outside by the sun in the form of electromagnetic waves, including visible light, ultraviolet rays, and infrared rays. The intensity of solar radiation directly affects the temperature of the concrete surface. For example, For example, solar radiation on a clear day will cause the concrete surface to heat up rapidly, and then affect the temperature field inside the concrete through heat conduction; wind speed refers to the distance that air flows per unit time. High wind speed will accelerate the air flow on the concrete surface, enhance the convective heat dissipation effect, and thus reduce the temperature of the concrete. On the contrary, low wind speed will weaken the heat dissipation effect and cause the concrete temperature to rise; ambient humidity refers to the water vapor content in the air, which mainly affects the evaporation rate of moisture and shrinkage performance of concrete. In a high humidity environment, the evaporation rate of moisture on the concrete surface is slower, which is conducive to keeping the concrete moist. In a low humidity environment, it is easy to cause the concrete surface to dry quickly, resulting in shrinkage cracks and other problems, thereby affecting the distribution of the temperature field inside the concrete.

[0044] Step S500, predicting the initial temperature of concrete, cooling water temperature, cooling water flow, and several monitored temperature time series information according to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, and generating several concrete predicted temperature time series information. The concrete temperature prediction model predicts the temperature change of concrete in the future based on ambient temperature, solar radiation, wind speed, ambient humidity, initial concrete temperature, cooling water temperature, cooling water flow and several monitoring temperature time series information. Among them, several monitoring temperature time series information are the concrete temperature data sequences collected by each temperature measurement pre-buried point of concrete. Specifically, the concrete temperature prediction model simulates and calculates the physical processes such as heat conduction, convection heat dissipation and radiation heat dissipation inside the concrete according to historical data and current environmental conditions, and predicts the temperature of concrete at various time points in the future. During the prediction process, the model considers the influence of different factors on the concrete temperature, such as changes in ambient temperature, the intensity of solar radiation, the size of wind speed, changes in ambient humidity, etc. At the same time, according to the adjustment of cooling measures (such as cooling water temperature and flow), the influence of cooling effect on concrete temperature is predicted, and finally several concrete predicted temperature time series information are generated. This information represents the temperature change of concrete in the future in the form of time series, and each time series corresponds to a pre-buried temperature measurement point of concrete.

[0045] Step S600, when the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information, a temperature warning signal is generated and sent to the user end. The concrete predicted temperature time series information corresponding to each topological node (i.e., the embedded temperature measuring point) is compared point by point with the corresponding expected temperature time series information. During the comparison process, the temperature change trend in the future period is considered to determine whether the deviation between the predicted temperature and the expected temperature exceeds the preset temperature deviation threshold. If the deviation between the predicted temperature time series information and the expected time series information of one or some topological nodes exceeds the preset temperature deviation threshold, a corresponding temperature warning signal is generated, which may include a warning level (such as low, medium, high, etc.), a warning reason (such as the predicted temperature is too high, too low, etc.), a warning time range, and an affected area, etc., and then the signal is sent to the user end through a predetermined communication channel (such as SMS, email, APP push, etc.), and the user end may be a monitoring center at the construction site or a mobile terminal device of a project manager.

[0046] In a possible implementation, step S600 further includes step S610, performing deviation analysis on the plurality of concrete predicted temperature time series information and the plurality of expected temperature time series information to obtain the percentage of the number of deviation temperature measurement points. By comparing the temperature values ​​of the two sets of data (predicted temperature and expected temperature) at the same time point at the same temperature measurement point, finding out the points where there are significant differences (i.e., deviations), and calculating the proportion of these points to the total number of points, the percentage of the number of deviation temperature measurement points is obtained. For example, assuming that there is a temperature monitoring project for a concrete structure, the predicted temperature and the expected temperature are recorded at 5 different time points (the same temperature measurement point), and the data are shown in the following table:

[0047]

[0048] Assume that the deviation threshold is 1°C, that is, if the absolute value of the deviation between the predicted temperature and the expected temperature is greater than or equal to 1°C, then the point is considered to be a deviation temperature measurement point. In the above table, time point 3 is identified as a deviation temperature measurement point. Finally, the ratio of the number of deviation temperature measurement points (1) to the total number of temperature measurement points (5) is calculated, that is, 1 / 5=20%.

[0049] Step S600 also includes step S620, obtaining a deviation temperature measurement point distribution coordinate set. Record the coordinates of each deviation temperature measurement point in space to form a deviation temperature measurement point distribution coordinate set. It also includes step S630, performing a distribution distance mean analysis on the deviation temperature measurement point distribution coordinate set to obtain a distribution distance coefficient. The deviation temperature measurement point distribution coordinate set is further analyzed to calculate the average distance between these points or some form of distance measurement (such as Euclidean distance, Manhattan distance, etc.) to obtain the distribution distance mean, and this mean is used as the distribution distance coefficient to evaluate the dispersion degree of the deviation temperature measurement points in space.

[0050] Step S600 also includes step S640, when the proportion of the number of deviation temperature measurement points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is less than the distribution distance threshold, a first-level temperature warning signal is generated and set as the temperature warning signal. When the proportion of the number of deviation temperature measurement points is high, it means that there is a large error in the concrete temperature prediction model or an abnormality in the concrete temperature distribution. When the distribution distance coefficient is small, it means that the deviation temperature measurement points are relatively concentrated, indicating that the temperature of the local area of ​​the concrete is abnormal. When these two conditions are met at the same time, a first-level temperature warning signal is generated, indicating that there is a temperature abnormality in the local area, but it has not reached the level where emergency measures need to be taken immediately, and it is set as a temperature warning signal.

[0051] Step S600 also includes step S650, when the proportion of the number of deviation temperature measurement points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is greater than or equal to the distribution distance threshold, a secondary temperature warning signal is generated and set as the temperature warning signal. If the proportion of the number of deviation temperature measurement points is greater than or equal to the warning proportion threshold, it means that there are many abnormal points in the temperature distribution, which may be a sign of improper overall temperature control or the existence of local hot spots / cold spots. If the distribution distance coefficient is greater than or equal to the distribution distance threshold, it means that the deviation temperature measurement points are not only large in number, but also relatively widely distributed in space, and may cover a large area. When both conditions are met at the same time, it means that the current temperature distribution state has exceeded the normal range and may have a greater impact on the stability and safety of the entire structure or system. A secondary temperature warning signal is generated, and its warning level is higher than the primary temperature warning signal, that is, more serious than the primary temperature warning signal, indicating that it is necessary to pay closer attention to temperature changes and take more urgent and effective measures to deal with it, and it is set as a temperature warning signal.

[0052] In a possible implementation, step S600 further includes step S660, when the proportion of the number of deviation temperature measurement points is less than the warning proportion threshold, the temperature warning signal is not generated. In concrete temperature monitoring, by statistically analyzing the deviation between the actual temperature measurement points and the expected or ideal temperature state, if the proportion of the number of deviation temperature measurement points does not reach the preset warning level (i.e., less than the warning proportion threshold), it is judged that the current temperature distribution state is still within the normal or acceptable range, that is, the current temperature deviation is not enough to constitute a threat to the stability and safety of the concrete structure, no immediate action is needed, and no temperature warning signal is generated.

[0053] In a possible implementation, before step S600, it further includes generating a temperature alarm signal and sending it to the user end when the plurality of monitored temperature timing information does not meet the plurality of expected temperature timing information. Compare the plurality of monitored temperature timing information with the plurality of expected temperature timing information, detect the deviation between the actual monitored temperature and the expected temperature, and evaluate whether the temperature deviation exceeds the preset alarm threshold (used to define when an alarm signal needs to be issued). If the deviation between the actual monitored temperature timing information and the expected timing information exceeds the preset alarm threshold, immediately generate a temperature alarm signal such as alarm time, alarm type (such as too high temperature, too low temperature, sudden temperature change, etc.), alarm level, affected area, etc., and send it to the user end.

[0054] In the above, refer to Figure 1 The concrete pouring and curing control method with auxiliary temperature field monitoring according to an embodiment of the present invention is described in detail. Figure 2A concrete pouring and curing control system with auxiliary temperature field monitoring according to an embodiment of the present invention is described.

[0055] The concrete pouring and curing control system with auxiliary temperature field monitoring according to the embodiment of the present invention is used to solve the technical problem that it is difficult to monitor the temperature change of concrete in real time and perform accurate prediction and curing control in the existing concrete pouring and curing control, realize comprehensive perception and intelligent prediction of the concrete temperature field, and achieve the technical effect of improving the accuracy and real-time of concrete temperature prediction and intelligent control of concrete pouring and curing. The concrete pouring and curing control system with auxiliary temperature field monitoring includes: a concrete pouring scene feature acquisition module 10, a concrete temperature prediction model construction module 20, an expected temperature time series information receiving module 30, an environmental monitoring information receiving module 40, a concrete predicted temperature time series information generation module 50, and a temperature warning signal generation module 60.

[0056] A concrete pouring scene feature acquisition module 10, wherein the concrete pouring scene feature acquisition module 10 is used to obtain concrete pouring scene features, wherein the concrete pouring scene features include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipe deployment topology;

[0057] A concrete temperature prediction model construction module 20, the concrete temperature prediction model construction module 20 is used to construct a concrete temperature prediction model based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology, wherein the concrete temperature prediction model has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model;

[0058] An expected temperature timing information receiving module 30, the expected temperature timing information receiving module 30 is used for an interactive user terminal to receive a plurality of expected temperature timing information, wherein the plurality of expected temperature timing information corresponds to the plurality of topological nodes one by one;

[0059] An environmental monitoring information receiving module 40, the environmental monitoring information receiving module 40 is used to receive environmental monitoring information, wherein the environmental monitoring information includes environmental temperature, solar radiation, wind speed, and environmental humidity;

[0060] A concrete predicted temperature time series information generating module 50, the concrete predicted temperature time series information generating module 50 is used to predict the concrete initial temperature, cooling water temperature, cooling water flow, and a plurality of monitored temperature time series information according to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, and generate a plurality of concrete predicted temperature time series information;

[0061] The temperature warning signal generating module 60 is used to generate a temperature warning signal and send it to the user end when the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information.

[0062] The following will describe in detail the specific configuration of the concrete predicted temperature time series information generating module 50. The concrete predicted temperature time series information generating module 50 further includes: when the plurality of monitored temperature time series information do not meet the plurality of expected temperature time series information, generating a temperature alarm signal and sending it to the user end.

[0063] The specific configuration of the concrete temperature prediction model construction module 20 will be described in detail below. The concrete temperature prediction model construction module 20 may further include: constructing a concrete temperature prediction network topology based on a graph neural network through the concrete pouring body equivalent model, wherein the concrete temperature prediction network topology has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model; taking the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology as constraints, retrieving the temperature field historical monitoring data, wherein any piece of temperature field historical monitoring data includes environmental temperature record information, solar radiation record information, wind speed record information, environmental humidity record information, concrete initial temperature record information, cooling water temperature record information, cooling water flow record information, and temperature field record time series information; performing multiple binary segmentations on the temperature field record time series information to construct multiple The method comprises the steps of: extracting the temperature of the plurality of temperature field front-end record time series information and the plurality of temperature field rear-end record time series information according to the plurality of pre-buried temperature measuring points of the concrete pouring body equivalent model, and generating a plurality of temperature field front-end record time series information and a plurality of temperature field rear-end record time series information; combining the ambient temperature record information, the solar radiation record information, the wind speed record information, the ambient humidity record information, the concrete initial temperature record information, the cooling water temperature record information, and the cooling water flow record information with the plurality of temperature field front-end record time series information, and generating a plurality of training input data; treating the plurality of temperature field rear-end record time series information as supervision data, treating the plurality of training input data as input data, training the concrete temperature prediction network topology, and generating the concrete temperature prediction model, wherein the input data of the corresponding temperature measuring point during the training process is input into the corresponding topological node.

[0064] The specific configuration of the concrete temperature prediction model building module 20 will be described in detail below. The concrete temperature prediction model building module 20 may further include: constructing a subnode training loss function:

[0065]

[0066] Among them, LOSS i Characterizes the loss value of the i-th topological node, t i,j Represents the predicted temperature value of the i-th topological node at the j-th moment, T i,j,0represents the supervised temperature value of the i-th topological node at the j-th moment, N represents the total number of output moments, and a represents a small constant;

[0067] Construct a global training loss function:

[0068]

[0069] Among them, M represents the total number of topological nodes, Q represents the training cycle of fitting the global loss, and k represents the kth training of any cycle; according to the sub-node training loss function and the global training loss function, the multiple temperature field back-end record time series information is regarded as supervision data, and the multiple groups of training input data are regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated.

[0070] The specific configuration of the concrete temperature prediction model construction module 20 will be described in detail below. The concrete temperature prediction model construction module 20 further includes: configuring a sub-node loss threshold and a global loss threshold; when the sub-node training loss function is less than or equal to the sub-node loss threshold in a number of consecutive training cycles, and the proportion of cycles in which the global training loss function is less than or equal to the global loss threshold is greater than or equal to the convergence ratio, the concrete temperature prediction model is output.

[0071] The specific configuration of the temperature warning signal generation module 60 will be described in detail below. The temperature warning signal generation module 60 further includes: performing deviation analysis on the several concrete predicted temperature time series information and the several expected temperature time series information to obtain the ratio of the number of deviation temperature measurement points; obtaining the deviation temperature measurement point distribution coordinate set; performing distribution distance mean analysis on the deviation temperature measurement point distribution coordinate set to obtain the distribution distance coefficient; when the ratio of the number of deviation temperature measurement points is greater than or equal to the warning ratio threshold, and the distribution distance coefficient is less than the distribution distance threshold, generating a first-level temperature warning signal, which is set as the temperature warning signal; when the ratio of the number of deviation temperature measurement points is greater than or equal to the warning ratio threshold, and the distribution distance coefficient is greater than or equal to the distribution distance threshold, generating a second-level temperature warning signal, which is set as the temperature warning signal; wherein the warning level of the second-level temperature warning signal is higher than that of the first-level temperature warning signal.

[0072] The specific configuration of the temperature warning signal generating module 60 will be described in detail below. The temperature warning signal generating module 60 may further include: when the proportion of the number of deviation temperature measurement points is less than the warning proportion threshold, not generating the temperature warning signal.

[0073] The concrete pouring and curing control system with auxiliary temperature field monitoring provided in the embodiment of the present invention can execute the concrete pouring and curing control method with auxiliary temperature field monitoring provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0074] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0075] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for controlling concrete pouring and curing with auxiliary temperature field monitoring, characterized in that: Applicable to servers, including: Acquire concrete pouring scene features, wherein the concrete pouring scene features include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipe deployment topology; A concrete temperature prediction model is constructed based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology, wherein the concrete temperature prediction model has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model; The interactive user terminal receives a plurality of expected temperature time series information, wherein the plurality of expected temperature time series information corresponds one-to-one to the plurality of topological nodes; Receiving environmental monitoring information, wherein the environmental monitoring information includes ambient temperature, solar radiation, wind speed, and ambient humidity; According to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, the concrete initial temperature, the cooling water temperature, the cooling water flow rate, and a plurality of monitored temperature time series information are predicted to generate a plurality of concrete predicted temperature time series information; When the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information, a temperature warning signal is generated and sent to a user terminal; A concrete temperature prediction model is constructed based on a graph neural network by using the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology, including: Through the concrete pouring body equivalent model, based on the graph neural network, a concrete temperature prediction network topology is constructed, wherein the concrete temperature prediction network topology has a plurality of topological nodes, the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model; Taking the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology as constraints, retrieve the temperature field historical monitoring data, wherein any temperature field historical monitoring data includes ambient temperature record information, solar radiation record information, wind speed record information, ambient humidity record information, concrete initial temperature record information, cooling water temperature record information, cooling water flow record information, and temperature field record time series information; Divide the temperature field recording time sequence information into two segments for multiple times to construct a plurality of temperature field front segment recording time sequence information and a plurality of temperature field rear segment recording time sequence information; Extract the temperature of the temperature measuring points from the plurality of temperature field front-end recording time series information and the plurality of temperature field rear-end recording time series information according to the plurality of pre-buried temperature measuring points of the concrete pouring body equivalent model, and generate a plurality of temperature field front-end recording time series information and a plurality of temperature field rear-end recording time series information; The environmental temperature record information, the solar radiation record information, the wind speed record information, the environmental humidity record information, the concrete initial temperature record information, the cooling water temperature record information, and the cooling water flow record information are respectively combined with the plurality of temperature field front-end record time series information to generate a plurality of sets of training input data; The multiple temperature field back-end record time series information is regarded as supervision data, the multiple sets of training input data are regarded as input data, the concrete temperature prediction network topology is trained to generate the concrete temperature prediction model, wherein the input data of the corresponding temperature measurement point during the training process is input into the corresponding topological node.

2. The concrete pouring and curing control method with auxiliary temperature field monitoring according to claim 1, predicting the concrete initial temperature, cooling water temperature, cooling water flow, and several monitoring temperature time series information based on the concrete temperature prediction model, the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, and generating several concrete predicted temperature time series information, which includes: When the plurality of monitored temperature timing information do not satisfy the plurality of expected temperature timing information, a temperature alarm signal is generated and sent to a user end.

3. The concrete pouring and curing control method with auxiliary temperature field monitoring as claimed in claim 1 is characterized in that: The plurality of temperature field back-end record time series information is regarded as supervision data, the plurality of sets of training input data is regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated, including: Construct subnode training loss function: ， in, Characterizes the loss value of the i-th topological node, Characterizes the predicted temperature value of the i-th topological node at the j-th moment, represents the supervised temperature value of the i-th topological node at the j-th moment, N represents the total number of output moments, and a represents a small constant; Construct a global training loss function: ， in, represents the total number of topological nodes, Q represents the training cycle of fitting the global loss, and k represents the kth training of any cycle; According to the sub-node training loss function and the global training loss function, the plurality of temperature field back-end record time series information is regarded as supervision data, the plurality of sets of training input data are regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated.

4. The concrete pouring and curing control method with auxiliary temperature field monitoring as claimed in claim 3 is characterized in that: According to the sub-node training loss function and the global training loss function, the plurality of temperature field back-end record time series information is regarded as supervision data, the plurality of sets of training input data are regarded as input data, the concrete temperature prediction network topology is trained, and the concrete temperature prediction model is generated, including: Configure sub-node loss threshold and global loss threshold; When, in a number of consecutive training cycles, the sub-node training loss function is less than or equal to the sub-node loss threshold, and the proportion of cycles in which the global training loss function is less than or equal to the global loss threshold is greater than or equal to the convergence ratio, the concrete temperature prediction model is output.

5. The concrete pouring and curing control method with auxiliary temperature field monitoring according to claim 1, characterized in that: When the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information, a temperature warning signal is generated and sent to a user end, including: Performing deviation analysis on the plurality of concrete predicted temperature time series information and the plurality of expected temperature time series information to obtain a ratio of the number of deviation temperature measurement points; Obtain a distribution coordinate set of deviation temperature measurement points; Performing a distribution distance mean analysis on the deviation temperature measurement point distribution coordinate set to obtain a distribution distance coefficient; When the proportion of the number of the deviation temperature measurement points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is less than the distribution distance threshold, a first-level temperature warning signal is generated and set as the temperature warning signal; When the proportion of the number of the deviation temperature measurement points is greater than or equal to the warning proportion threshold, and the distribution distance coefficient is greater than or equal to the distribution distance threshold, a secondary temperature warning signal is generated and set as the temperature warning signal; Among them, the warning level of the second-level temperature warning signal is higher than that of the first-level temperature warning signal.

6. The concrete pouring and curing control method with auxiliary temperature field monitoring as claimed in claim 5, characterized in that: Also includes: When the proportion of the number of deviation temperature measurement points is less than the warning proportion threshold, the temperature warning signal is not generated.

7. A concrete pouring and curing control system with auxiliary temperature field monitoring, characterized in that: The system is used to implement the concrete pouring and curing control method with auxiliary temperature field monitoring according to any one of claims 1 to 6, and the system comprises: A concrete pouring scene feature acquisition module, wherein the concrete pouring scene feature acquisition module is used to obtain concrete pouring scene features, wherein the concrete pouring scene features include cement type, aggregate type, cement content, aggregate particle size, water-cement ratio, fly ash content, slag powder content, concrete pouring body equivalent model, and cooling pipe deployment topology; A concrete temperature prediction model construction module, the concrete temperature prediction model construction module is used to construct a concrete temperature prediction model based on a graph neural network through the cement type, the aggregate type, the cement content, the aggregate particle size, the water-cement ratio, the fly ash content, the slag powder content, the concrete pouring body equivalent model, and the cooling pipe deployment topology, wherein the concrete temperature prediction model has a plurality of topological nodes, and the plurality of topological nodes are all long short-term memory neural networks, and the plurality of topological nodes correspond one-to-one to a plurality of pre-buried temperature measurement points of the concrete pouring body equivalent model; An expected temperature timing information receiving module, the expected temperature timing information receiving module is used for interacting with the user end, receiving a plurality of expected temperature timing information, wherein the plurality of expected temperature timing information corresponds to the plurality of topological nodes one by one; An environmental monitoring information receiving module, the environmental monitoring information receiving module is used to receive environmental monitoring information, wherein the environmental monitoring information includes environmental temperature, solar radiation, wind speed, and environmental humidity; A concrete predicted temperature time series information generation module, the concrete predicted temperature time series information generation module is used to predict the concrete initial temperature, cooling water temperature, cooling water flow, and a plurality of monitored temperature time series information according to the concrete temperature prediction model, based on the ambient temperature, the solar radiation, the wind speed, and the ambient humidity, and generate a plurality of concrete predicted temperature time series information; The temperature warning signal generation module is used to generate a temperature warning signal and send it to the user end when the plurality of concrete predicted temperature time series information do not meet the plurality of expected temperature time series information.

Images