Concrete adiabatic temperature rise prediction method, device and electronic equipment

Abstract

The present invention provides a method, device and electronic device for predicting adiabatic temperature rise of concrete, which are applied to the field of digital data processing technology. The corresponding component ratio image is generated by the component data of the concrete to be predicted, and the component ratio image is input into a trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data of the concrete to be predicted at key time points, and then the adiabatic temperature rise prediction data curve of the concrete is determined. The method does not rely on test equipment, is easy to operate, and the model parameters are universal and have a wide range of applications. In addition, the adiabatic temperature rise prediction model can approximate any continuous function with arbitrary precision, thereby improving the accuracy of the adiabatic temperature rise prediction method of concrete.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical digital data processing, and in particular to a method, a device and an electronic device for predicting the thermal insulation temperature rise of concrete. Background Art

[0002] The adiabatic temperature rise of concrete is the temperature rise of concrete produced during the cement hardening process, assuming that the boundary is insulated. The adiabatic temperature rise is an important parameter of concrete, which directly affects the temperature peak value inside the concrete and the temperature difference between the inside and the outside. When the temperature difference between the inside and outside of the concrete is large, it may cause temperature cracks in the concrete, thus affecting the safety and normal use of the engineering structure.

[0003] In the related art, the adiabatic temperature rise of concrete is measured by using a concrete adiabatic temperature rise measuring instrument.

[0004] However, measuring the adiabatic temperature rise of concrete by an adiabatic temperature rise measuring instrument requires special testing equipment, and the test process is cumbersome, complicated and time-consuming. At the same time, the test results are easily affected by the test conditions and the accuracy is difficult to guarantee. Summary of the invention

[0005] The embodiments of the present invention provide a method, device and electronic device for predicting the adiabatic temperature rise of concrete, so as to solve the problem that it is difficult to quickly and accurately determine the adiabatic temperature rise of concrete.

[0006] In a first aspect, an embodiment of the present invention provides a method for predicting adiabatic temperature rise of concrete, comprising:

[0007] Generate a component proportion image according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted;

[0008] The component proportion image is input into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model; wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on a plurality of concrete component proportion images and a plurality of adiabatic temperature rise data corresponding to each concrete;

[0009] According to each adiabatic temperature rise prediction data, an adiabatic temperature rise prediction data curve of the concrete to be predicted is determined; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time.

[0010] In a possible implementation, determining the adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data includes:

[0011] Determine mutation data in the adiabatic temperature rise prediction data, and use the mutation data as a data segmentation point to perform data segmentation processing on each adiabatic temperature rise prediction data to obtain multiple segments of data to be fitted; wherein each segment of the data to be fitted includes adiabatic temperature rise prediction values ​​corresponding to multiple preset key time points, and the mutation data is adiabatic temperature rise prediction data with a slope change rate greater than a preset threshold;

[0012] Based on a preset function and the least squares method, curve fitting is performed on each segment of the data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function and a spline function;

[0013] If the fitting error of the fitting curve meets the preset condition, the fitting curve is used as the adiabatic temperature rise prediction data curve of the concrete to be predicted.

[0014] In a possible implementation, the adiabatic temperature rise prediction model has multiple groups of model parameters, and each group of model parameters corresponds to a preset key time point;

[0015] The step of inputting the component proportion image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted outputted by the adiabatic temperature rise prediction model includes:

[0016] Based on each group of model parameters of the adiabatic temperature rise prediction model, extracting feature information corresponding to the component proportion image, wherein the feature information includes one or more of edge features, corner features, color features, and geometric shape features;

[0017] The characteristic information is integrated to obtain a predicted adiabatic temperature rise value of the concrete to be predicted at a corresponding key time point.

[0018] In a possible implementation, the component data includes the proportion of each component in the mix ratio of the concrete to be predicted; and generating a component proportion image according to the component data of the concrete to be predicted includes:

[0019] According to the proportion of each component in the mix ratio of the concrete to be predicted, determine the area proportion of each partition in the component proportion image; wherein each component of the concrete to be predicted corresponds to a partition in the component proportion image;

[0020] Generate a component ratio image based on the area ratio of each partition.

[0021] In a possible implementation, before inputting the component proportion image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, the method further includes:

[0022] Inputting component proportion images of multiple concretes into the adiabatic temperature rise prediction model respectively to obtain multiple adiabatic temperature rise prediction data corresponding to each concrete;

[0023] According to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function, the model parameters of the adiabatic temperature rise prediction model are adjusted to obtain the trained adiabatic temperature rise prediction model.

[0024] In a possible implementation, adjusting the model parameters of the adiabatic temperature rise prediction model according to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function to obtain the trained adiabatic temperature rise prediction model includes:

[0025] Determining a value of the loss function based on a difference between a plurality of adiabatic temperature rise prediction data corresponding to each concrete and a plurality of adiabatic temperature rise data corresponding to each concrete;

[0026] It is determined whether the value of the loss function is less than a preset threshold value. If the value of the loss function is greater than or equal to the preset threshold value, the model parameters of the adiabatic temperature rise prediction model are adjusted, and based on the adjusted adiabatic temperature rise prediction model, the step of respectively inputting the component proportion images of multiple concretes into the adiabatic temperature rise prediction model is re-executed until the value of the loss function is less than the preset threshold value, thereby obtaining a trained adiabatic temperature rise prediction model.

[0027] In a possible implementation, the component proportion image is a pie chart, which includes sector partitions. Each component of the concrete to be predicted corresponds to a sector partition, and the area proportion of each sector partition represents the mass proportion of the corresponding component in the concrete. Each sector partition has a different filling color.

[0028] In a possible implementation, after determining the adiabatic temperature rise prediction data curve of the concrete to be predicted, the method further includes:

[0029] Based on the adiabatic temperature rise prediction data curve, cracking prevention and control is performed on the concrete to be predicted.

[0030] In a second aspect, an embodiment of the present invention provides a device, including:

[0031] A generating module, used to generate a component proportion image according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted;

[0032] The first processing module is used to input the component ratio image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model; wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on a plurality of concrete component ratio images and a plurality of adiabatic temperature rise data corresponding to each concrete;

[0033] The second processing module is used to determine the adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time.

[0034] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the method described in the first aspect or any possible implementation manner of the first aspect.

[0035] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation method of the first aspect are implemented.

[0036] The embodiment of the present invention provides a method, device and electronic device for predicting the adiabatic temperature rise of concrete. The corresponding component ratio image is generated through the component data of the concrete to be predicted, and the component ratio image is input into a trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data of the concrete to be predicted at key time points, and then the adiabatic temperature rise prediction data curve of the concrete is determined. Among them, this embodiment does not rely on test equipment, is easy to operate, and the model parameters are universal and have a wide range of applications. In addition, the adiabatic temperature rise prediction model can approximate any continuous function with arbitrary precision, thereby improving the accuracy of the adiabatic temperature rise prediction method of concrete. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 It is a flow chart of the implementation of the concrete adiabatic temperature rise prediction method provided in the embodiment of the present application;

[0039] Figure 2 A schematic diagram of a convolutional neural network structure provided in an embodiment of the present application;

[0040] Figure 3 A schematic diagram of adiabatic temperature rise prediction data curve provided in an embodiment of the present application;

[0041] Figure 4 This is a specific implementation flow chart of the concrete adiabatic temperature rise prediction method provided in the embodiment of the present application;

[0042] Figure 5 It is a structural schematic diagram of a concrete insulation temperature rise prediction device provided in an embodiment of the present application;

[0043] Figure 6 It is a schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0045] In the related art, there are two main methods for testing the adiabatic temperature rise of concrete: the indirect method and the direct method. The direct method uses a concrete adiabatic temperature rise measuring instrument to directly measure the adiabatic temperature rise of concrete; the indirect method mainly calculates the adiabatic temperature rise of concrete based on parameters such as cement hydration heat, cement dosage, and concrete specific heat, and relies on theoretical calculations. After research, the inventors found that the direct method of concrete adiabatic temperature rise testing requires special test equipment, the test process is cumbersome and complicated, and time-consuming. At the same time, the test results are easily affected by the test conditions, and the accuracy is difficult to guarantee; the indirect method relies on theoretical calculations, and its accuracy is greatly affected by the theoretical abstract model. In addition, in many related theoretical models, the cementitious material in concrete only contains cement, which leads to a limited application range of the theoretical abstract model.

[0046] In order to improve the accuracy, universality and ease of operation of the adiabatic temperature rise prediction method of concrete, in the implementation mode of the present application, the corresponding component proportion image is generated through the component data of concrete, and the component proportion image is input into the trained neural network model to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete. The method is independent of test equipment and easy to operate. The characteristic of neural network that can approximate any continuous function with arbitrary precision is utilized, thereby improving the accuracy of the adiabatic temperature rise prediction method of concrete and making the scope of application easier to expand.

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0048] See also Figure 1 , which shows a flow chart of the implementation of the concrete adiabatic temperature rise prediction method provided by an embodiment of the present invention, which is described in detail as follows:

[0049] In step 101, a component proportion image is generated according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted.

[0050] Here, the component data of concrete may include the proportion of each component in the mix ratio of concrete to be predicted, and each component in the concrete mix ratio may include ordinary Portland cement, fly ash, slag, silica fume, water, fine aggregate, coarse aggregate, etc.

[0051] In this embodiment, the component proportion image may be an image representing the proportion of each component in the mix ratio of concrete to be predicted, such as a pie chart, a bar chart, a sector chart, etc.

[0052] The concrete to be predicted can be understood as the concrete that requires adiabatic temperature rise prediction, such as large volume concrete.

[0053] Exemplarily, after receiving the component data of the concrete to be predicted, the embodiment of the present application can generate a corresponding component proportion image based on the received component data of the concrete to be predicted, so that the adiabatic temperature rise of the concrete can be predicted based on the component proportion image later, without relying on test equipment, which is convenient and fast.

[0054] In step 102, the component proportion image is input into the trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on the component proportion images of the plurality of concretes and the plurality of adiabatic temperature rise data corresponding to each concrete.

[0055] Among them, the preset key time point in the embodiment of the present application is a time point in a representative concrete adiabatic temperature rise curve found through research. Here, the time point in the representative concrete adiabatic temperature rise curve can be determined based on the change of adiabatic temperature rise data of multiple concretes over time, such as taking the time point with a large change in adiabatic temperature rise value as the time point in the representative concrete adiabatic temperature rise curve, or taking the time point that can reflect the change trend of concrete adiabatic temperature rise value as the time point in the representative concrete adiabatic temperature rise curve.

[0056] In this embodiment, the trained adiabatic temperature rise prediction model can be a trained convolutional neural network model. The adiabatic temperature rise prediction model is trained based on multiple concrete component ratio images and multiple adiabatic temperature rise data corresponding to each concrete. Therefore, the diversity of the components of multiple concretes in the training data can expand the application scope of the model, and the influence of admixtures can be included in the prediction process, thereby achieving a wider prediction range; the accuracy of the model is improved by the number of training data. The adiabatic temperature rise prediction model is trained with the input item being the component ratio image of the concrete and the output item being the adiabatic temperature rise value of the concrete at a key time point.

[0057] Exemplarily, the embodiment of the present application inputs the generated component proportion image into a trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, that is, the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point.

[0058] In step 103, an adiabatic temperature rise prediction data curve of the concrete to be predicted is determined according to each adiabatic temperature rise prediction data; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time.

[0059] Exemplarily, the embodiment of the present application can generate an adiabatic temperature rise prediction data curve of the concrete to be predicted based on various adiabatic temperature rise prediction data of the concrete to be predicted output by the adiabatic temperature rise prediction model, and the curve can represent the change of the adiabatic temperature rise value of the concrete to be predicted over time.

[0060] The adiabatic temperature rise prediction data curve can be obtained by fitting the adiabatic temperature rise prediction data of each concrete to be predicted. The adiabatic temperature rise prediction data curve can indicate the trend of the adiabatic temperature rise of the concrete to be predicted over time, and then, after verification of the fitting error, it can be applied to the crack prevention and control of concrete to ensure the safety of the engineering structure.

[0061] The embodiment of the present invention generates a corresponding component ratio image through the component data of the concrete to be predicted, and inputs the component ratio image into a trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data of the concrete to be predicted at key time points, and then determines the adiabatic temperature rise prediction data curve of the concrete. The method does not rely on test equipment, is easy to operate, and has an easily expandable application range. The adiabatic temperature rise prediction model can approximate any continuous function with arbitrary precision, thereby improving the accuracy of the concrete adiabatic temperature rise prediction method.

[0062] The following further describes how to determine the adiabatic temperature rise prediction data curve of the concrete to be predicted based on various adiabatic temperature rise prediction data.

[0063] In some embodiments, determining an adiabatic temperature rise prediction data curve of concrete to be predicted based on various adiabatic temperature rise prediction data includes:

[0064] First, the mutation data in the adiabatic temperature rise prediction data is determined, and the mutation data is used as the data segmentation point to perform data segmentation processing on each adiabatic temperature rise prediction data to obtain multiple segments of data to be fitted; wherein each segment of data to be fitted includes adiabatic temperature rise prediction values ​​corresponding to multiple preset key time points, and the mutation data is adiabatic temperature rise prediction data with a slope change rate greater than a preset threshold. Then, based on the preset function and the least squares method, curve fitting processing is performed on each segment of data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function, and a spline function.

[0065] If it is determined that the fitting error of the fitting curve meets the preset condition, the fitting curve is determined to be the adiabatic temperature rise prediction data curve of the concrete to be predicted.

[0066] Here, in order to improve the accuracy of data fitting, the embodiment of the present application can perform segmented processing on each adiabatic temperature rise prediction data: the adiabatic temperature rise prediction values ​​of multiple key time points corresponding to the concrete to be predicted are arranged in chronological order, and then the preset thresholds of the slope and the slope change rate between the adiabatic temperature rise prediction values ​​of each two adjacent key time points are determined, and the mutation data in the adiabatic temperature rise prediction data are determined, wherein the mutation data characterizes that the adiabatic temperature rise value of the concrete to be predicted has experienced a sudden increase or decrease in value exceeding the preset threshold.

[0067] Then, based on a preset function and the least square method, curve fitting processing is performed on each segment of the data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function and a spline function.

[0068] Finally, the fitting error of the fitting curve is calculated. If it is determined that the fitting error of the fitting curve meets the preset conditions, the fitting curve is determined to be the adiabatic temperature rise prediction data curve of the concrete to be predicted; if it is determined that the fitting error of the fitting curve does not meet the preset conditions, the curve fitting is performed again.

[0069] In a possible implementation, the types of functions that can be used include exponential functions, low-order power functions (such as quadratic or cubic power functions), spline functions, etc. The preset condition for fitting can be the average absolute percentage error required for the accuracy of the fitting. Should be less than 3%. Among them, the mean absolute percentage error The expression is as follows:

[0070]

[0071] in is the adiabatic temperature rise prediction data to be fitted, is the data corresponding to the fitting formula, and n is the number of fitted adiabatic temperature rise prediction data.

[0072] In a possible implementation, the root mean square error and the correlation coefficient may be used to judge the fitting effect. If the root mean square error is smaller and the correlation coefficient is larger, the fitting effect is better.

[0073] In some embodiments, after determining the adiabatic temperature rise prediction data curve of the concrete to be predicted, this embodiment can also prevent and control cracking of the concrete to be predicted based on the adiabatic temperature rise prediction data curve.

[0074] Illustratively, after determining the adiabatic temperature rise prediction data curve of the concrete to be predicted, the embodiments of the present application can prevent and control cracking of structures and projects using the concrete to be predicted based on the adiabatic temperature rise prediction data curve to ensure structural safety.

[0075] This embodiment uses mutation data as a benchmark, segments the adiabatic temperature rise prediction data corresponding to the concrete to be predicted, generates a fitting curve based on a preset function and the least squares method, verifies the fitting error, and finally determines the adiabatic temperature rise prediction data curve, thereby improving the accuracy of the adiabatic temperature rise prediction data curve and further ensuring the safety of the structure.

[0076] The above-mentioned embodiment introduces how to generate an adiabatic temperature rise prediction curve based on multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted. The following further introduces how to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model based on the component proportion image and the trained adiabatic temperature rise prediction model.

[0077] In some embodiments, the adiabatic temperature rise prediction model has multiple groups of model parameters, each group of model parameters corresponds to a preset key time point. In this embodiment, when the component proportion image is input into the trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, the corresponding component proportion image feature information can be first extracted based on each group of model parameters of the adiabatic temperature rise prediction model, wherein the feature information includes one or more of edge features, corner features, color features, and geometric shape features. Then, the corresponding feature information is integrated to obtain the adiabatic temperature rise prediction value of the concrete to be predicted at the corresponding key time point.

[0078] Among them, the adiabatic temperature rise prediction model has multiple groups of model parameters, each group of model parameters corresponds to a preset key time point, and each group of model parameters is used to calculate and output the adiabatic temperature rise prediction value corresponding to the key time point after inputting the component proportion image.

[0079] The embodiment of the present application extracts the corresponding component proportion image feature information based on each group of model parameters of the adiabatic temperature rise prediction model, including one or more of edge features, corner features, color features, and geometric shape features. Then, the corresponding feature information is integrated and processed to obtain the adiabatic temperature rise prediction value of the concrete to be predicted at the corresponding key time point, thereby improving the accuracy of the adiabatic temperature rise prediction.

[0080] In a possible implementation, the adiabatic temperature rise prediction model in this embodiment can be a convolutional neural network, and the weights and biases of each layer are obtained through training. To improve the prediction accuracy, the number of parameters to be determined for the convolutional neural network is not less than 3 million, and it should include at least three convolutional layers, and each convolutional layer is used in conjunction with a pooling layer, and the entire convolutional neural network includes at least one nonlinear activation function. The edges, corners, colors, geometric shapes and other features of the input image are extracted layer by layer through convolution operations.

[0081] Among them, the adiabatic temperature rise prediction model may include an input layer, a hidden layer and an output layer. The input layer is used to obtain the component proportion image; the hidden layer includes a number of sequentially connected convolutional layers, pooling layers and fully connected layers, each convolutional layer is used for the image features of the component proportion image, the pooling layer is used to reduce the data dimension of the image features and extract the main features, and the fully connected layer is used to integrate all features to generate regression prediction results; the output layer is used to output the adiabatic temperature rise prediction data; an activation function (Relu) is also set between the convolutional layer and the pooling layer, and between each fully connected layer, and an activation function (Linear) is set between the fully connected layer and the output layer. For example, Figure 2The present embodiment provides a convolutional neural network, which includes a hidden layer, and the hidden layer includes several sequentially connected convolution layers, pooling layers and fully connected layers. As shown in the figure, it includes convolution layers 1, 2, 3, 4 and 5, maximum pooling layers 1, 2, 3, 4 and 5, and a fully connected layer.

[0082] Optionally, in some embodiments, the component data includes the proportion of each component in the mix ratio of the concrete to be predicted; generating a component proportion image according to the component data of the concrete to be predicted includes:

[0083] According to the proportion of each component in the mix ratio of the concrete to be predicted, the area proportion of each partition in the component proportion image is determined; wherein each component of the concrete to be predicted corresponds to a partition in the component proportion image. According to the area proportion of each partition, a component proportion image is generated.

[0084] Among them, the component data of the concrete to be predicted can be the proportion of each component in the mix ratio of the concrete to be predicted, for example, the proportion of ordinary Portland cement, the proportion of fly ash, the proportion of slag, the proportion of silica fume, the proportion of water, the proportion of fine aggregate and the proportion of coarse aggregate in the mix ratio.

[0085] Among them, each component in the mix ratio independently corresponds to a partition in the component proportion image. Therefore, this embodiment determines the area proportion of each partition in the component proportion image according to the proportion of each component in the mix ratio of the concrete to be predicted, and then generates the component proportion image according to the area proportion of each partition.

[0086] Optionally, in some embodiments, the component proportion image is a pie chart, which includes sector-shaped partitions corresponding to each component, the area proportion of each sector-shaped partition represents the mass proportion of the corresponding component in the concrete, and each sector-shaped partition has a different filling color.

[0087] Exemplarily, this embodiment generates a component proportion image in the form of a pie chart based on the component data of the concrete to be predicted. In the pie chart, each component in the concrete mix corresponds to a sector-shaped partition, and the area proportion of each sector-shaped partition represents the mass proportion of the corresponding component in the concrete, that is, the mass percentage. For further distinction, each sector-shaped partition has a different filling color.

[0088] In a possible implementation, each sector is colored with a different color system (RGB) value; the sector colors of the pie chart should be greatly differentiated. For example, the RGB value range of the sector partition corresponding to ordinary Portland cement in the pie chart is (0, 0, 1~127); the RGB value range of the sector partition corresponding to fly ash in the pie chart is (0, 0, 128~255); the RGB value range of the sector partition corresponding to slag in the pie chart is (0, 1~127, 0); the RGB value range of the sector partition corresponding to silica ash in the pie chart is (0, 128~255, 0); the RGB value range of the sector partition corresponding to water in the pie chart is (1~85, 0, 0); the RGB value range of the sector partition corresponding to fine aggregate in the pie chart is (86~170, 0, 0); the RGB value range of the sector partition corresponding to coarse aggregate in the pie chart is (171~255, 0, 0).

[0089] This embodiment generates a corresponding component proportion image based on the component data of the concrete to be predicted, and strengthens the distinction between each component in the component proportion image, and then extracts the edge features, corner features, color features and geometric shape features of the corresponding component proportion image based on each set of model parameters of the adiabatic temperature rise prediction model, and then integrates the corresponding feature information to obtain the adiabatic temperature rise prediction value of the concrete to be predicted at the corresponding key time point. This process can further improve the recognition ability of the adiabatic temperature rise prediction model by strengthening the distinction between the proportions of each component in the concrete mix ratio, thereby improving the accuracy of the output of the adiabatic temperature rise prediction model; the adiabatic temperature rise prediction model has multiple sets of model parameters, and each set of parameters is used to obtain the adiabatic temperature rise prediction value of the concrete to be predicted at a key time point, which can improve the calculation speed and accuracy.

[0090] The above embodiment introduces how to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model based on the component proportion image and the trained adiabatic temperature rise prediction model. The following further introduces how to obtain the trained adiabatic temperature rise prediction model.

[0091] In some embodiments, before inputting the component proportion image into the trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, the method further includes:

[0092] The component proportion images of multiple concretes are respectively input into the adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data corresponding to each concrete.

[0093] According to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function, the model parameters of the adiabatic temperature rise prediction model are adjusted to obtain the trained adiabatic temperature rise prediction model.

[0094] In some embodiments, according to a plurality of adiabatic temperature rise prediction data corresponding to each concrete, a plurality of adiabatic temperature rise data corresponding to each concrete and a loss function, a model parameter of the adiabatic temperature rise prediction model is adjusted to obtain a trained adiabatic temperature rise prediction model, including:

[0095] The value of the loss function is determined based on the difference between the plurality of adiabatic temperature rise prediction data corresponding to each concrete and the plurality of adiabatic temperature rise data corresponding to each concrete.

[0096] It is determined whether the value of the loss function is less than a preset threshold. If the value of the loss function is greater than or equal to the preset threshold, the model parameters of the adiabatic temperature rise prediction model are adjusted, and based on the adjusted adiabatic temperature rise prediction model, the above steps of inputting the component proportion images of multiple concretes into the adiabatic temperature rise prediction model are re-executed until the value of the loss function is less than the preset threshold, thereby obtaining a trained adiabatic temperature rise prediction model.

[0097] Exemplarily, the embodiment of the present application pre-generates a large amount of concrete adiabatic temperature rise data, and generates corresponding component proportion images based on these data, and then inputs these multiple concrete component proportion images into the adiabatic temperature rise prediction model. For example, in the convolutional neural network model, multiple adiabatic temperature rise prediction data corresponding to each concrete are obtained based on the output of the adiabatic temperature rise prediction model, and then, based on the difference between the multiple adiabatic temperature rise prediction data corresponding to each concrete and the multiple actual adiabatic temperature rise data corresponding to each concrete, the value of the loss function is determined; then, it is determined whether the value of the loss function is less than a preset threshold value. If the value of the loss function is greater than or equal to the preset threshold value, the model parameters of the adiabatic temperature rise prediction model are adjusted, and the training is performed again until the value of the loss function is less than the preset threshold value, so as to obtain a trained adiabatic temperature rise prediction model.

[0098] Here, this embodiment performs training based on training data, loss function and model output in the model training process, so that the adiabatic temperature rise prediction of concrete can be predicted based on the trained model in the future, and the characteristics of any continuous function can be approximated with arbitrary precision, thereby improving the accuracy of the adiabatic temperature rise prediction method of concrete. And the obtained model has good reusability. After the model parameter training is completed, the adiabatic temperature rise prediction can be carried out for many concretes with the same components and different proportions.

[0099] In a feasible implementation, this embodiment collects no less than 5000 concrete adiabatic temperature rise curves in advance, calculates the mass percentage of each component in the mix ratio corresponding to each curve, and represents the mass percentage of each component with a pie chart, and each sector is colored with a different RGB value; then reads the adiabatic temperature rise values ​​of 1 hour, 2 hours, 4 hours, 8 hours, 16 hours, 32 hours, 64 hours, and 128 hours from each adiabatic temperature rise curve, and rounds it to the single digit; then selects the convolutional neural network model as the adiabatic temperature rise prediction model, uses each pie chart as a graphic input, and uses the adiabatic temperature rise of the ith hour as a label for training, and obtains a total of 8 groups of parameters; wherein the classification categories of the convolutional neural network are numbers from 1 to 60. Exemplary, such as Figure 2 The convolutional neural network shown.

[0100] Optionally, the RGB value range of the color of the sector partition corresponding to ordinary Portland cement in the pie chart is (0, 0, 1~127); the RGB value range of the color of the sector partition corresponding to fly ash in the pie chart is (0, 0, 128~255); the RGB value range of the color of the sector partition corresponding to slag in the pie chart is (0, 1~127, 0); the RGB value range of the color of the sector partition corresponding to silica ash in the pie chart is (0, 128~255, 0); the RGB value range of the color of the sector partition corresponding to water in the pie chart is (1~85, 0, 0); the RGB value range of the color of the sector partition corresponding to fine aggregate in the pie chart is (86~170, 0, 0); the RGB value range of the color of the sector partition corresponding to coarse aggregate in the pie chart is (171~255, 0, 0).

[0101] Optionally, before training the convolutional neural network, the training data should be enhanced. The pie chart can be rotated at any angle without changing its shape and color, thereby generating more usable input images. In this process, the corresponding label value remains unchanged.

[0102] After obtaining the model parameters of each group of trained adiabatic temperature rise prediction models, adiabatic temperature rise prediction data of the concrete to be predicted can be obtained based on each group of model parameters.

[0103] In a feasible implementation method, the concrete to be predicted is the basement floor of a high-rise building. The concrete mix ratio is ordinary Portland cement: fly ash: slag: silica ash: water: fine aggregate: coarse aggregate = 400: 72: 85: 50: 151: 1060: 570. It can be obtained that the mass percentage of each component is 16.75%, 3.015%, 3.559%, 2.094%, 6.323%, 44.389%, and 23.869%, respectively. Based on this, a component proportion image is generated, such as a color pie chart, and each component corresponds to a different color sector partition in the figure. The color pie chart is input into the trained convolutional neural network, and based on each group of model parameters, the adiabatic temperature rise prediction values ​​of the 1st, 2nd, 4th, 8th, 16th, 32nd, 64th, and 128th hours are obtained. The spline function is used for smoothing to obtain the adiabatic temperature rise curve of the concrete to be predicted within 128 hours as shown in FIG. Figure 3 As shown, Figure 3 This is a schematic diagram of the adiabatic temperature rise prediction data curve provided in an embodiment of the present application. It can be seen from the figure that after about 100 hours of hydration, the temperature no longer increases.

[0104] In the embodiment of the present application, an adiabatic temperature rise prediction model is trained based on real concrete adiabatic temperature rise data, which can improve the accuracy of model prediction. In addition, the obtained adiabatic temperature rise prediction model with model parameters is universal and can repeatedly predict the adiabatic temperature rise of concrete with different mix ratios. It has the advantages of accurate, convenient and efficient analysis results, and can provide new ideas and new tools for concrete adiabatic temperature rise prediction.

[0105] Figure 4 A specific implementation flow chart of the concrete adiabatic temperature rise prediction method provided in the embodiment of the present application is shown, and the details are as follows:

[0106] In step 401, component proportion images of a plurality of concretes are respectively input into an adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to each concrete.

[0107] In step 402, according to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function, the model parameters of the adiabatic temperature rise prediction model are adjusted to obtain a trained adiabatic temperature rise prediction model.

[0108] In one example, step 402 includes the following steps:

[0109] The value of the loss function is determined based on the difference between the plurality of adiabatic temperature rise prediction data corresponding to each concrete and the plurality of adiabatic temperature rise data corresponding to each concrete.

[0110] It is determined whether the value of the loss function is less than a preset threshold value. If the value of the loss function is greater than or equal to the preset threshold value, the model parameters of the adiabatic temperature rise prediction model are adjusted, and based on the adjusted adiabatic temperature rise prediction model, the step of respectively inputting the component proportion images of the plurality of concretes into the adiabatic temperature rise prediction model is re-executed until the value of the loss function is less than the preset threshold value, thereby obtaining a trained adiabatic temperature rise prediction model.

[0111] The adiabatic temperature rise prediction model has multiple groups of model parameters, and each group of model parameters corresponds to a preset key time point.

[0112] In step 403, a component proportion image is generated according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted.

[0113] In one example, the component data includes the proportion of each component in the mix ratio of the concrete to be predicted; step 403 includes the following:

[0114] According to the proportion of each component in the mix ratio of the concrete to be predicted, the area proportion of each partition in the component proportion image is determined; wherein each component of the concrete to be predicted corresponds to a partition in the component proportion image.

[0115] Generate a component ratio image based on the area ratio of each partition.

[0116] In one example, the component proportion image is a pie chart, which includes sector-shaped partitions corresponding to each component. The area proportion of each sector-shaped partition represents the mass proportion of the corresponding component in the concrete, and each sector-shaped partition has a different filling color.

[0117] In step 404, based on each group of model parameters of the adiabatic temperature rise prediction model, the corresponding component proportion image feature information is extracted, wherein the feature information includes one or more of edge features, corner features, color features, and geometric shape features.

[0118] In step 405, the corresponding characteristic information is integrated to obtain the predicted adiabatic temperature rise value of the concrete to be predicted at the corresponding key time point.

[0119] In step 406, mutation data in the adiabatic temperature rise prediction data is determined, and the mutation data is used as a data segmentation point to perform data segmentation processing on each adiabatic temperature rise prediction data to obtain multiple segments of data to be fitted; wherein each segment of data to be fitted includes adiabatic temperature rise prediction values ​​corresponding to multiple preset key time points, and the mutation data is adiabatic temperature rise prediction data whose slope change rate is greater than a preset threshold.

[0120] In step 407, based on a preset function and the least square method, curve fitting is performed on each segment of the data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function and a spline function.

[0121] In step 408, if the fitting error of the fitting curve meets the preset condition, the fitting curve is used as the adiabatic temperature rise prediction data curve of the concrete to be predicted.

[0122] In step 409, based on the adiabatic temperature rise prediction data curve, crack prevention and control is performed on the concrete to be predicted.

[0123] The detailed implementation process of the above steps 401 to 409 can refer to the relevant descriptions of the above embodiments, which will not be repeated here. The execution order of the above steps 401 to 409 can be set according to actual needs and is not specifically limited here.

[0124] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0125] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0126] Figure 5 The schematic diagram of the structure of the concrete adiabatic temperature rise prediction device provided by the embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0127] like Figure 5 As shown, the concrete adiabatic temperature rise prediction device 500 includes:

[0128] A generating module 501 is used to generate a component proportion image according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted;

[0129] The first processing module 502 is used to input the component proportion image into the trained adiabatic temperature rise prediction model to obtain multiple adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on the component proportion images of multiple concretes and the multiple adiabatic temperature rise data corresponding to each concrete.

[0130] The second processing module 503 is used to determine the adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time.

[0131] In a possible implementation, the second processing module 503 is specifically configured to:

[0132] Determine the mutation data in the adiabatic temperature rise prediction data, and use the mutation data as the data segmentation point to perform data segmentation processing on each adiabatic temperature rise prediction data to obtain multiple segments of data to be fitted; wherein each segment of data to be fitted includes adiabatic temperature rise prediction values ​​corresponding to multiple preset key time points, and the mutation data is the adiabatic temperature rise prediction data whose slope change rate is greater than a preset threshold.

[0133] Based on a preset function and the least squares method, curve fitting is performed on each segment of the data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function and a spline function.

[0134] If the fitting error of the fitting curve meets the preset conditions, the fitting curve is used as the adiabatic temperature rise prediction data curve of the concrete to be predicted.

[0135] In a possible implementation, the adiabatic temperature rise prediction model has multiple groups of model parameters, each group of model parameters corresponds to a preset key time point; the first processing module 502 is specifically used to:

[0136] Based on each group of model parameters of the adiabatic temperature rise prediction model, the corresponding component proportion image feature information is extracted, wherein the feature information includes one or more of edge features, corner features, color features and geometric shape features.

[0137] The corresponding characteristic information is integrated and processed to obtain the predicted adiabatic temperature rise value of the concrete to be predicted at the corresponding key time point.

[0138] In a possible implementation, the component data includes the proportion of each component in the mix ratio of the concrete to be predicted; the generating module 501 is specifically used for:

[0139] According to the proportion of each component in the mix ratio of the concrete to be predicted, the area proportion of each partition in the component proportion image is determined; wherein each component of the concrete to be predicted corresponds to a partition in the component proportion image.

[0140] Generate a component ratio image based on the area ratio of each partition.

[0141] In a possible implementation, before the first processing module 502, the apparatus 500 further includes:

[0142] The first training module is used to: input component ratio images of multiple concretes into the adiabatic temperature rise prediction model respectively to obtain multiple adiabatic temperature rise prediction data corresponding to each concrete.

[0143] The second training module is used to adjust the model parameters of the adiabatic temperature rise prediction model according to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function to obtain the trained adiabatic temperature rise prediction model.

[0144] In a possible implementation, the second training module is specifically used for:

[0145] The value of the loss function is determined based on the difference between the plurality of adiabatic temperature rise prediction data corresponding to each concrete and the plurality of adiabatic temperature rise data corresponding to each concrete.

[0146] It is determined whether the value of the loss function is less than a preset threshold value. If the value of the loss function is greater than or equal to the preset threshold value, the model parameters of the adiabatic temperature rise prediction model are adjusted, and based on the adjusted adiabatic temperature rise prediction model, the step of respectively inputting the component proportion images of the plurality of concretes into the adiabatic temperature rise prediction model is re-executed until the value of the loss function is less than the preset threshold value, thereby obtaining a trained adiabatic temperature rise prediction model.

[0147] In a possible implementation, after the second processing module 503, the apparatus 500 further includes:

[0148] The monitoring module is used to prevent and control cracking of the predicted concrete based on the adiabatic temperature rise prediction data curve.

[0149] Figure 6 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 6 As shown, the electronic device 6 of this embodiment includes: a processor 60 and a memory 61. The memory 61 is used to store a computer program 62, and the processor 60 is used to call and run the computer program 62 stored in the memory 61 to perform the steps in the above-mentioned concrete adiabatic temperature rise prediction method embodiments, such as Figure 1 Alternatively, the processor 60 is used to call and run the computer program 62 stored in the memory 61 to implement the functions of each module / unit in the above-mentioned device embodiments, such as Figure 5 Functions of modules 501 to 503 are shown.

[0150] Exemplarily, the computer program 62 may be divided into one or more modules / units, one or more modules / units are stored in the memory 61 and executed by the processor 60 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments that can implement specific functions, and the instruction segments are used to describe the execution process of the computer program 62 in the electronic device 6. For example, the computer program 62 may be divided into Figure 5 Modules 501 to 503 are shown.

[0151] The electronic device 6 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 6 may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will appreciate that Figure 6 It is only an example of the electronic device 6 and does not constitute a limitation of the electronic device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device 6 may also include input and output devices, network access devices, buses, etc.

[0152] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0153] The memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. The memory 61 may also be an external storage device of the electronic device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 6. Further, the memory 61 may also include both an internal storage unit of the electronic device 6 and an external storage device. The memory 61 is used to store computer programs and other programs and data required by the terminal. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0154] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0155] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0156] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0157] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0158] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0159] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0160] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned concrete adiabatic temperature rise prediction method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0161] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for predicting adiabatic temperature rise of concrete, characterized in that: The method comprises: According to the proportion of each component in the mix ratio of the concrete to be predicted, determine the area proportion of each partition in the component proportion image; wherein each component of the concrete to be predicted corresponds to a partition in the component proportion image; Generate a component proportion image according to the area proportion of each partition; wherein the component proportion image is a pie chart, and the pie chart includes sector-shaped partitions; The component proportion image is input into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model; wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on a plurality of concrete component proportion images and a plurality of adiabatic temperature rise data corresponding to each concrete; Determine an adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time; Wherein, the adiabatic temperature rise prediction model has multiple groups of model parameters, and each group of model parameters corresponds to a preset key time point; The step of inputting the component proportion image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted outputted by the adiabatic temperature rise prediction model includes: Based on each group of model parameters of the adiabatic temperature rise prediction model, feature information corresponding to the component proportion image is extracted, wherein the feature information includes edge features, corner features, color features, and geometric shape features; The characteristic information is integrated to obtain a predicted adiabatic temperature rise value of the concrete to be predicted at a corresponding key time point.

2. The method according to claim 1, characterized in that The step of determining the adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data comprises: Determine mutation data in the adiabatic temperature rise prediction data, and use the mutation data as a data segmentation point to perform data segmentation processing on each adiabatic temperature rise prediction data to obtain multiple segments of data to be fitted; wherein each segment of the data to be fitted includes adiabatic temperature rise prediction values ​​corresponding to multiple preset key time points, and the mutation data is adiabatic temperature rise prediction data with a slope change rate greater than a preset threshold; Based on a preset function and the least squares method, curve fitting is performed on each segment of the data to be fitted to generate a fitting curve; wherein the preset function is any one or more of an exponential function, a power function and a spline function; If the fitting error of the fitting curve meets the preset condition, the fitting curve is used as the adiabatic temperature rise prediction data curve of the concrete to be predicted.

3. The method according to any one of claims 1 to 2, characterized in that Before inputting the component proportion image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model, the method further includes: Inputting component proportion images of multiple concretes into the adiabatic temperature rise prediction model respectively to obtain multiple adiabatic temperature rise prediction data corresponding to each concrete; According to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function, the model parameters of the adiabatic temperature rise prediction model are adjusted to obtain the trained adiabatic temperature rise prediction model.

4. The method according to claim 3, characterized in that The method of adjusting the model parameters of the adiabatic temperature rise prediction model according to the multiple adiabatic temperature rise prediction data corresponding to each concrete, the multiple adiabatic temperature rise data corresponding to each concrete and the loss function to obtain the trained adiabatic temperature rise prediction model includes: Determining a value of the loss function based on a difference between a plurality of adiabatic temperature rise prediction data corresponding to each concrete and a plurality of adiabatic temperature rise data corresponding to each concrete; It is determined whether the value of the loss function is less than a preset threshold value. If the value of the loss function is greater than or equal to the preset threshold value, the model parameters of the adiabatic temperature rise prediction model are adjusted, and based on the adjusted adiabatic temperature rise prediction model, the step of respectively inputting the component proportion images of multiple concretes into the adiabatic temperature rise prediction model is re-executed until the value of the loss function is less than the preset threshold value, thereby obtaining a trained adiabatic temperature rise prediction model.

5. The method according to claim 1, characterized in that Each component of the concrete to be predicted corresponds to a sector-shaped partition, the area ratio of each sector-shaped partition represents the mass ratio of the corresponding component in the concrete, and each sector-shaped partition has a different filling color.

6. The method according to any one of claims 1 to 2, characterized in that After determining the adiabatic temperature rise prediction data curve of the concrete to be predicted, the method further includes: Based on the adiabatic temperature rise prediction data curve, cracking prevention and control is performed on the concrete to be predicted.

7. A concrete adiabatic temperature rise prediction device, which executes the concrete adiabatic temperature rise prediction method according to any one of claims 1 to 6, characterized in that: include: A generating module, used to generate a component proportion image according to the component data of the concrete to be predicted; wherein the component proportion image is used to represent the proportion of each component in the mix ratio of the concrete to be predicted; The first processing module is used to input the component ratio image into a trained adiabatic temperature rise prediction model to obtain a plurality of adiabatic temperature rise prediction data corresponding to the concrete to be predicted output by the adiabatic temperature rise prediction model; wherein the adiabatic temperature rise prediction data is the adiabatic temperature rise prediction value corresponding to the concrete to be predicted at a preset key time point; the adiabatic temperature rise prediction model is trained based on a plurality of concrete component ratio images and a plurality of adiabatic temperature rise data corresponding to each concrete; The second processing module is used to determine the adiabatic temperature rise prediction data curve of the concrete to be predicted according to each adiabatic temperature rise prediction data; wherein the adiabatic temperature rise prediction data curve is used to characterize the change of the adiabatic temperature rise value of the concrete to be predicted over time; The adiabatic temperature rise prediction model has multiple groups of model parameters, each group of model parameters corresponds to a preset key time point; the first processing module is also used for: Based on each group of model parameters of the adiabatic temperature rise prediction model, the corresponding component proportion image feature information is extracted, wherein the feature information includes edge features, corner features, color features and geometric shape features; The corresponding characteristic information is integrated and processed to obtain the predicted adiabatic temperature rise value of the concrete to be predicted at the corresponding key time point.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the concrete adiabatic temperature rise prediction method according to any one of claims 1 to 6.

Images