Intelligent condenser layout system based on machine learning

Abstract

The present invention relates to the technical field of concrete pouring, and discloses an intelligent layout system for condenser pipes based on machine learning, comprising a condition input layer, a layout analysis layer, a data recording layer and a result output layer; the condition input layer is responsible for receiving control conditions that affect concrete temperature changes input by users, the layout analysis layer trains a temperature change prediction model, and the model can predict the maximum temperature rise of concrete during the hardening process based on the condition data input by users, the layout analysis module calls the prediction model according to these condition data, obtains the predicted maximum temperature value, and sends it to the data recording layer together with the control conditions input by users, the data recording layer is used to save these data, and the result output layer displays the input conditions and prediction results to users in a visual form. The system can accurately determine a reasonable condenser pipe layout method before pouring large-volume concrete, effectively improve construction quality, and reduce the cost of later maintenance and adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete pouring, and in particular to a condenser pipe intelligent layout system based on machine learning. Background Art

[0002] In large-volume concrete construction, since concrete will generate a large amount of hydration heat during the hardening process, if the heat cannot be dissipated in a timely and effective manner, it may cause the temperature inside the concrete to be too high, thereby causing temperature cracks, which seriously affects the strength and durability of the concrete structure. In order to solve this problem, condensation pipes are usually used for cooling during construction to promptly remove the heat generated during the hardening process, thereby maintaining a stable temperature inside the concrete.

[0003] However, the layout design of the condenser is a highly complex issue, which is affected by many factors, such as the diameter of the condenser, the water flow rate, the layout method, etc. Traditional design methods rely on the experience and intuition of engineers, and it is difficult to ensure the optimal layout. In actual construction, unreasonable condenser layout may lead to poor cooling effect, unsatisfactory temperature control inside the concrete, and even cause quality problems.

[0004] If the unreasonable layout of the condenser pipe is found after the concrete is poured, it will be extremely difficult to adjust it, and it may cause damage to the poured concrete. Therefore, it is very important to accurately determine the reasonable layout of the condenser pipe before construction, which can not only improve the construction quality, but also effectively reduce the cost of later maintenance and adjustment. To this end, we propose a condenser pipe intelligent layout system based on machine learning. Summary of the invention

[0005] The purpose of the present invention is to provide a condenser tube intelligent layout system based on machine learning to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: a machine learning-based intelligent layout system for condenser tubes, comprising a condition input layer, a layout analysis layer, a data recording layer, and a result output layer;

[0007] The condition input layer includes a user interaction module, which is used to receive control conditions that affect concrete temperature changes input by the user and transmit these condition data to the layout analysis layer;

[0008] The layout analysis layer includes a machine learning module and a layout analysis module. The machine learning module is used to train a temperature change prediction model. The temperature change prediction model can predict the maximum temperature rise of concrete during the hardening process based on the condition data input by the user. The layout analysis module is responsible for receiving the condition data transmitted by the condition input layer, and calling the temperature change prediction model in the machine learning module according to these data to obtain the predicted maximum temperature value, and then sending the predicted maximum temperature value and the control condition input by the user to the data recording layer;

[0009] The data recording layer includes a storage module, and the storage module is used to store the condition data input by the user and the prediction results output by the temperature change prediction model;

[0010] The result output layer includes a display module, which can access the data stored in the storage module and display all the maximum temperature values ​​predicted based on specific conditions to the user in a visual form;

[0011] Among them, the user can dynamically adjust the input control conditions through the user interaction module to obtain different predicted maximum temperature values, and judge whether the predicted results meet the production requirements by comparing them with the maximum temperature value expected in actual production. When the predicted results meet the production requirements, the control conditions at this time are determined as the optimal layout parameters of the condenser.

[0012] Preferably, the user interaction module allows the user to input conditions including the diameter of the condenser tube, tube wall thickness, spacing, length, as well as the concrete volume, flow rate and starting temperature, wherein the flow rate represents the flow speed of the cooling material in the condenser tube, and the starting temperature represents the temperature of the concrete at the time of pouring.

[0013] Preferably, the steps of constructing and training the temperature change prediction model include:

[0014] S1: Data collection, collecting layout parameters of various combinations of condenser pipes during the concrete hardening process, and recording the maximum temperature rise data of concrete under each combination of layout parameters, wherein the layout parameters include the pipe diameter, pipe wall thickness, spacing, length of the condenser pipe, as well as the concrete volume, flow rate and starting temperature;

[0015] S2: Preprocess the collected data, including data cleaning, standardization and normalization, to eliminate the impact of outliers and dimensional differences on model training and form a regular data set;

[0016] S3: Divide the data set into training set, validation set and test set. The training set is used for model training, the validation set is used for model effect verification and optimization, and the test set is used to verify the generalization ability of the model on unknown data sets.

[0017] S4: Model selection, according to the nature of the temperature change prediction problem, select a suitable machine learning model, which should be able to predict the maximum value of temperature rise based on the input conditional data;

[0018] S5: Initialize model parameters, set initial parameters for the selected prediction model, and prepare for training;

[0019] S6: Model training: Use the training set to train the temperature change prediction model, and continuously adjust the model parameters through the optimization algorithm to minimize the gap between the predicted temperature and the actual temperature;

[0020] S7: Model validation and optimization: Use the validation set to validate the trained model and adjust the model parameters or structure based on the validation results to improve the accuracy of the prediction.

[0021] S8: Model evaluation: Use the test set to comprehensively evaluate the optimized model using the set evaluation indicators to ensure that the model performance meets the expected standards;

[0022] S9: Model saving and deployment: Save the trained and verified temperature change prediction model and deploy it to the machine learning module of the layout analysis layer for subsequent real-time prediction users to use, thereby achieving accurate prediction of the maximum temperature rise during the concrete hardening process.

[0023] Preferably, according to the nature of the temperature change prediction problem, a linear regression model is selected as the prediction model:

[0024] y=β0+β1χ1+β2χ2+...+β n χ n +∈

[0025] Among them, y represents the predicted maximum temperature, x1, x2, ..., x n represents the control conditions that affect the temperature change of concrete input by the user, such as the diameter, wall thickness, spacing, length of the condenser pipe, as well as the concrete volume, flow rate and starting temperature; β1, β2, ..., β n is the parameter of the model, which needs to be determined through training; β0 is the intercept term, that is, the expected value of y when all control terms x are 0; ∈ is the error term, that is, the difference between the actual observed value and the model predicted value; through the linear regression model, the maximum temperature generated by concrete hardening can be predicted based on the input conditional data, thereby assisting users in determining the optimal layout of the condenser pipe.

[0026] Preferably, when initializing the model parameters, the machine learning module is used to initialize the model parameters for each feature x in the linear regression model. i The corresponding parameter β iAssign an initial value, which is set to a random number close to 0, and also assign an initial value to the intercept term β0; these initial parameter values ​​will serve as the starting point for model training, and will be continuously optimized and adjusted through subsequent training processes until the goal of minimizing the gap between the predicted temperature and the actual temperature is achieved;

[0027] The machine learning module uses the data in the training set to continuously iteratively update the parameters of the linear regression model through a gradient descent algorithm to minimize the loss function value between the predicted temperature and the actual temperature; in each iteration, the value of the model parameter is adjusted according to the gradient information of the loss function to the model parameter until the loss function converges or reaches a preset number of iterations.

[0028] Preferably, the machine learning module uses the data in the test set to calculate the error index between the maximum temperature value predicted by the model and the actual maximum temperature value; the prediction accuracy, stability and generalization ability of the model are comprehensively evaluated through the error index, so as to determine the performance of the linear regression model in the prediction task.

[0029] Preferably, the error indicators include mean square error, mean absolute error and determination coefficient R 2 .

[0030] Preferably, the user interaction module provides intuitive and easy-to-use interactive operations through a graphical user interface; the user can input the layout parameters of the condenser through the interface, and the interface provides data verification and error prompt functions to ensure that the data entered by the user is legal and valid.

[0031] Preferably, the storage module uses relational database technology to implement data storage functions; wherein the conditional data input by the user and the prediction results of the temperature change prediction model are structured and stored in the form of data tables, each data table containing multiple fields, corresponding to various conditions input by the user and the prediction output of the model; at the same time, the transaction processing mechanism of the database is used to ensure the consistency and integrity of the data; the storage module also provides a data query interface, allowing other modules to retrieve and access the stored data according to needs.

[0032] Preferably, the display module uses Web front-end technology to realize the visualization display function of data; by calling the chart library, the user input condition data and the prediction results of the temperature change temperature prediction model stored in the storage module are displayed in the form of charts, including line charts, bar charts and scatter charts; at the same time, the display module also provides interactive functions, allowing users to select the displayed data range, chart type and custom chart style and configuration through interface operations to meet users' different data display needs.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The system uses a temperature change prediction model trained by a machine learning module to accurately predict the maximum temperature rise of concrete during the hardening process based on multiple control conditions input by the user. This prediction capability enables engineers to scientifically evaluate the layout effect of condenser pipes before construction, thereby avoiding the uncertainty of traditional methods that rely on experience and intuition for design.

[0035] By dynamically adjusting the control conditions input by the user and observing the temperature prediction values ​​output by the prediction model in real time, the user can intuitively understand the impact of different layout schemes on concrete temperature control. This interactive design method not only improves the design flexibility, but also helps engineers quickly find the best layout of condenser pipes that meet production needs, and realizes the intelligent optimization of condenser pipe layout, which not only improves the construction quality, but also effectively reduces the cost of subsequent maintenance and adjustment, bringing significant economic benefits to large-volume concrete construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is the overall structure diagram of the present invention;

[0037] Figure 2 Flowchart for building and training a temperature change prediction model using a linear regression model. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] See also Figure 1-2 ,The present invention provides a technical solution: a machine learning-based condenser tube intelligent layout system, comprising a condition input layer, a layout analysis layer, a data recording layer, and a result output layer;

[0040] The condition input layer includes a user interaction module, which is used to receive control conditions that affect concrete temperature changes input by the user and transmit these condition data to the layout analysis layer;

[0041] The layout analysis layer includes a machine learning module and a layout analysis module. The machine learning module is used to train a temperature change prediction model. The temperature change prediction model can predict the maximum temperature rise of concrete during the hardening process based on the condition data input by the user. The layout analysis module is responsible for receiving the condition data transmitted by the condition input layer, and calling the temperature change prediction model in the machine learning module according to these data to obtain the predicted maximum temperature value, and then sending the predicted maximum temperature value and the control condition input by the user to the data recording layer;

[0042] The data recording layer includes a storage module, and the storage module is used to store the condition data input by the user and the prediction results output by the temperature change prediction model;

[0043] The result output layer includes a display module, which can access the data stored in the storage module and display all the maximum temperature values ​​predicted based on specific conditions to the user in a visual form;

[0044] Among them, the user can dynamically adjust the input control conditions through the user interaction module to obtain different predicted maximum temperature values, and judge whether the predicted results meet the production requirements by comparing them with the maximum temperature value expected in actual production. When the predicted results meet the production requirements, the control conditions at this time are determined as the optimal layout parameters of the condenser.

[0045] As attached Figure 1 As shown in the figure, in the condition input layer, the user inputs the control conditions that affect the temperature change of concrete through the interactive module. The machine learning module in the layout analysis layer uses the pre-trained temperature change prediction model to predict the maximum temperature rise of concrete during the hardening process based on the condition data input by the user. This prediction result, together with the condition data input by the user, is transmitted to the data recording layer for storage; in the result output layer, the display module presents this data to the user in a visual way to help the user intuitively understand the temperature prediction under different control conditions. The user can dynamically adjust the input conditions according to the prediction results to obtain the prediction results that meet the production needs, and determine the best condenser layout method accordingly.

[0046] This system uses machine learning technology to quickly and accurately predict the impact of different layout conditions on concrete temperature, greatly improving design efficiency. This precise prediction capability helps users find the optimal condenser layout parameters, thereby ensuring that heat can be efficiently removed and improving design quality. This system determines the optimal layout during the design phase, effectively avoiding the cost of dismantling and reinstalling after installation, and reducing potential safety risks. At the same time, intuitive visualization and powerful user interaction functions allow users to flexibly adjust and optimize design solutions.

[0047] The present invention will be further described below in conjunction with Examples 1 to 3:

[0048] Embodiment 1:

[0049] The user interaction module allows the user to input conditions including the diameter of the condenser tube, tube wall thickness, spacing, length, as well as the concrete volume, flow rate and starting temperature, where the flow rate represents the flow speed of the cooling material in the condenser tube, and the starting temperature represents the temperature of the concrete at the time of pouring.

[0050] The above user input conditions will directly affect the cooling efficiency and layout design of the condenser:

[0051] The small diameter of the condenser tube directly affects the flow state of the fluid in the condenser tube and the heat exchange efficiency. A larger diameter can reduce the resistance of the fluid flow, allowing the fluid to pass through the condenser tube more smoothly, thereby improving the cooling efficiency. A large diameter will also increase material costs and space occupancy, which may not be economical in practical applications.

[0052] The wall thickness has an important influence on the thermal conductivity and mechanical strength of the condenser. A thinner wall can reduce thermal resistance, making it easier for heat to be transferred from the fluid inside the tube to the cooling medium outside the tube, thereby improving cooling efficiency. However, too thin a wall may reduce the mechanical strength and pressure resistance of the condenser, increasing the risk of leakage or rupture. When selecting the wall thickness, cooling efficiency and safety need to be considered comprehensively.

[0053] The size of the spacing between the condenser tubes is directly related to the cooling coverage and the efficiency of heat transfer. Smaller spacing means that the area between the condenser tubes is closer, which can disperse the cooling effect more evenly and ensure that all parts inside the concrete can be cooled in a timely and effective manner. However, too small a spacing will not only increase the number of condenser tubes, thereby increasing material costs and installation complexity, but may also cause thermal interference between adjacent condenser tubes, reducing the overall cooling efficiency; on the other hand, if the spacing between the condenser tubes is too large, although it will reduce the use and cost of the condenser tubes, it may cause some concrete areas to not be fully cooled, which in turn causes local overheating and increases the risk of temperature cracks in the concrete. In addition, large spacing may also cause heat to accumulate in the concrete, making it difficult to dissipate in time, affecting the overall cooling effect of the condenser tube.

[0054] The length of the condenser tube directly affects the surface area and time of heat exchange. A longer condenser tube provides a larger heat exchange area, allowing the fluid to have more time to exchange heat with the cooling medium, thereby improving the cooling efficiency. However, a condenser tube that is too long will also increase the residence time of the fluid in the tube, which may cause the fluid to flow out of the condenser tube before it is fully cooled. When designing, it is necessary to find the optimal condenser tube length to balance the cooling efficiency and fluid residence time.

[0055] A faster flow rate can speed up the flow of fluid in the condenser tube, allowing heat to be transferred to the tube wall faster and carried away by the cooling medium. Too high a flow rate may also cause the fluid to flow out of the condenser tube before it fully contacts the tube wall, thereby reducing the cooling efficiency. A balance point needs to be found when selecting the flow rate to ensure that the fluid stays in the condenser tube for a sufficient time for sufficient heat exchange.

[0056] The starting temperature indicates the temperature of the cooled material when it enters the condenser inlet. A higher starting temperature means that more cooling energy is required to cool the fluid to the target temperature, which may affect the cooling efficiency. At the same time, the starting temperature also determines the temperature difference between the inside and outside of the condenser. The greater the temperature difference, the stronger the driving force for heat exchange, which is conducive to the rapid transfer of heat. However, excessive temperature difference may also lead to increased thermal stress on the condenser, and put higher requirements on the material and service life of the condenser. When designing the condenser layout, it is necessary to comprehensively consider the impact of the starting temperature on the cooling efficiency and the safety of the condenser.

[0057] Combined with the above-mentioned influencing conditions, through the pre-trained model, the machine learning module can give the prediction result of the maximum temperature change of concrete in a very short time. The user can choose the best condenser pipe layout method for use in production installation according to the actual production constraints.

[0058] A specific implementation example is given below:

[0059] When a construction site was pouring a large concrete foundation, in order to ensure that the concrete would not crack due to excessive temperature during the hardening process, the engineer decided to use a temperature change prediction model to predict the maximum temperature that may be generated when the concrete hardens, and accurately design the layout of the condenser based on this predicted value.

[0060] Before pouring began, engineers entered the following key parameters through the user interaction module of the condenser tube intelligent layout system:

[0061] Concrete volume: According to the architectural design drawings, the volume of concrete poured this time is 600 cubic meters.

[0062] Starting temperature: Taking into account the environmental conditions of the construction site and the mix ratio of concrete, the starting temperature is set at 25 degrees Celsius.

[0063] Condenser tube diameter: In order to provide sufficient cooling capacity, engineers selected a condenser tube with an inner diameter of 8 cm.

[0064] Condenser pipe spacing: Based on the concrete pouring area and the cooling capacity of the condenser pipe, the spacing was initially set to 200 cm.

[0065] Condenser pipe length: According to the pouring depth and shape of concrete, the length of the condenser pipe is set to 15 meters.

[0066] Water flow rate: To ensure effective heat exchange, the water flow rate in the condenser is set at 1.5 meters per second.

[0067] After entering these parameters, the condenser intelligent layout analysis system starts the temperature change prediction model. Based on the input parameter conditions, the model begins to predict the maximum temperature value that the concrete may reach during the hardening process. Assume that the prediction results show that under the current condenser layout, the maximum temperature value during the hardening process of the concrete will reach 65 degrees Celsius, which exceeds the safety temperature threshold of 60 degrees Celsius set by the engineer.

[0068] In order to reduce the predicted maximum temperature value, the engineers decided to adjust the layout parameters of the condenser tubes. They first adjusted the spacing of the condenser tubes to 160 cm to increase the cooling coverage. Then they re-ran the prediction model and found that the maximum temperature value was reduced to 62 degrees Celsius. Although it was an improvement, it still did not reach the safety threshold. Therefore, the engineers further increased the length of the condenser tube to 18 meters and slightly increased the water flow rate to 1.7 meters per second to enhance the cooling effect. After these adjustments, the maximum temperature value shown by the prediction model finally dropped to 58 degrees Celsius, which was lower than the safety threshold and met the requirements of the engineers.

[0069] Through this precise condenser pipe layout design, engineers not only ensured the temperature control of the concrete during the hardening process, but also effectively avoided the risk of temperature cracks, thereby ensuring the quality and safety of the building foundation.

[0070] Embodiment 2:

[0071] As attached Figure 2 As shown, the steps of constructing and training the temperature change prediction model include:

[0072] S1: Data collection, collect the layout parameters of various combinations of condenser pipes during the concrete hardening process, and record the maximum temperature rise data of concrete under each layout parameter combination; in order to collect the data of condenser pipe layout parameters and concrete temperature rise, a series of experiments are required. In the experiment, the diameter of the condenser pipe (such as 6mm, 8mm, 10mm), pipe wall thickness (such as 1mm, 1.5mm, 2mm), spacing (such as 150cm, 200cm, 250cm), length (such as 10m, 15m, 20m), and the volume of concrete (such as 600m 3 、800m 3 , 1000m 3 ), flow rate (such as 0.5m / s, 1m / s, 1.5m / s) and starting temperature (such as 20℃, 25℃, 30℃). For each parameter combination, the temperature rise during the concrete hardening process is monitored and the maximum temperature rise is recorded.

[0073] S2: Preprocess the collected data, including data cleaning, standardization and normalization, to eliminate the impact of outliers and dimensional differences on model training and form a regular data set; the collected raw data contains some outliers, such as abnormally high or low temperature readings due to sensor failure. During the data preprocessing stage, these outliers are identified and cleaned, that is, they are replaced with reasonable values ​​or deleted. Due to the different dimensions of each parameter, the data also needs to be standardized and normalized to ensure that all parameters are in the same numerical range, thereby eliminating the impact of dimensional differences on model training.

[0074] S3: Divide the data set. The preprocessed data set is randomly divided into three parts: training set, validation set and test set. Among them, 70% of the data is used as a training set to train the temperature change prediction model; 15% of the data is used as a validation set to verify the performance of the model and adjust the parameters during the model training process; the remaining 15% of the data is used as a test set to evaluate its generalization ability after the model training is completed.

[0075] S4: Model selection,According to the nature of the temperature change prediction problem, the linear regression model is selected as the machine learning model, which can predict the maximum temperature change during concrete pouring based on the layout parameters of the condenser pipe and the conditions such as concrete volume, flow rate, and starting temperature. The linear regression model is simple and easy to interpret, and is suitable for exploring the linear relationship between each parameter and the maximum temperature.

[0076] S5: Initialize model parameters. In the linear regression model, the parameters that need to be initialized include the regression coefficient and the intercept. These parameters are randomly initialized before model training begins, or are set to reasonable initial values ​​based on prior knowledge.

[0077] S6: Model training: Use the training set to train the temperature change prediction model, and continuously adjust the model parameters through the optimization algorithm to minimize the gap between the predicted temperature and the actual temperature.

[0078] S7: Model verification and optimization. During the model training process, the model is verified using the validation set data. If the model performance on the validation set is found to be poor, such as a large prediction error, the model parameters will be adjusted, or the model structure will be changed (such as adding or reducing features, introducing regularization terms, etc.) to improve the accuracy of the prediction.

[0079] S8: Model evaluation: Use the test set to comprehensively evaluate the optimized model through the set evaluation indicators to ensure that the model performance meets the expected standards.

[0080] S9: Model saving and deployment. The trained and verified temperature change prediction model is saved as a file and deployed to the machine learning module of the layout analysis layer. In actual applications, users can input the layout parameters and condition data of the condenser through the interactive module, and the machine learning module will call the trained model for real-time prediction.

[0081] According to the characteristics of the temperature change prediction problem, the linear regression model is selected as the prediction model:

[0082] y=β0+β1χ1+β2χ2+...+β n χ n +∈

[0083] Among them, y represents the predicted maximum temperature, x1, x2, ..., x n represents the control conditions that affect the temperature change of concrete input by the user, such as the diameter, wall thickness, spacing, length of the condenser pipe, as well as the concrete volume, flow rate and starting temperature; β1, β2, ..., β n is a parameter of the model, which needs to be determined through training; β0 is the intercept term, that is, the expected value of y when all control terms x are 0; ∈ is the error term, that is, the difference between the actual observed value and the model predicted value; through the linear regression model, the maximum temperature of concrete hardening can be predicted based on the input conditional data, thereby assisting users in determining the optimal layout of the condenser pipe. Linear regression is a statistical predictive analysis used to estimate the relationship between two or more variables; in the temperature change prediction model, the goal is to predict the maximum temperature y of the concrete, which is based on multiple features (control conditions) x1, x2, ..., x n These control conditions are important factors affecting the maximum temperature of concrete. The model combines these factors with the corresponding coefficients β1, β2, ..., β n Multiply them and add an intercept term β0 to estimate the maximum temperature.

[0084] When initializing the model parameters, the machine learning module is used to initialize each feature x in the linear regression model. i The corresponding parameter β iAssign an initial value, which is set to a random number close to 0, and also assign an initial value to the intercept term β0; these initial parameter values ​​will serve as the starting point for model training, and will be continuously optimized and adjusted through subsequent training processes until the goal of minimizing the gap between the predicted temperature and the actual temperature is achieved; the machine learning module uses the data in the training set and continuously iterates and updates the parameters of the linear regression model through the gradient descent algorithm to minimize the loss function value between the predicted temperature and the actual temperature; in each iteration, the machine learning module adjusts the values ​​of these parameters according to the gradient information of the loss function to the current model parameters. The gradient information indicates how the model parameters should be changed in order to reduce the loss function value. This process will continue until the loss function converges, that is, the value of the loss function no longer decreases significantly, or the preset number of iterations is reached; in this way, the machine learning module can train a linear regression model that can accurately predict the maximum temperature value of concrete based on the input conditional data.

[0085] The machine learning module uses the data in the test set to calculate the error index between the maximum temperature value predicted by the model and the actual maximum temperature value; the error index includes but is not limited to the mean squared error (MSE), the mean absolute error (MAE) and the determination coefficient R 2 (Coefficient of Determination), which comprehensively evaluates the prediction accuracy, stability, and generalization ability of the model through error indicators, thereby determining the performance of the linear regression model in the prediction task.

[0086] The mean square error (MSE) calculates the average of the squares of the differences between the predicted values ​​and the actual values. This metric is more sensitive to large errors because the errors are squared, so large errors will be magnified. The smaller the MSE, the closer the model's predicted values ​​are to the actual values, and the higher the prediction accuracy.

[0087] The mean absolute error (MAE) directly calculates the average of the differences between the predicted value and the actual value, regardless of whether the error is positive or negative. Compared with MSE, MAE treats all errors equally and does not give different weights to the size of the error. Therefore, MAE reflects more of the average deviation of the model's prediction.

[0088] Coefficient of determination R 2 Indicates the degree of fit of the model prediction value. R 2 The value of is between 0 and 1. The closer the value is to 1, the better the model fit is, that is, the model can explain more data variation. 2It not only takes into account the difference between the predicted value and the actual value, but also takes into account the degree of variation of the data itself. Therefore, it is a relatively comprehensive evaluation indicator.

[0089] Through the comprehensive analysis of these error indicators, we can fully understand the performance of the linear regression model in the prediction task. These indicators can not only evaluate the prediction accuracy of the model, but also reflect the stability and generalization ability of the model. If the model performs well in these indicators, then it can be considered that the model is reliable and can be used in practical applications.

[0090] Embodiment 3:

[0091] The user interaction module provides intuitive and easy-to-use interactive operations through a graphical user interface; the user can input the layout parameters of the condenser through the interface, and the interface provides data verification and error prompt functions to ensure that the data entered by the user is legal and valid. When the data entered by the user does not meet the preset requirements of the system or contains potential errors, the system will immediately issue a corresponding error prompt through the interface, thereby guiding the user to make necessary data corrections, ensuring the correctness and validity of all input data, and reducing subsequent problems caused by erroneous data input.

[0092] The storage module uses relational database technology to implement data storage functions, such as selecting MySQL to create a database instance; the conditional data entered by the user and the prediction results of the temperature change prediction model are structured and stored in the form of data tables, each of which contains multiple fields, corresponding to various conditions entered by the user and the prediction output of the model; at the same time, the transaction processing mechanism of the database is used to ensure the consistency and integrity of the data; the storage module also provides a data query interface, allowing other modules to retrieve and access the stored data according to needs.

[0093] By writing database access code to implement the data insertion function, JDBC or ORM frameworks such as Hibernate or Entity Framework can be used to insert the conditional data and prediction results entered by the user into the corresponding data table. In this process, the database transaction processing mechanism is used to ensure data consistency and integrity. When inserting, updating or deleting data, a transaction will be started and multiple database operation statements will be executed. The transaction will only be committed when all operations are successfully completed. If any of the statements fail to execute, the transaction will be rolled back to ensure data integrity.

[0094] The storage module also provides a data query interface, allowing other modules to retrieve and access the stored data as needed. For example, you can create a REST Ful API interface, and other modules can use HTTP requests to obtain conditional data and prediction results for a specific user.

[0095] The display module uses Web front-end technology to realize the visualization display function of data, such as using HTML, CSS and JavaScript to build an intuitive and easy-to-operate user interface; by calling the chart library, the user input condition data and the prediction results of the temperature change prediction model saved in the storage module are displayed in the form of charts, including but not limited to line charts, bar charts and scatter charts; at the same time, the display module also provides interactive functions, allowing users to select the displayed data range, chart type and custom chart style and configuration through interface operations to meet users' different data display needs.

[0096] Powerful chart libraries such as Chart.js or D3.js can be called through JavaScript. These chart libraries can provide a variety of chart types and can flexibly meet various data display needs.

[0097] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0098] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A condenser tube intelligent layout system based on machine learning, characterized in that: It includes condition input layer, layout analysis layer, data recording layer and result output layer; The condition input layer includes a user interaction module, which is used to receive control conditions that affect concrete temperature changes input by the user and transmit these condition data to the layout analysis layer; The layout analysis layer includes a machine learning module and a layout analysis module. The machine learning module is used to train a temperature change prediction model. The temperature change prediction model can predict the maximum temperature rise of concrete during the hardening process based on the condition data input by the user. The layout analysis module is responsible for receiving the condition data transmitted by the condition input layer, and calling the temperature change prediction model in the machine learning module according to these data to obtain the predicted maximum temperature rise, and then sending the predicted maximum temperature and the control conditions input by the user to the data recording layer; The data recording layer includes a storage module, and the storage module is used to store the condition data input by the user and the prediction results output by the temperature change prediction model; The result output layer includes a display module, which can access the data stored in the storage module and display all the maximum temperature values ​​predicted based on specific conditions to the user in a visual form; The user can dynamically adjust the input control conditions through the user interaction module to obtain different predicted maximum temperature values, and compare them with the expected maximum temperature value of actual production to determine whether the predicted results meet the production requirements. When the predicted results meet the production requirements, the control conditions at this time are determined as the optimal layout parameters of the condenser pipe; The temperature change prediction model was constructed using a linear regression model; The control conditions received by the user interaction module include the diameter, wall thickness, spacing, length of the condenser pipe, as well as the concrete volume, flow rate and starting temperature.

2. The condenser tube intelligent layout system based on machine learning according to claim 1 is characterized in that: The flow rate represents the flow rate of the cooling material in the condenser pipe, and the starting temperature represents the temperature of the concrete at the time of pouring.

3. The machine learning-based condenser tube intelligent layout system according to claim 2, characterized in that: The steps of constructing and training the temperature change prediction model include: S1: Data collection, collecting layout parameters of various combinations of condenser pipes during the concrete hardening process, and recording the maximum temperature rise data of concrete under each combination of layout parameters, wherein the layout parameters include the diameter, wall thickness, spacing, length of the condenser pipes, as well as the concrete volume, flow rate and starting temperature; S2: Preprocess the collected data, including data cleaning, standardization and normalization, to eliminate the impact of outliers and dimensional differences on model training and form a regular data set; S3: Divide the data set into training set, validation set and test set. The training set is used for model training, the validation set is used for model effect verification and optimization, and the test set is used to verify the generalization ability of the model on unknown data sets. S4: Model selection, according to the nature of the temperature change prediction problem, select a suitable machine learning model, which should be able to predict the maximum value of temperature rise based on the input conditional data; S5: Initialize model parameters, set initial parameters for the selected prediction model, and prepare for training; S6: Model training: Use the training set to train the temperature change prediction model, and continuously adjust the model parameters through the optimization algorithm to minimize the gap between the predicted temperature and the actual temperature; S7: Model validation and optimization: Use the validation set to validate the trained model and adjust the model parameters or structure based on the validation results to improve the accuracy of the prediction. S8: Model evaluation: Use the test set to comprehensively evaluate the optimized model using the set evaluation indicators to ensure that the model performance meets the expected standards; S9: Model saving and deployment: Save the trained and verified temperature change prediction model and deploy it to the machine learning module of the layout analysis layer for subsequent real-time prediction users to use, thereby achieving accurate prediction of the maximum temperature rise during the concrete hardening process.

4. The machine learning-based condenser tube intelligent layout system according to claim 3, characterized in that: According to the nature of the temperature change prediction problem, the linear regression model is selected as the temperature change prediction model: y=β0+β1x1+β2x2+...+β n x n +∈ Among them, y represents the predicted maximum temperature, x1, x2, ..., x n represents the control conditions that affect the temperature change of concrete input by the user, such as the diameter, wall thickness, spacing, length of the condenser pipe, as well as the concrete volume, flow rate and starting temperature; β1, β2, ..., β n is the parameter of the model, which needs to be determined through training; β0 is the intercept term, that is, the expected value of y when all control terms x are 0; ∈ is the error term, that is, the difference between the actual observed value and the model predicted value; through the linear regression model, the maximum temperature generated by concrete hardening can be predicted based on the input conditional data, thereby assisting users in determining the optimal layout of the condenser pipe.

5. The machine learning-based condenser tube intelligent layout system according to claim 4, characterized in that: When initializing the model parameters, the machine learning module is used to initialize the model parameters for each feature x in the linear regression model. i The corresponding parameter β i Assign an initial value, which is set to a random number close to 0, and also assign an initial value to the intercept term β0; these initial parameter values ​​will serve as the starting point for model training, and will be continuously optimized and adjusted through subsequent training processes until the goal of minimizing the gap between the predicted temperature and the actual temperature is achieved; The machine learning module uses the data in the training set to continuously iteratively update the parameters of the linear regression model through a gradient descent algorithm to minimize the loss function value between the predicted temperature and the actual temperature; In each iteration, the values ​​of the model parameters are adjusted according to the gradient information of the loss function with respect to the model parameters until the loss function converges or the preset number of iterations is reached.

6. The machine learning-based condenser tube intelligent layout system according to claim 5, characterized in that: The machine learning module uses the data in the test set to calculate the error index between the maximum temperature value predicted by the model and the actual maximum temperature value; the prediction accuracy, stability and generalization ability of the model are comprehensively evaluated through the error index, so as to determine the performance of the linear regression model in the prediction task.

7. The machine learning-based condenser tube intelligent layout system according to claim 6, characterized in that: The error indicators include mean square error, mean absolute error and determination coefficient R 2 .

8. The machine learning-based condenser tube intelligent layout system according to claim 1, characterized in that: The user interaction module provides intuitive and easy-to-use interactive operations through a graphical user interface; the user can input the layout parameters of the condenser through the interface, and the interface provides data verification and error prompt functions to ensure that the data entered by the user is legal and valid.

9. The machine learning-based condenser tube intelligent layout system according to claim 1, characterized in that: The storage module uses relational database technology to realize data storage function; wherein, the condition data input by the user and the prediction results of the temperature change prediction model are structured and stored in the form of data tables, each data table contains multiple fields, corresponding to various conditions input by the user and the prediction output of the model; at the same time, the transaction processing mechanism of the database is used to ensure the consistency and integrity of the data; the storage module also provides a data query interface, allowing other modules to retrieve and access the stored data according to needs.

10. The machine learning-based condenser tube intelligent layout system according to claim 1, characterized in that: The display module uses Web front-end technology to realize the visualization display function of data; by calling the chart library, the user input condition data and the prediction results of the temperature change temperature prediction model saved in the storage module are displayed in the form of charts, including line charts, bar charts and scatter charts; at the same time, the display module also provides interactive functions, allowing users to select the displayed data range, chart type and custom chart style and configuration through interface operations to meet users' different data display needs.

Images