A method for determining the length of a bin in mass concrete construction by the skip method

Abstract

The application discloses a method for determining the length of a bin in a large-volume concrete skip method construction, adopts a big data analysis mode, combines actual conditions of a construction site, limits a plurality of indexes such as raw material composition of the large-volume concrete and overall specifications of the large-volume concrete, builds a database for preparing the large-volume concrete skip method construction, then analyzes and calculates by using a machine learning calculation model, finds out influence indexes, inputs related indexes of the large-volume concrete in a to-be-constructed project in a built graphical user interface, and adjusts the influence indexes according to an output result, so that the determination of the most suitable bin length is realized. The length determination method adopted by the application comprehensively analyzes actual properties of the large-volume concrete, the calculation result is accurate, the calculation process is low in cost, and the length determination method has important practical application value.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of concrete construction, and particularly relates to a method for determining the length of a bin in a skip-bin method for constructing mass concrete. BACKGROUND

[0002] Concrete is one of the most widely used materials in construction engineering. In some projects, it is often necessary to pour concrete with a very large volume, such as residential building foundation slabs, terminal floor slabs, long-span bridge pedestals, etc. Such concrete is referred to as mass concrete. Concrete releases a lot of heat during the hydration process, which in turn affects the stress distribution within the concrete, leading to cracking of the concrete. This phenomenon is particularly pronounced in mass concrete.

[0003] At the current stage, to improve the temperature rise cracking problem of mass concrete, a post-pouring strip is usually used to avoid the cracking problem. However, the use of post-pouring strip construction affects the overall construction, the process is relatively complex, and the effect is not ideal. Therefore, some scholars have proposed using the skip-bin method for constructing mass concrete. The skip-bin method does not require the use of post-pouring strips, but rather divides the mass concrete into multiple bins, synchronously pours the non-adjacent bins during construction, and then pours the concrete of adjacent bins after a certain period of curing. The principle of the skip-bin method is to create a time difference in the internal temperature release of mass concrete, thereby regulating the "resistance" and "release" effects of the internal stress of the concrete. In the skip-bin method construction, a good bin length can ensure the smooth pouring and construction of the concrete of each bin and make the performance of the concrete superior. However, when dividing the bin, the corresponding bin length calculation formula given in the Technical Specification for Skip-Bin Method for Super-Long Mass Concrete Structures is usually used, the actual plot characteristics are referred to, and the bin is divided by calculating the temperature cracking index.

[0004] However, there are still some problems in applying this method to divide the bin at the current stage:

[0005] Firstly, although the calculation of the temperature cracking index can ensure the accuracy of the bin division, it requires multiple tests for testing and analysis, a corresponding test site, the production of corresponding test blocks, and the acquisition of a large number of technical parameters, which consumes a large amount of manpower and resources.

[0006] Secondly, the existing division method does not consider the problem of the characteristics of the concrete raw materials that have the greatest impact on the cracking of mass concrete.

[0007] Thirdly, when the actual construction length of the bin in the existing project exceeds the length given in the specification, it is difficult to predict the demand for the most suitable bin length of mass concrete in the project under the specification at a low cost.

[0008] Therefore, it is of great significance to seek a bin length determination method that is convenient, fast, and highly implementable. SUMMARY

[0009] The purpose of the present application is to provide a method for determining the length of a bin in the construction of mass concrete by the skip bin method, to solve the problem that the existing method for determining the length of a bin in the skip bin method requires the calculation of a temperature cracking index and multiple test analyses, consumes a large amount of manpower and resources, and does not take into account the characteristics of the raw materials of the concrete, and to solve the technical problem of predicting the available length of a bin at a low cost.

[0010] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0011] A method for determining the length of a bin in the construction of mass concrete by the skip bin method, the determination steps being as follows:

[0012] Step one: collect the characteristic index data of existing mass concrete engineering plots, including the raw material performance index of the concrete of each plot, the overall specification index of the concrete of each plot, and the actual construction bin length of each plot;

[0013] Step two: remove the data in the non-adaptive range of the characteristic index in step one, and then establish a database of the characteristic index;

[0014] Step three: based on the data in the database, analyze using a machine learning calculation model that has been optimized in parameters to determine a final feasibility model; wherein the raw material performance index of the concrete of each plot and the overall specification index of the concrete of each plot are input parameters, and the actual construction bin length of each plot is an output parameter;

[0015] Step four: based on the final feasibility model, analyze the degree of influence of each input parameter on the output parameter using the SHAP method, and determine the characteristic index that has an influence as an influence index;

[0016] Step five: based on the final feasibility model, construct a graphical user interface, collect the characteristic index data of the mass concrete engineering plot to be constructed, and after inputting the data in the graphical user interface, output the predicted bin length L out ;

[0017] Step six: verify whether the predicted bin length L out meets the requirement of the maximum bin length L' of the mass concrete engineering plot to be constructed in the specification:

[0018] Case one: when L out ≤ L', the construction uses L out as the bin length;

[0019] Case two: when L out > L',

[0020] If the concrete mixing work steps have been completed, the construction uses L' as the bin length;

[0021] If the concrete mixing work steps have not been carried out, the numerical value of the influence index is adjusted within the adaptive interval, and the bin length L is output again on the graphical user interface new , adjust until L new ≤L', then the concrete is mixed with the adjusted characteristic index data, and the construction uses L new as the bin length.

[0022] The applicable concrete strength grade in the database ranges from 30MPa to 50MPa, the construction temperature condition ranges from 15℃ to 25℃, and the construction relative humidity condition ranges from 60% to 80%.

[0023] The raw material performance indicators of the concrete include water-cement ratio, cement content, fly ash proportion, mineral powder proportion, fine aggregate proportion, coarse aggregate proportion and water reducing agent content, and the overall specification indicators of the concrete include the volume and pouring thickness of the concrete.

[0024] The machine learning calculation model uses the extreme gradient boosting model;

[0025] The parameter optimization method of the extreme gradient boosting model uses the particle swarm optimization algorithm to obtain the objective function of the final feasible model; the graphical user interface is constructed based on the objective function, and the Tkinter module in Python software is used for construction.

[0026] The influence indexes are sorted in descending order of influence degree as follows: fly ash content, concrete volume, water reducing agent content, water-cement ratio, mineral powder content, cement content and pouring thickness.

[0027] In step six, a single influence index, i.e. the fly ash content with the largest influence degree, is adjusted. The fly ash content and the bin length have a positive correlation influence trend, i.e. the larger the fly ash content, the longer the bin length of the concrete.

[0028] In step six, two or more influence indexes are adjusted, and the adjustment order is in descending order of influence degree.

[0029] The adaptive interval of each characteristic index is as follows:

[0030] The water-cement ratio interval is 0.3-0.8, the cement content interval is 180m 3 -360m 3, the fly ash accounts for 10% to 60%, the mineral powder accounts for 0% to 40%, the ratio of the fine aggregate content and the cement content is 2 to 4.5, the ratio of the coarse aggregate content and the cement content is 3 to 6, the water reducing agent content is 0 to 12 kg / m3, the volume of the concrete is 5000m 3 ~ 15000m 3 , and the pouring thickness is 0.4m to 1.8m.

[0031] Compared with the prior art, the present application has the following characteristics and beneficial effects:

[0032] The present application discards the calculation idea of using temperature cracking index for bin length division in a single project, collects relevant data information of existing engineering examples of mass concrete skip bin method construction, extracts nine material performance indexes of concrete in each mass concrete skip bin method construction plot and corresponding bin length as characteristic indexes, and establishes an information database of skip bin method construction; at the same time, a machine learning calculation model optimized by parameters is used to analyze the data of the database to obtain a final feasibility model, the SHAP method is used based on the final feasibility model to analyze the influence degree of each characteristic index on the bin length, and the characteristic index that produces influence is determined as an influence index; then, based on the final feasibility model, a graphical user interface is established by using the Tkinter module in the Python algorithm. Finally, the characteristic indexes of the mass concrete in the project to be constructed are input into the graphical user interface, and the bin length can be output, and the most suitable bin length suitable for the engineering project can be obtained by adjusting each characteristic index.

[0033] The present application can avoid most of the tests involved in calculating the temperature cracking risk, significantly reduce the test cost, and fully consider the influence of cement, fly ash, mineral powder, aggregate and other factors on the bin length division when determining the bin length, so as to subsequently adjust the proportion according to the demand. The present application can realize the prediction of the most suitable bin length of mass concrete skip bin method construction according to different plot characteristics, assist construction, and promote the engineering structure to achieve the best quality effect.

[0034] The present application can comprehensively analyze the performance of related raw materials involved in the concrete and the concrete specification, give the most suitable bin length of the construction plot, ensure the best quality of the mass concrete, and the method is simple and low in cost, and has important engineering practical popularization value. BRIEF DESCRIPTION OF DRAWINGS

[0035] The present application will be further described in detail below with reference to the drawings.

[0036] Figure 1 is the correlation heat map between the characteristic indexes of the present application.

[0037] Figure 2 is a ranking diagram of the influence of each feature indicator of the present application on the prediction result.

[0038] Figure 3 is a diagram of the influence of a single feature indicator of the present application on the prediction result.

[0039] Figure 4 is a SHAP value development result of each feature indicator of the present application.

[0040] Figure 5 is a schematic diagram of a graphical user interface of embodiment one of the present application.

[0041] Figure 6 is a schematic diagram of a graphical user interface of embodiment two of the present application.

[0042] Figure 7 is a schematic diagram of a graphical user interface of embodiment three of the present application. Figure 1 .

[0043] Figure 8 is a schematic diagram of a graphical user interface of embodiment three of the present application. Figure 2 . DETAILED DESCRIPTION

[0044] At the current stage, machine learning methods can quickly extract target parameters through big data analysis, so as to achieve the purposes of performance prediction and optimization. Therefore, it is of great significance to use machine learning methods to study the division of bins in the construction of skip floor method for improving work efficiency and guiding construction.

[0045] Database construction: before machine learning, a qualified database needs to be established. First, the selection of indicators in the database: there are many factors that affect the construction of mass concrete skip method. In the division of skip method construction, the performance of concrete materials has an important influence. The influencing factors of the performance of concrete materials are mainly cementitious materials, water-cement ratio and aggregate, etc. Therefore, when selecting the indicators, the first characteristic indicators considered by the present application are water-cement ratio (w_c) and cement content (ccon). In mass concrete, fly ash is usually added as a cementitious material to reduce the hydration heat, and sometimes mineral powder is also added as a compensatory cementitious material for later strength. Therefore, fly ash content (fmh) and mineral powder content (kf) are also used as characteristic indicators. In concrete, according to the particle size difference, aggregate can be divided into fine aggregate and coarse aggregate, which are also important components of concrete. Therefore, the present application sets the mass ratio of fine aggregate to cement (X / c) and the mass ratio of coarse aggregate to cement (C / c) as characteristic indicators. Then, the admixture in the concrete is considered, and the present application only considers water reducing agent as an admixture, so the water reducing agent content (jsj) is set as a characteristic indicator. Finally, for mass concrete, thickness and volume are also important indicators. Therefore, the present application sets the thickness (h) and volume (tj) as characteristic indicators. At the same time, the present application also sets the actual skip length (L) in the skip method construction that meets the specification as a characteristic indicator.

[0046] Secondly, for the data in the database, in addition to the statistical characteristics, the correlation between various parameters also needs to be calculated. If there is a significant linear relationship between the parameter indicators, it will reduce the prediction accuracy of the machine learning model. The present application uses Pearson correlation coefficient as the basis for judgment to evaluate the relationship between the parameter indicators. The Pearson correlation coefficient is simply denoted as PCC, and the calculation formula of PCC is shown in formula (1). In the formula, i represents the data sample; N represents the number of data samples; y i and y i represent the true value and the predicted value, respectively; and represent the mean of the true value and the predicted value, respectively. According to the judgment principle of PCC, the P value of two characteristic indicators can be between -1 and 1. If the P value is closer to 0, it means that the correlation between the two characteristic indicators is weaker.

[0047]

[0048] Through the calculation and analysis of PCC value, the correlation heat map between the characteristic indicators is obtained, as shown in Figure 1 It can be seen that the P value between the characteristic indicators is mostly between -0.4 and 0.4, which indicates that the linear relationship between the characteristic indicators is weak, so the influence on the prediction effect of the machine learning model is small, which ensures the smooth construction of the model.

[0049] The present application adopts the current widely used extreme gradient boosting (XGB) model as the algorithm of ensemble learning. The ensemble learning model adopts decision tree as the base learner. When using the model for calculation and analysis, the objective function expression is

[0050] The objective function of the XGB model is represented by

[0051] L represents the loss function, y x represents the actual value of the xth sample, u x represents the xth input variable, Ω represents the regularization term, and M represents a constant. When predicting the bin length, parameter optimization of the machine learning model is needed. For the XGB model, the optimization method used is the particle swarm optimization method (PSO). Using this method, the main adjustment parameters corresponding to different databases include four, the initial control interval needs to be set: the eta value (learning rate of the machine learning model) is 0.02-0.8; the gamma value (minimum loss function decrease value required for leaf node to continue splitting in the machine learning model) is 0.03-0.9; the max_depth value (maximum depth of the decision tree in the machine learning model) is 2-36; and the n_estimators value (number of trained decision trees in the machine learning model) is 12-680.

[0052] Global feature index analysis: In order to explain the prediction results of the machine learning model, the present application uses Shapley additive explanation (SHAP). The SHAP method is based on the SHAP value of cooperative game theory as a quantitative evaluation index. The SHAP value of a feature is calculated by the sum of the influence of the combination of the feature and other features on the regression result, and the calculation method is shown in formula (2).

[0053]

[0054] wherein φj(f, x) represents the SHAP value corresponding to feature J in the machine learning model of type f, x represents the sample explained by the SHAP method. J represents the feature set, and S is the feature set not containing feature j.

[0055] The SHAP analysis method calculates the contribution of each feature index to the model prediction result by arranging and combining the feature values, thereby explaining the prediction result of the model.

[0056] Using the XGB model, the influence of each feature index on the prediction result is determined by SHAP analysis, which is the influence of the feature index on the bin length first, and the results are shown in Figure 2 and Figure 3 Figure 2 ​It can be seen that the most influential factor on the prediction result is the fly ash content, followed by the volume of mass concrete, the water reducing agent content, the water-cement ratio, the mineral powder content, the cement content and the thickness. The influence of coarse aggregate and fine aggregate can be ignored. For mass concrete, the main performance to be considered is the temperature effect affecting cracking. Fly ash has been proven to have a lower heat release during hydration, which can well regulate the release rate of cement hydration heat. Therefore, when dividing the length of the bin, the first consideration is the content of fly ash. The second influencing factor is the volume of mass concrete, because the thickness of mass concrete changes little during the construction of the bottom plate, and the volume index is mainly controlled by the length and width, so the overall length and width directly affect the length of the bin. Subsequently, the water reducing agent and water-cement ratio and other parameters, these indicators mainly affect the flow performance and strength of concrete and other indicators, and then affect the construction and maintenance of concrete foundation. The influence of aggregate ratio is small, because it does not participate in the hydration heat effect of cementitious materials in mass concrete, so it will not affect the length division.

[0057] Referring to Figure 3 For each characteristic index, its influence on the prediction result of the bin length increases with the SHAP value, which can be considered as positive correlation in the red part, and negative correlation in the blue part. As can be seen from the figure, except for individual differences, the relationship between the fly ash content (fmh) and the bin length is positively correlated, that is, the greater the fly ash content, the longer the bin length of the concrete.

[0058] Single feature index analysis: referring to Figure 4 The SHAP value development result of each characteristic index is shown in the figure. Each dot in the figure represents a sample data point corresponding to the characteristic index. From Figure 4 It can be seen that for each characteristic index, the distribution of SHAP value will show a certain trend with the change of the value, which is related to the division of bin length. Figure 4 (a) is the SHAP value of fly ash content (fmh) and the relationship diagram of bin length. Figure 4 (b) is the SHAP value of volume (tj) and the relationship diagram of bin length. Figure 4 (c) is the SHAP value of water reducing agent content (jsj) and the relationship diagram of bin length. Figure 4 (d) is the SHAP value of water-cement ratio (w_c) and the relationship diagram of bin length. Figure 4 (e) is the SHAP value of mineral powder content (kf) and the relationship diagram of bin length. Figure 4 (f) is the SHAP value of cement content (ccon) and the relationship diagram of bin length.Figure 4 (j) is a graph showing the relationship between the SHAP value of thickness (h) and cell length. Figure 4 (h) is a graph showing the relationship between the SHAP value of the mass ratio of coarse aggregate to cement (C / c) and the length of the storage cell. Figure 4 (i) is a graph showing the relationship between the SHAP value of the mass ratio of fine aggregate to cement (X / c) and the length of the storage cell.

[0059] For example, for Figure 4 In the graph of the relationship between the SHAP value of fly ash content (fmh) and the silo length in (a), it can be seen that, except for two points of difference around 15% fly ash content, the SHAP values ​​of the remaining data points show an approximately increasing trend with the increase of fly ash content. This indicates that the silo length tends to increase with the increase of fly ash content. The changing trends of other characteristic parameters that can affect silo length can be compared with the fly ash analysis. When the changing trends of characteristic indicators are not obvious or the rules are not clear, the amount of data in the database can be increased to make the model analysis more comprehensive and the model calculation more instructive.

[0060] The graphical user interface was built using the Tkinter module in Python to create a GUI for predicting the optimal cell length in the skip-cell construction method. See also... Figure 5- Figure 8 As shown, the GUI includes two modules: an input module and an output module. After inputting the feature index into the input module, clicking "Calculate" will generate the predicted cell length for the concrete.

[0061] Example 1:

[0062] Based on the construction of the exterior wall of a large-volume concrete basement in a building construction project, this invention determines the optimal cell length for the skip-cell method in this project. The project site covers an area of ​​6575m². 2 The concrete pouring thickness is 0.4m. The concrete strength is 35MPa.

[0063] The specific construction steps are as follows:

[0064] Step 1: Collect data on the characteristic indicators of existing large-volume concrete engineering sites. This generally involves collecting relevant data from existing literature on engineering examples of skip-pour construction. Characteristic indicators include the raw material performance indicators of concrete in each site, the overall specifications of concrete in each site, and the actual construction cell length of each site under the specified standards. The raw material performance indicators of concrete include water-cement ratio, cement content, fly ash ratio, mineral powder ratio, fine aggregate ratio, coarse aggregate ratio, and water-reducing agent content. The overall specifications of concrete include the volume and pouring thickness.

[0065] Step two, remove the data in the non-adaptive interval of the characteristic index in step one, and then establish a database of the characteristic index; the applicable concrete strength grade range in the database is 30MPa-50MPa, the construction temperature condition range is 15℃-25℃, and the construction relative humidity condition range is 60%-80%.

[0066] Step three, based on the data in the database, an extreme gradient boosting model is selected as the model for calculation and application, and then a particle swarm optimization method is used to optimize the parameters of the database, and at the same time, the best parameter values after particle swarm optimization are as follows: the eta value is 0.1; the gamma value is 0.6; the max_depth value is 12; and the n_estimators value is 240. Based on the determination of the optimal parameters, the extreme gradient model is verified and analyzed to determine the final feasibility model, and the objective function of the final feasibility model is obtained. Among them, the raw material performance index of each plot of concrete and the overall specification index of each plot of concrete are input parameters, and the actual construction bin length under the specification of each plot is an output parameter.

[0067] Step four, based on the final feasibility model, the SHAP method is used to analyze the influence degree of each input parameter on the output parameter, and the characteristic index that produces influence is determined as the influence index; the influence index is sorted in descending order of influence degree as follows: fly ash content, concrete volume, water reducing agent content, water-cement ratio, mineral powder content, cement content and pouring thickness.

[0068] Step five, based on the final feasibility model, a graphical user interface is constructed, and the characteristic index data of the mass concrete engineering plot to be constructed are collected. After inputting the data in the graphical user interface, the predicted bin length L out is output.

[0069] The detailed information of the mass concrete in construction in this embodiment mainly includes environmental conditions and characteristic indexes. The site temperature is 22℃, and the air relative humidity is 76%. The detailed information of the raw material components of the concrete is as follows: the water-cement ratio of the concrete is 0.42, the cement content is 224kg / m 3 , the fly ash content accounts for 41% of the cement, the mineral powder content is 0%, the fine aggregate accounts for 3.3% of the cement, the coarse aggregate accounts for 4.77% of the cement, and the water reducing agent content is 7.7kg / m 3 . The calculated volume of the mass concrete is 3945m 3 .

[0070] Step six, whether the predicted bin length L out meets the requirement of the maximum bin length L' of the mass concrete engineering plot to be constructed in the specification:

[0071] Case one: when Lout If ≤L', then construction shall adopt L. out As the length of the storage compartment;

[0072] Scenario 2: When L out >L',

[0073] If the concrete mixing process has been completed, then L' is used as the length of the storage compartment during construction.

[0074] If the concrete mixing process has not yet been carried out, adjust the values ​​of the influencing indicators within the adaptability range, and re-output the cell length L on the graphical user interface. new Adjust until output L new If the concrete composition is ≤L', then the adjusted characteristic index data will be used for mixing, and L' will be used during construction. new As the length of the storage compartment.

[0075] In step six, the single influencing index is adjusted, namely the fly ash content, which has the greatest impact. The fly ash content has a positive correlation with the length of the concrete compartments, meaning that the higher the fly ash content, the longer the concrete compartments.

[0076] In step six, adjust two or more influencing indicators in descending order of their impact.

[0077] The adaptability ranges for each characteristic index are as follows: water-cement ratio range is 0.3 to 0.8, and cement content range is 180 mg / m³. 3 ~360kg / m 3 The fly ash content ranges from 10% to 60%, the mineral powder content ranges from 0% to 40%, the ratio of fine aggregate to cement content ranges from 2 to 4.5, the ratio of coarse aggregate to cement content ranges from 3 to 6, and the water-reducing agent content ranges from 0 to 12 kg / m³. 3 The volume range of the concrete is 5000m³. 3 ~15000m 3 The pouring thickness ranges from 0.4m to 1.8m.

[0078] This embodiment refers to Figure 5 As shown, by substituting the feature indicators into the input parameter section of the graphical user interface and clicking the prediction button, the output cell length L is obtained. out =46.55m, which, compared with the recommended value L'=50m for the maximum compartment length of large-volume concrete in wall structures in the "Technical Specification for Skip-Pour Method of Ultra-Long and Large-Volume Concrete Structures" (DB11T 1200-2023), meets the maximum limit specified in the specification, i.e., L out <L', therefore the most suitable cell length for this project is 46.55m.

[0079] Therefore, as long as the length of the bin is less than or equal to the number when the bin is divided, the concrete in each bin can exhibit good performance and no cracking phenomenon occurs. In addition, the crack conditions of the mass concrete during the maintenance period and the post-maintenance stage are monitored, and it is found that the concrete in each bin is in good condition and has no adverse conditions such as cracking during the above period.

[0080] Example two:

[0081] Based on the construction of the mass concrete foundation slab of a housing project, the most suitable bin length for the skip bin construction of the project is determined by the present application. The project engineering land area is 8560m 2 , and the pouring thickness of the concrete is 0.6m. The strength of the concrete is 40MPa.

[0082] Different from example one, the optimal parameter values in the model after particle swarm optimization are as follows: the eta value is 0.2; the gamma value is 0.7; the max_depth value is 6; and the n_estimators value is 148.

[0083] The detailed information of the mass concrete during construction in this embodiment mainly includes environmental conditions and characteristic indexes. The site temperature is 17℃, and the air relative humidity is 68%. The detailed information of the raw material components of the concrete is as follows: the water-cement ratio of the concrete is 0.46, the cement content is 185kg / m 3 , the fly ash content accounts for 35% of the cement, the mineral powder content is 10%, the fine aggregate accounts for 3.1% of the cement, the coarse aggregate accounts for 4.5% of the cement, and the water reducing agent content is 8.4kg / m 3 . The volume of the mass concrete is calculated to be 5260m 3 .

[0084] Referring to Figure 6 , the above characteristic indexes are substituted into the input parameter part of the graphical user interface, and the output bin length L out =42.74m is obtained by clicking the prediction key. Compared with the recommended value L' = 40m of the maximum bin length of the mass concrete of the foundation slab in the “DB11T 1200-2023 Technical Specification for Skip Bin Method of Ultra-long Mass Concrete Structure”, it exceeds the maximum length of 40m specified in the specification. In this embodiment, the concrete deployment work steps have been completed at this time, so the input indexes are not adjusted, and L' = 40m can be used as the bin length for construction.

[0085] Example three:

[0086] Based on the mass concrete construction of a certain infrastructure construction project, the most suitable bin length of the project skip construction method is determined by the method. The project land area is 19600m 2 , the pouring thickness of the concrete is 0.8m. The strength of the concrete is 45MPa.

[0087] Different from example one and example two, the optimal parameter values of the model after particle swarm optimization are as follows: the eta value is 0.4; the gamma value is 0.4; the max_depth value is 22; and the n_estimators value is 425.

[0088] The detailed information of the mass concrete in construction in the embodiment mainly includes environmental conditions and characteristic indexes. The site temperature is 24℃, and the air relative humidity is 80%. The detailed information of the raw material components of the concrete is as follows: the water-cement ratio of the concrete is 0.36, the cement content is 232kg / m 3 , the fly ash content accounts for 48% of the cement, the mineral powder content is 15%, the fine aggregate accounts for 3.05% of the cement, the coarse aggregate accounts for 4.8% of the cement, and the water reducing agent content is 10.4kg / m 3 . The volume of the mass concrete is calculated to be 13460m 3 .

[0089] Referring to Figure 7 , the above characteristic indexes are substituted into the input parameter part of the graphical user interface, and the output bin length L out =41.15m is obtained by clicking the prediction key. Compared with the recommended value L' =40m of the maximum bin length of the foundation mass concrete in the technical specification for skip construction of super-long mass concrete structure DB11T 1200-2023, it does not meet the maximum value limit specified in the specification.

[0090] In this embodiment, the concrete deployment work steps have not been carried out at this time, so the values of the influence indexes are adjusted within the adaptive interval, and the bin length L new is output again on the graphical user interface. The adjustment is carried out until L new ≤L', then the concrete is deployed with the adjusted characteristic index data, and the construction adopts L new as the bin length.

[0091] According to the actual engineering requirements, the concrete proportioning adjustment can be carried out in combination with the mixing station work in this embodiment, and the fly ash content which has the greatest influence is adjusted, as shown in Figure 8 , when the fly ash content is adjusted to 32%, the bin length L new= 38.46m < L' = 40m, meet the maximum length range in the specification. Therefore, the actual bin length value of the project is adjusted to 38.46m, which is the most appropriate bin length, and the concrete is adjusted with the adjusted characteristic index data.

Claims

1. A method for determining the length of a bay for construction of mass concrete by the jump bay method, characterized by, The determining step is as follows: Step one, collect the characteristic index data of the existing mass concrete engineering plots, the characteristic indexes including the raw material performance indexes of the concrete of each plot, the overall specification indexes of the concrete of each plot, and the actual construction bin length of each plot; Step two, remove the data in the non-adaptive interval of the characteristic indexes in step one, and then establish a database of the characteristic indexes; Step three, based on the data in the database, analyze by using a machine learning calculation model after parameter optimization to determine the final feasibility model; wherein the raw material performance indexes of the concrete of each plot and the overall specification indexes of the concrete of each plot are input parameters, and the actual construction bin length of each plot is an output parameter; Step four, based on the final feasibility model, analyze the influence degree of each input parameter on the output parameter by using the SHAP method, and determine the characteristic indexes that have an impact as impact indexes; Step five, based on the final feasibility model to build a graphical user interface, collect the characteristic index data of the mass concrete engineering plot to be constructed, and input the data in the graphical user interface to output the predicted bin length L out ; Step six, verify the predicted bin length L out Whether the maximum bin length L' of the mass concrete project site in the specification is met: Case 1: When L out ≤ L', then the construction adopts L out as the bin length; Case 2: When L out > L', If the concrete deployment work step has been completed, the construction adopts L' as the bin length; If the concrete batching procedure has not been performed, adjust the values of the influencing parameters within the adaptation interval and output the resulting bin length L on the graphical user interface new , adjust until L new ≤ L', then the concrete is batched with the adjusted characteristic parameter data and the construction uses L new as the bin length.

2. The method for determining the length of a bin in a mass concrete jump bin method construction according to claim 1, characterized in that: The applicable concrete strength grade range in the database is 30MPa-50MPa, the construction temperature condition range is 15℃-25℃, and the construction relative humidity condition range is 60%-80%.

3. The method for determining the length of a bin in the mass concrete jump bin method construction according to claim 1, characterized in that: The raw material performance indexes of the concrete include water-cement ratio, cement content, fly ash proportion, mineral powder proportion, fine aggregate proportion, coarse aggregate proportion, and water reducing agent content, and the overall specification indexes of the concrete include the volume and pouring thickness of the concrete.

4. The bin length determination method for mass concrete skip bin method construction according to claim 1, characterized in that: The machine learning calculation model uses an extreme gradient boosting model; The parameter optimization method of the extreme gradient boosting model uses a particle swarm optimization algorithm to obtain the objective function of the final feasibility model; Based on the objective function, a graphical user interface is constructed, which is constructed by using the Tkinter module in Python software.

5. The method for determining the length of a compartment for the placement of mass concrete according to claim 3 or 4, characterized in that: The impact indexes are sorted in descending order of influence degree as follows: fly ash content, concrete volume, water reducing agent content, water-cement ratio, mineral powder content, cement content, and pouring thickness.

6. The method for determining the length of a bin in a mass concrete jump-form construction according to claim 5, characterized in that: In step six, a single impact index, i.e. the fly ash content with the largest influence degree, is adjusted. The fly ash content and the bin length have a positive correlation in the influence trend, i.e. the larger the fly ash content, the longer the bin length of the concrete.

7. The method for determining the length of a bin in a mass concrete jump-form construction according to claim 5, characterized in that: In step six, two or more impact indexes are adjusted, and the adjustment order is in descending order of influence degree.

8. The method of determining the length of a compartment for the construction of mass concrete by the jump-forming method according to claim 6 or 7, characterized in that: The adaptive interval of each characteristic index is as follows: The water-cement ratio interval is 0.3-0.8, the cement content interval is 180m 3 -360m 3 , the fly ash proportion interval is 10%-60%, the mineral powder proportion interval is 0%-40%, the ratio interval of fine aggregate content and cement content is 2-4.5, the ratio interval of coarse aggregate content and cement content is 3-6, the water reducing agent content interval is 0-12kg / m 3 , the concrete volume interval is 5000m 3 -15000m 3 , and the pouring thickness interval is 0.4m-1.8m.

Images