Early strength prediction and regulation method and system for solid waste concrete based on transfer learning and digital twinning

Abstract

The application discloses a solid waste concrete early strength prediction and regulation method and system based on transfer learning and digital twinning, belongs to the technical field of building material detection, and comprises the following steps: S10, a mixed data set about the space and strength of a solid waste concrete sample is constructed, the space is divided, including being divided into diagonal spaces, the strength of the solid waste concrete sample is acquired in the diagonal spaces, the strength acquisition steps are as follows: the end point of the solid waste concrete is selected, the center of the solid waste concrete is taken as the strength acquisition path from the end point, the length of the acquisition path is measured, the length is divided into three path segments of the same length, the strength is acquired in each path segment, and the solid waste concrete sample is subjected to pressure test at the middle position of each path segment. Through in-depth analysis of the correlation between the material dosage and the strength, the application realizes accurate prediction and flexible regulation of the concrete strength.

Description

Technical Field

[0001] This invention relates to the field of building material testing technology, and in particular to a method and system for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins. Background Technology

[0002] With the increasing emphasis on sustainable development in the construction industry, the application of solid waste concrete (such as phosphogypsum concrete, steel fiber concrete, steel slag, recycled brick aggregate, coal gangue, and iron tailings) is becoming increasingly widespread. However, the accuracy of early strength prediction for solid waste concrete directly affects the feasibility and safety of its engineering applications. Therefore, it is necessary to predict the early strength of solid waste concrete to ensure its continued feasibility and safety in engineering applications.

[0003] Regarding this research, application CN202511083154.6 provides a method for predicting the fatigue life of solid waste fiber reinforced concrete. The technical solution includes the following steps: Step 1: Prepare m concrete samples using solid waste, and perform alkali activation on the concrete samples to calculate their physical properties; Step 2: Establish a testing environment, place the concrete samples in the testing environment, and calculate the properties of the concrete samples according to the specified time intervals, including n indicators, forming a data set; Step 3: Use the data set as calculation data, perform data standardization, calculate the information content, and input it into the constructed concrete life prediction model. This technical solution predicts the strength of concrete by simulating changes in indicators under a simulated environment, ultimately achieving the prediction of concrete life.

[0004] Another application, CN202510610883.6, provides a transfer learning optimization system and method for predicting the early-age strength of concrete. This technical solution includes multi-scale data perception, a deep neural network constrained by physical information, adaptive transfer learning of the formula, Bayesian optimization prediction, and federated learning feedback modules, realizing the entire process from data acquisition to model optimization. Through this system, the prediction errors of concrete strength at very early age and standard age are reduced to ±5% and ±3%, respectively. At the same time, the number of concrete specimens used in the experiment is reduced by 85%, saving material and labor costs. This method not only improves the prediction accuracy, but also continuously optimizes the prediction model through continuous learning and feedback, providing a prediction solution for practical engineering.

[0005] However, the above-mentioned technical solutions are mostly based on the assumption of linear superposition of components and lack the ability to analyze nonlinear coupling effects (such as those between solid waste, cement and aggregate). This leads to underfitting problems under small sample conditions, resulting in insufficient accuracy in predicting the early strength of solid waste concrete. Summary of the Invention

[0006] In view of the problems existing in the field of building material testing technology, the present invention is proposed.

[0007] Therefore, one of the objectives of this invention is to provide a method and system for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins. By constructing a hybrid spatial and strength dataset, it deeply analyzes the correlation between material usage and strength, thereby achieving accurate prediction and flexible control of concrete strength. At the same time, it can also promote the optimization and standardization of material proportions, provide scientific decision support for concrete production, and help reduce the consumption of natural resources and waste emissions.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] On the one hand, this invention provides a method for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins, including the following steps:

[0010] S10: Construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include:

[0011] The space is divided, including dividing it into diagonal spaces;

[0012] The strength of the solid waste concrete sample is obtained in the diagonal space, and the steps are as follows:

[0013] Select the endpoint of the solid waste concrete, and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path;

[0014] Measure the length of the acquired path and divide the length into three path segments of equal length;

[0015] Strength is obtained at each of the aforementioned path segments, including pressure testing of solid waste concrete samples at the midpoint of each path segment.

[0016] The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained;

[0017] S20: Based on the maximum pressure, obtain relevant data, including the materials and quantities corresponding to the manufacture of the solid waste concrete sample, the materials including solid waste, cement and aggregate; and analyze the correlation effect on the maximum pressure based on the changes in the quantities of materials corresponding to the manufacture of the solid waste concrete sample.

[0018] S30: Perform relevant processing based on the aforementioned correlation effects, the relevant processing including:

[0019] The dosage is differentiated to obtain the maximum pressure corresponding to different dosages;

[0020] Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on the different dosages;

[0021] The control is based on the pressure difference, and the control is divided into central control and outer central control. The central control is based on the I path segment, and the maximum pressure on the solid waste concrete sample is set at a future time according to the pressure difference, and the amount of material is adjusted based on the maximum pressure.

[0022] S40: The two path segments other than path segment I are marked as path segment II and path segment III. Path segment II is the path segment between path segment I and path segment III. The external center control is based on path segment III. The control is performed after relevant comparison.

[0023] In a preferred embodiment of the present invention, in step S20, the correlation between the change in the amount of material used in manufacturing the solid waste concrete sample and the maximum pressure is analyzed and calculated according to the following formula:

[0024] ;

[0025] In the formula, Indicates maximum pressure;

[0026] Indicates the amount of solid waste added;

[0027] Indicates the amount of cement used;

[0028] Indicates the amount of aggregate used;

[0029] , , , Represents the regression coefficient;

[0030] This indicates the error term.

[0031] In a preferred embodiment of the present invention, the following method is further included: [The following formula is used to calculate the result].

[0032] ;

[0033] In the formula, express The quadratic term.

[0034] In a preferred embodiment of the present invention, a sensitivity analysis is performed based on the results of the analysis and calculation, wherein the sensitivity analysis is based on the dosage, as shown below:

[0035] ;

[0036] ;

[0037] In the formula, Representing variables The first-order sensitivity index;

[0038] Representing variables and The second-order sensitivity index;

[0039] Indicates variance;

[0040] Indicates a prospective outlook.

[0041] In a preferred embodiment of the present invention, information is acquired based on the calculation results. The acquired information includes the amount of material that is most correlated with causing a certain maximum pressure, based on different maximum pressures. This amount of material is marked as the reference material amount. When the solid waste concrete sample needs to reach the corresponding maximum pressure in the future, the amount of this material is first proportioned.

[0042] In a preferred embodiment of the present invention, in step S40, adjustment is performed after correlation comparison, wherein the correlation comparison step includes:

[0043] Based on the relevant data obtained from the maximum pressure, the maximum pressure that path segment III can withstand is determined.

[0044] Obtain the pressure difference between the maximum pressure that path segment III can withstand and the maximum pressure that path segment I can withstand; mark the pressure difference as the control pressure difference;

[0045] Obtain the relevant factors that cause this pressure difference, including calculating the amount of material corresponding to the maximum pressure that path segment III can withstand based on the material and amount corresponding to the maximum pressure that path segment I can withstand, and marking this amount as the reference amount;

[0046] When the solid waste concrete sample needs to reach the maximum pressure that the III path segment can withstand in the future, the amount of material should be proportioned based on the reference dosage.

[0047] In a preferred embodiment of the present invention, the amount of material corresponding to the maximum pressure that path segment III can withstand is calculated based on the material and amount corresponding to the maximum pressure that path segment I can withstand, and is obtained according to the following formula:

[0048] ;

[0049] ;

[0050] In the formula, Indicates the maximum pressure that path segment I can withstand;

[0051] Indicates and The corresponding amount of materials used;

[0052] This indicates the maximum pressure that path segment III can withstand;

[0053] Indicates and The corresponding amount of materials used;

[0054] The slope of the linear relationship represents the degree to which changes in material usage affect the maximum pressure.

[0055] The intercept represents the linear relationship, and the base pressure value is represented when the amount of material used is zero.

[0056] In a preferred embodiment of the present invention, the distance between the center of path segment I and the center of path segment III is obtained in the solid waste concrete sample. When the solid waste concrete sample needs to reach the same pressure difference as the control pressure difference in the future, the same amount of material as path segment I and path segment III is used for mixing.

[0057] On the other hand, the present invention provides a system for application to the method for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins as described above, comprising:

[0058] The data construction module is used to construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include:

[0059] The space is divided, including dividing it into diagonal spaces;

[0060] The strength of the solid waste concrete sample is obtained in the diagonal space, and the steps are as follows:

[0061] Select the endpoint of the solid waste concrete, and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path;

[0062] Measure the length of the acquired path and divide the length into three path segments of equal length;

[0063] Strength is obtained at each of the aforementioned path segments, including pressure testing of solid waste concrete samples at the midpoint of each path segment.

[0064] The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained;

[0065] The data analysis module is used to acquire relevant data based on the maximum pressure, including the materials and quantities corresponding to the manufacture of the solid waste concrete sample, the materials including solid waste, cement and aggregate; and to analyze the correlation effect of changes in the quantities of materials corresponding to the manufacture of the solid waste concrete sample on the maximum pressure.

[0066] A data fusion processing module, comprising a processing unit and a control unit;

[0067] The processing unit is used to perform correlation processing based on the correlation effect, and the correlation processing includes:

[0068] The dosage is differentiated to obtain the maximum pressure corresponding to different dosages;

[0069] Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on the different dosages;

[0070] The control unit is used to control the pressure difference, and the control is divided into central control and outer central control. The central control is based on the I path segment, and sets the maximum pressure on the solid waste concrete sample at a future time according to the pressure difference, and adjusts the amount of material based on the maximum pressure.

[0071] Beneficial effects:

[0072] 1. By constructing a hybrid dataset of spatial and strength data, especially by subdividing the diagonal space into three equal-length path segments for strength testing, the distribution of internal strength of solid waste concrete can be captured more accurately. This meticulous data collection method can improve the accuracy and reliability of the data, providing a foundation for subsequent strength prediction and control.

[0073] 2. By analyzing the correlation between changes in the amount of solid waste, cement, and aggregates and the maximum pressure, the specific contribution of each material to the concrete strength was revealed. This analysis helps to understand the formation mechanism of concrete strength and provides a scientific basis for optimizing material proportions.

[0074] 3. Furthermore, based on the analysis of maximum pressure and related data, it is possible to predict the maximum pressure of concrete under different material proportions, and through the central control and external central control strategies, adjust the material dosage according to actual needs to achieve the expected strength, thereby helping to improve the quality and performance stability of concrete.

[0075] 4. By distinguishing between central control and external central control, and conducting control based on path segment I and path segment III respectively, flexible adjustments can be made according to the intensity requirements of different locations. At the same time, through relevant comparison steps, the pressure difference between different path segments and its related factors can be accurately calculated, providing more comprehensive and accurate information support for control.

[0076] 5. Through sensitivity analysis and information acquisition, the amount of material most correlated with causing a certain maximum pressure (reference material amount) can be determined, and the amount of this material can be given priority in the mix design at future times. This can promote the optimization and standardization of material mix design and improve the efficiency and consistency of concrete production. Attached Figure Description

[0077] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the modular structure of the early strength prediction and control system for solid waste concrete based on transfer learning and digital twins, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the method flow for step S10 in an embodiment of the present invention;

[0078] The diagram is labeled as follows: 110 - Data construction module; 120 - Data analysis module; 130 - Data fusion and processing module; 1301 - Processing unit; 1302 - Control unit. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0080] Because existing technologies are mostly based on the assumption of linear superposition of components, they lack the ability to analyze nonlinear coupling effects (such as those between solid waste, cement, and aggregates), which can easily lead to underfitting problems under small sample conditions, resulting in insufficient accuracy in predicting the early strength of solid waste concrete.

[0081] Based on this, the present invention proposes a method and system for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins. By constructing a hybrid spatial and strength dataset, it deeply analyzes the correlation between material usage and strength, so as to achieve accurate prediction and flexible control of concrete strength. At the same time, it can also promote the optimization and standardization of material proportions, and provide scientific decision support for concrete production.

[0082] The present solution will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0083] Reference Figures 1 to 3 This is one embodiment of the present invention, which provides a method for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins, including the following steps:

[0084] S10: Construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include:

[0085] Divide the space, including dividing it into diagonal spaces;

[0086] The strength of solid waste concrete samples was obtained in the diagonal space, and the steps are as follows:

[0087] Select the endpoint of the solid waste concrete (the endpoint is selected at the edge of the solid waste concrete sample), and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path (this acquisition path is a straight path, and the acquisition path is from the endpoint to the center of the solid waste concrete, which is a diagonal line, that is, the endpoint to the center is a straight diagonal line).

[0088] Measure the length of the obtained path and divide the length into 3 path segments of equal length;

[0089] Strength was obtained at each path segment, including pressure testing of solid waste concrete samples at the middle position of each path segment.

[0090] The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained.

[0091] It should be noted that, in this embodiment, the space of the solid waste concrete sample is divided into a diagonal space, and the path from the endpoint to the center is selected as the strength acquisition path.

[0092] The path was divided into three equal-length segments, and stress tests were conducted at the middle of each segment to obtain intensity data at different locations.

[0093] The path segment closest to the center is marked as path segment I, and its maximum pressure capacity is recorded.

[0094] This method of subdividing the space and testing at multiple points can more accurately capture the spatial distribution of the internal strength of solid waste concrete samples, thereby improving the accuracy and reliability of the data.

[0095] This multi-point testing method allows for a more comprehensive assessment of the internal structure of concrete, helping to understand the underlying mechanisms of strength changes.

[0096] S20: Obtain relevant data based on the maximum pressure, including the materials and quantities corresponding to the manufacture of solid waste concrete samples, including solid waste, cement, and aggregates; and analyze the correlation between the changes in the quantities of materials corresponding to the manufacture of solid waste concrete samples and the maximum pressure (in a feasible implementation, the relationship between variables can be quantified through a mathematical model, for example, by performing analysis and calculation through relevant formulas, which will be illustrated and explained later).

[0097] In reality, when preparing solid waste concrete samples, the amount of solid waste, cement, and aggregate used can affect the sample's strength. For example:

[0098] Steel slag, due to its high hardness, can improve the compressive strength of concrete when added to it. When the aggregate-to-ash ratio is 3.0, the compressive strength of steel slag aggregate concrete can reach 53.4 MPa, but when the aggregate-to-ash ratio increases to 6.0, the strength drops sharply by 60%.

[0099] Recycled brick aggregate has low strength; with an aggregate-to-ash ratio of 3.0, the compressive strength is only 16.7 MPa, and with an aggregate-to-ash ratio of 6.0, the strength decreases by 48%.

[0100] Coal gangue, with a low admixture content (10%), has a slightly improved strength, but a high admixture content will lead to a decrease in strength;

[0101] Iron tailings have mechanical properties similar to ordinary concrete, but attention should be paid to their strength and crushing value.

[0102] Furthermore, increasing the amount of solid waste typically leads to a decrease in strength. For example, when the replacement rate of recycled coarse aggregate increases from 0 to 22.5%, the compressive strength of concrete decreases by 0.7 MPa, 2.4 MPa, and 4.5 MPa, respectively. This is mainly because solid waste has many surface defects and high water absorption, which weakens the bond between the aggregate and the cement paste.

[0103] As a cementitious material, increasing the amount of cement used will increase the amount of hydration products, thereby enhancing the strength of concrete. For example, when the amount of cement increases, the water-cement ratio decreases, the density of concrete increases, and the compressive strength rises accordingly. However, excessive cement can lead to the accumulation of heat of hydration, causing internal cracks and actually reducing long-term strength.

[0104] For aggregates, their own strength is fundamental to the strength of concrete. For example, concrete made with high-strength lithological aggregates (such as basalt) has higher strength. As the aggregate particle size increases, the total porosity increases, and the strength decreases. For instance, concrete with 5mm–8mm aggregate has higher strength than concrete with 10mm–13mm aggregate.

[0105] Furthermore, aggregates with rough surfaces (such as crushed stone) have stronger adhesion to cement paste and higher strength than aggregates with smooth surfaces (such as pebbles). Well-graded aggregates with an appropriate sand ratio can form a dense skeleton, increasing strength. For example, poor gradation can lead to increased porosity and decreased strength in concrete.

[0106] Therefore, obtaining relevant data based on maximum pressure is of practical significance.

[0107] The analysis of the correlation between the changes in the amount of materials used in the manufacture of solid waste concrete samples and the maximum pressure has multi-dimensional significance, covering aspects such as technology optimization, cost control, environmental benefits, standard setting, and long-term performance evaluation.

[0108] In addition, by analyzing the correlation between changes in the amount of solid waste, cement, and aggregates on strength, the synergistic or antagonistic effects between different materials can be revealed.

[0109] For example, it was found that a certain type of solid waste (such as steel slag) can enhance strength when added at low dosage, but when added at high dosage, the porosity increases due to excessive water absorption, which in turn reduces strength.

[0110] Determine the optimal ratio of cement dosage to solid waste admixture to ensure that hydration products fully fill the surface defects of solid waste and improve interfacial adhesion.

[0111] Furthermore, based on the quantitative relationship between variations in dosage and strength, mathematical models (such as regression analysis and machine learning models) can be established to predict the strength of concrete under different mix proportions.

[0112] For example, by fitting experimental data to derive the functional relationship between strength and the ratio of bone ash to cement usage, a rapid mix design tool can be provided for engineering applications. This allows for the optimization of mix proportions using models, reducing trial-and-error costs and shortening the research and development cycle.

[0113] Furthermore, by analyzing the impact of variations in dosage on strength, the optimal formulation that meets performance requirements while minimizing cost can be identified. For example:

[0114] It was found that increasing the amount of solid waste reduced the strength, but the strength loss could be partially compensated by adjusting the amount of cement used, while reducing the amount of cement used (accounting for about 40% to 60% of the cost) and reducing material costs.

[0115] Determine the critical point between aggregate gradation and sand ratio to avoid overuse of high-cost fine aggregates while ensuring strength.

[0116] At the same time, by optimizing usage, the utilization of solid waste resources can be maximized, reducing dependence on natural aggregates. For example:

[0117] While meeting strength requirements, the solid waste content was increased from 30% to 50%, which reduced the consumption of natural aggregates and reduced the environmental risks caused by solid waste stockpiling.

[0118] Therefore, analyzing the correlation between changes in the amount of materials used in the manufacture of solid waste concrete samples and the maximum pressure is of practical significance.

[0119] It should be noted that this embodiment provides a scientific basis for optimizing material ratios by quantitatively analyzing the relationship between material usage and strength.

[0120] The established mathematical model can predict the maximum pressure of concrete under different material ratios, providing a basis for subsequent control.

[0121] S30: Perform relevant processing based on the associated impact, including:

[0122] Differentiate the dosage to obtain the maximum pressure corresponding to different dosages;

[0123] Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on different dosages (obviously, the difference between corresponding dosages can also be deduced from the pressure difference).

[0124] The control is based on the pressure difference, and the control is divided into central control and external central control. The central control is based on path segment I. The maximum pressure on the solid waste concrete sample is set at a future time according to the pressure difference (this maximum pressure is the maximum pressure set for path segment I), and the amount of material is adjusted based on the maximum pressure (i.e., the ratio of the amount of material is adjusted).

[0125] In this embodiment, by distinguishing the strength requirements and pressure differences at different locations, precise control of material usage is achieved to improve the quality and performance stability of concrete.

[0126] The combined use of central control and external central control makes the control strategy more flexible and adaptable to different engineering needs.

[0127] S40: The two path segments outside path segment I are marked as path segment II and path segment III. Path segment II is the path segment between path segments I and III. Outer center control is based on path segment III (of course, control can also be based on path segment II; this is flexible and can be adjusted as needed. In reality, if area is considered, the area of ​​the manufactured solid waste concrete samples varies. For example, in a small area, path segment II and path segment I are close together, and the pressure difference between their maximum pressures is not significant. In this case, it is not necessary to base control on path segment II; it is more appropriate to base control on path segment III. In a large area, path segment II and path segment I are separated by a certain distance. In this case, control can be based on path segment II, or it can be based on path segment III. In short, the choice is flexible and can be adjusted flexibly according to needs and / or purposes). Control is performed after relevant comparison. The steps of relevant comparison include:

[0128] Based on the relevant data obtained from the maximum pressure (recorded in S20), the maximum pressure that path segment III can withstand is determined.

[0129] Obtain the pressure difference between the maximum pressure that path segment III can withstand and the maximum pressure that path segment I can withstand; mark the pressure difference as the control pressure difference.

[0130] Obtain the relevant factors that cause this pressure difference, including calculating the amount of material corresponding to the maximum pressure that path segment III can withstand based on the material and amount corresponding to the maximum pressure that path segment I can withstand, and marking this amount as the reference amount;

[0131] When the solid waste concrete sample needs to reach the maximum pressure that path segment III can withstand in the future, the amount of material should be proportioned based on the reference dosage.

[0132] In this embodiment, the introduction of external center control makes the control range wider, enabling comprehensive control of different locations inside the concrete.

[0133] By comparing relevant steps and calculating reference dosages, the material ratio can be optimized, improving the uniformity and overall performance of concrete. In this way, the controllability and stability of the production process can be improved.

[0134] In S20, the correlation between the change in the amount of material used in manufacturing the solid waste concrete sample and the maximum pressure is analyzed and calculated according to the following formula:

[0135] ;

[0136] In the formula, This represents the maximum pressure (MPa, dependent variable, i.e., early intensity).

[0137] This indicates the amount of solid waste added (kg / m³ or mass percentage, such as when 20% of solid waste replaces natural aggregate). =20);

[0138] Indicates the amount of cement used;

[0139] This indicates the amount of aggregate used (kg / m³), or the percentage of sand content, such as when the sand content is 45%. =45);

[0140] , , , Represents the regression coefficients (used to reflect the contribution of each variable to the intensity);

[0141] This represents the error term (which follows a normal distribution).

[0142] This formula is used to quantify the linear effect of the amount of solid waste, cement, and aggregate on strength;

[0143] pass Determine the change in strength when the amount of solid waste added increases by 1 unit (e.g.) =-0.5, indicating that for every 1% increase in solid waste content, the strength decreases by 0.5 MPa.

[0144] Based on the above, it also includes the following formula for analysis and calculation:

[0145] ;

[0146] In the formula, express The quadratic term (this is to capture the non-linear relationship between the input and output variables).

[0147] In practical applications, such as concrete strength prediction, It can represent a certain influencing factor (such as the amount of solid waste added), while This indicates the nonlinear effect of the square of the factor on the intensity;

[0148] By introducing quadratic terms, the model can better describe the curvilinear relationship between intensity and factors.

[0149] The two calculation formulas mentioned above are closely related. The first formula is a multiple linear regression model used to analyze the impact of changes in the amount of materials such as solid waste, cement, and aggregates on the maximum pressure of concrete. This formula (model) quantifies the contribution of each material to concrete strength, providing a foundation for subsequent strength prediction and control.

[0150] The second calculation formula adds a quadratic term for material usage to the first formula to more accurately describe the nonlinear relationship between material usage and maximum pressure.

[0151] The purpose of adding quadratic terms is to capture the potential nonlinear effects between material usage and concrete strength. For example, when the usage of a certain material increases to a certain extent, its effect on improving concrete strength may weaken or even have a negative impact. By introducing quadratic terms, the model can more accurately describe this complex nonlinear relationship, thereby improving the accuracy of predictions.

[0152] In practical applications, the first calculation formula can be used for preliminary analysis to understand the basic contribution of each material to the strength of concrete. If it is found that the linear relationship is insufficient to accurately describe the actual situation, the second calculation formula can be used for more in-depth analysis and prediction.

[0153] Based on the above, a sensitivity analysis is further conducted according to the results of the analysis and calculation. The sensitivity analysis is based on the dosage, as shown below:

[0154] ;

[0155] ;

[0156] In the formula, Representing variables (such as solid waste content) The first-order sensitivity index (the contribution rate to the intensity variance when acting alone);

[0157] Representing variables and (like and The second-order sensitivity index (the contribution rate of the interaction to the strength variance).

[0158] Indicates variance;

[0159] An index used to measure the dispersion of values ​​of a random variable, indicating the degree to which the variable deviates from its expected value, in order to quantify model output (such as concrete strength). Total variability.

[0160] Its physical significance lies in the fact that, in the strength analysis of solid waste concrete, This reflects the range of fluctuation in strength values. A large variance indicates that the strength is significantly affected by changes in the amount of material used; a small variance indicates that the strength is relatively stable.

[0161] Indicates a prospective outlook.

[0162] Conditional expectation is the expected value of one variable given the values ​​of some other variables.

[0163] Used to calculate partial variance, which is the contribution of the remaining variables to the output variation when some variables are fixed;

[0164] In solid waste concrete ( The solid waste content (as opposed to the amount of solid waste added) can be understood as "the average level of strength at a specific solid waste content." If this value increases with... The significant changes indicate that the amount of solid waste added has a crucial impact on the strength.

[0165] This formula is used to identify key variables (such as...) > This indicates that the amount of solid waste added has a greater impact on strength than the amount of cement used.

[0166] This can quantify the interaction of multiple factors (such as the synergistic hydration effect of solid waste and cement).

[0167] Information is acquired based on the calculation results. The acquired information includes the amount of material that is most correlated with the maximum pressure, and the amount of this material is marked as the reference material amount. When the solid waste concrete sample needs to reach the corresponding maximum pressure in the future, the amount of this material is first proportioned.

[0168] Calculate the material usage corresponding to the maximum pressure that path segment III can withstand based on the material and usage corresponding to the maximum pressure that path segment I can withstand, using the following formula:

[0169] ;

[0170] ;

[0171] In the formula, Indicates the maximum pressure that path segment I can withstand;

[0172] Indicates and The corresponding amount of materials (such as solid waste, cement, aggregates, etc.) used;

[0173] This indicates the maximum pressure that path segment III can withstand;

[0174] Indicates and The corresponding amount of materials used;

[0175] The slope of the linear relationship represents the degree to which changes in material usage affect the maximum pressure.

[0176] The intercept represents the linear relationship, and the base pressure value when the amount of material used is zero (it may not be zero, as other factors may also affect the pressure).

[0177] The distance from the center of path segment I to the center of path segment III is obtained in the solid waste concrete sample. When the solid waste concrete sample (at the position corresponding to path segment I and path segment III) needs to reach the same pressure difference as the control pressure difference at a future time, the same amount of material as path segment I and path segment III is used for mixing.

[0178] Based on the above, this application constructs a hybrid spatial and strength dataset to deeply analyze the correlation between material usage and strength, thereby achieving accurate prediction and flexible control of concrete strength, which helps to reduce the consumption of natural resources and waste emissions.

[0179] This embodiment, combining the above-mentioned method for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins, also proposes a working system applied to this method, as follows:

[0180] Data construction module 110 is used to construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include:

[0181] Divide the space, including dividing it into diagonal spaces;

[0182] The strength of solid waste concrete samples was obtained in the diagonal space, and the steps are as follows:

[0183] Select the endpoint of the solid waste concrete, and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path;

[0184] Measure the length of the obtained path and divide the length into 3 path segments of equal length;

[0185] Strength was obtained at each path segment, including pressure testing of solid waste concrete samples at the middle position of each path segment.

[0186] The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained.

[0187] The data analysis module 120 is used to acquire relevant data based on the maximum pressure. The relevant data includes the materials and quantities corresponding to the manufacture of solid waste concrete samples. The materials include solid waste, cement, and aggregates. The module also analyzes the correlation between changes in the quantities of materials corresponding to the manufacture of solid waste concrete samples and the maximum pressure.

[0188] The data fusion processing module 130 includes a processing unit 1301 and a control unit 1302.

[0189] Processing unit 1301 is used to perform relevant processing based on the associated impact, including:

[0190] Differentiate the dosage to obtain the maximum pressure corresponding to different dosages;

[0191] Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on the different dosages;

[0192] The control unit 1302 is used to control according to the pressure difference. The control is divided into central control and outer central control. The central control is based on path segment I. The maximum pressure on the solid waste concrete sample is set according to the pressure difference at a future time, and the amount of material is adjusted based on the maximum pressure.

[0193] In summary, this invention constructs a hybrid spatial and strength dataset to deeply analyze the correlation between material usage and strength, thereby enabling accurate prediction and flexible control of concrete strength. Simultaneously, it promotes the optimization and standardization of material proportions, providing scientific decision support for concrete production and helping to reduce the consumption of natural resources and waste emissions.

[0194] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for predicting and regulating early strength of solid waste concrete based on transfer learning and digital twinning, characterized in that, Includes the following steps: S10: Construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include: The space is divided, including dividing it into diagonal spaces; The strength of the solid waste concrete sample is obtained in the diagonal space, and the steps are as follows: Select the endpoint of the solid waste concrete, and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path; Measure the length of the acquired path and divide the length into three path segments of equal length; Strength is obtained at each of the aforementioned path segments, including pressure testing of solid waste concrete samples at the midpoint of each path segment. The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained; S20: Based on the maximum pressure, obtain relevant data, including the materials and quantities corresponding to the manufacture of the solid waste concrete sample, the materials including solid waste, cement and aggregate; and analyze the correlation effect on the maximum pressure based on the changes in the quantities of materials corresponding to the manufacture of the solid waste concrete sample. S30: Perform relevant processing based on the aforementioned correlation effects, the relevant processing including: The dosage is differentiated to obtain the maximum pressure corresponding to different dosages; Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on the different dosages; The control is based on the pressure difference, and the control is divided into central control and outer central control. The central control is based on the I path segment, and the maximum pressure on the solid waste concrete sample is set at a future time according to the pressure difference, and the amount of material is adjusted based on the maximum pressure. S40: The two path segments other than path segment I are marked as path segment II and path segment III. Path segment II is the path segment between path segment I and path segment III. The external center control is based on path segment III. The control is performed after relevant comparison.

2. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 1, wherein In step S20, the correlation between the change in the amount of material used in manufacturing the solid waste concrete sample and the maximum pressure is analyzed and calculated according to the following formula: ； In the formula, Indicates maximum pressure; represents the solid waste content; represents the amount of cement; represents the amount of aggregate; , , , denotes the regression coefficient; denotes the error term.

3. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 2, wherein, It also includes calculations based on the following formula: ； In the formula, express The quadratic term.

4. The method for predicting and controlling the early strength of solid waste concrete based on transfer learning and digital twins as described in any one of claims 2 to 3, characterized in that, Sensitivity analysis was performed based on the results of the analysis and calculations. The sensitivity analysis was based on the dosage, as shown below: ； ； wherein denotes the first order sensitivity index of the variable denotes the first order sensitivity index of the variable denotes a variable with a second order sensitivity index; denotes the variance; represents a conditional expectation.

5. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 4, wherein, Information is acquired based on the calculation results. The acquired information includes the amount of material that is most correlated with the maximum pressure, and the amount of this material is marked as the reference material amount. When the solid waste concrete sample needs to reach the corresponding maximum pressure in the future, the amount of this material is first proportioned.

6. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 1, wherein, In step S40, adjustment is performed after correlation comparison. The correlation comparison step includes: Based on the relevant data obtained from the maximum pressure, the maximum pressure that path segment III can withstand is determined. Obtain the pressure difference between the maximum pressure that path segment III can withstand and the maximum pressure that path segment I can withstand; mark the pressure difference as the control pressure difference; Obtain the relevant factors that cause this pressure difference, including calculating the amount of material corresponding to the maximum pressure that path segment III can withstand based on the material and amount corresponding to the maximum pressure that path segment I can withstand, and marking this amount as the reference amount; When the solid waste concrete sample needs to reach the maximum pressure that the III path segment can withstand in the future, the amount of material should be proportioned based on the reference dosage.

7. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 6, wherein, Calculate the material usage corresponding to the maximum pressure that path segment III can withstand based on the material and usage corresponding to the maximum pressure that path segment I can withstand, using the following formula: ； ； wherein represents the maximum pressure that the section of path I can withstand; indicates the amount of the corresponding material; corresponding material; Pmax represents the maximum pressure that the section of path III can withstand; Indicates and The corresponding amount of materials used; The slope of the linear relationship represents the degree to which changes in material usage affect the maximum pressure. The intercept represents the linear relationship, and the base pressure value is represented when the amount of material used is zero.

8. The solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning of claim 6, wherein, The distance between the center of path segment I and the center of path segment III is obtained in the solid waste concrete sample. When the solid waste concrete sample needs to reach the same pressure difference as the control pressure difference in the future, the same amount of material as path segment I and path segment III is used for mixing.

9. The system applied to the solid waste concrete early strength prediction and regulation method based on transfer learning and digital twinning according to claim 1, characterized in that, include: The data construction module is used to construct a hybrid dataset of spatial and strength data for solid waste concrete samples. The steps include: The space is divided, including dividing it into diagonal spaces; The strength of the solid waste concrete sample is obtained in the diagonal space, and the steps are as follows: Select the endpoint of the solid waste concrete, and take the path from the endpoint to the center of the solid waste concrete as the strength acquisition path; Measure the length of the acquired path and divide the length into three path segments of equal length; Strength is obtained at each of the aforementioned path segments, including pressure testing of solid waste concrete samples at the midpoint of each path segment. The path segment closest to the center of the solid waste concrete is marked as path segment I, and the maximum pressure that path segment I can withstand is obtained; The data analysis module is used to acquire relevant data based on the maximum pressure, including the materials and quantities corresponding to the manufacture of the solid waste concrete sample, the materials including solid waste, cement and aggregate; and to analyze the correlation effect of changes in the quantities of materials corresponding to the manufacture of the solid waste concrete sample on the maximum pressure. A data fusion processing module, comprising a processing unit and a control unit; The processing unit is used to perform correlation processing based on the correlation effect, and the correlation processing includes: The dosage is differentiated to obtain the maximum pressure corresponding to different dosages; Obtain relevant information from the differentiated dosages, including the pressure difference for obtaining the maximum pressure based on the different dosages; The control unit is used to control the pressure difference, and the control is divided into central control and outer central control. The central control is based on the I path segment, and sets the maximum pressure on the solid waste concrete sample at a future time according to the pressure difference, and adjusts the amount of material based on the maximum pressure.

Images