A fan foundation concrete proportioning optimization method and system based on data analysis

Abstract

The application provides a fan foundation concrete proportioning optimization method and system based on data analysis, and relates to the field of engineering data analysis, and comprises the following steps: collecting historical concrete proportioning records and preprocessing to obtain a proportioning data set; evaluating the scene similarity of each historical concrete proportioning record and a target project, and calculating sample weights; weighting modeling the proportioning data set according to the sample weights to obtain a performance prediction model; discretizing and generating a candidate concrete proportioning scheme set within the value range of each proportioning design variable, obtaining the performance prediction value of each candidate scheme based on the performance prediction model, and calculating the predicted internal-external temperature difference of each candidate scheme; and screening according to the predicted internal-external temperature difference based on strength constraints, workability constraints, durability constraints and temperature control constraints to obtain a recommended concrete proportioning scheme. The application realizes parallel solving of temperature control compliance verification and cost optimization in the proportioning design stage, and improves the efficiency and compliance reliability of proportioning design.

Description

Technical Field

[0001] This invention relates to the field of engineering data analysis technology, and in particular to a method and system for optimizing the concrete mix proportion of wind turbine foundations based on data analysis. Background Technology

[0002] Wind turbine foundation piers are typical large-volume concrete structures with typically large minimum cross-sectional dimensions. The heat released during cement hydration is difficult to dissipate, and the temperature difference between the internal peak and surface of the structure can easily exceed the specified limits. Current standards for large-volume concrete construction mandate that the temperature difference between the inside and outside of the poured concrete should not exceed 25°C. This temperature control indicator is a rigid constraint that the mix design of wind turbine foundations must meet; violating this requirement will constitute a potential safety hazard for engineering quality.

[0003] In the field of concrete mix design, data-driven methods are increasingly widely used. The basic idea is to establish a predictive model between mix proportion parameters and concrete performance from historical mix proportion data, transforming the mix optimization problem into a mathematical problem of solving for the optimal mix proportion under constraints. Compared with the traditional trial-and-adjustment empirical design method, this significantly reduces experimental workload and improves design efficiency. However, due to the diversity of historical data sources, historical mix proportion records accumulated in engineering practice cover various application scenarios such as thin-plate structures, ordinary building foundations, and industrial components. These historical records differ fundamentally from those of wind turbine foundations in terms of structural cross-sectional heat dissipation conditions, durability design life requirements, and cementitious material system selection logic. How to reasonably establish predictive models suitable for specific application scenarios from heterogeneous historical data is a practical problem faced by data-driven mix design methods in engineering implementation.

[0004] Chinese invention patent application number 202210762533.8 discloses an intelligent design method for concrete mix proportions based on machine learning. This method uses concrete strength and flowability performance indicators as input variables to a neural network model, and the amount of each raw material as output variables, establishing an inverse mapping model between concrete performance and mix proportions. Based on this, it enumerates combinations of performance parameters and sorts them by cost, combining them with material unit prices to output a mix proportion design that meets strength and flowability requirements while maintaining low cost. However, this approach does not consider the differences between different scenarios, nor does it incorporate internal and external temperature differences into the mix proportion optimization process. This could lead to the risk of non-compliant concrete mix proportions requiring rework. Summary of the Invention

[0005] In view of this, the present invention provides a data analysis-based method and system for optimizing the concrete mix proportion of wind turbine foundations. It solves the technical problem in the prior art where the decoupling between temperature control constraints and mix proportion optimization leads to difficulty in ensuring the accuracy of concrete mix proportions. It enables the parallel solution of temperature control compliance verification and cost optimization in the mix proportion design stage, thereby improving the efficiency and compliance reliability of mix proportion design.

[0006] The technical solution of this invention is implemented as follows: On the one hand, this invention provides a data analysis-based method for optimizing the concrete mix proportions of wind turbine foundations, including: S1. Collect historical concrete mix proportion records from the historical concrete mix proportion database of wind turbine foundation projects, preprocess the historical concrete mix proportion records, and obtain the mix proportion dataset. S2. Based on the scenario characteristics of the target wind turbine foundation project, evaluate the scenario similarity between each historical concrete mix ratio record in the mix ratio dataset and the target project, and calculate the weight of each historical concrete mix ratio record based on the scenario similarity to obtain the sample weight of each historical concrete mix ratio record. S3. Based on the sample weights of each historical concrete mix design record, a weighted model is constructed for the mix design dataset to obtain a performance prediction model with mix design parameters as independent variables. S4. Discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Input each candidate scheme in the set of candidate concrete mix design schemes into the performance prediction model to obtain the performance prediction value of each candidate scheme. Calculate the predicted internal and external temperature difference of each candidate scheme based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project. S5. Based on the predicted internal and external temperature difference, the candidate concrete mix design schemes are screened according to strength constraints, workability constraints, durability constraints and temperature control constraints to obtain the recommended concrete mix design scheme.

[0007] Based on the above technical solutions, preferably, the historical concrete mix proportion records include concrete mix proportion parameters and scene characteristics, wherein the concrete mix proportion parameters include cement dosage. Mineral powder content fly ash content Water-to-glue ratio and sand ratio The scene characteristics include the minimum size of the structural cross section, the design service life, and the total amount of mineral admixtures.

[0008] Based on the above technical solutions, preferably, step S2 specifically includes: For the first in the proportion dataset Each historical concrete mix design record was compared with the scene feature parameters corresponding to the target project to obtain the cross-sectional feature similarity. Durability design level similarity Similarity to cementitious material systems ; Based on cross-sectional feature similarity Durability design level similarity Similarity to cementitious material systems By merging the data, a comprehensive scene similarity score can be obtained. : Combine scene similarity Transformed into the first exponential form Sample weights of historical records :

[0009] in This is the sensitivity coefficient for similarity-weight transformation.

[0010] Based on the above technical solutions, the preferred formula for calculating the comprehensive scene similarity is:

[0011]

[0012]

[0013]

[0014]

[0015] in, The similarity decay coefficient for cross-sectional dimensions represents the rate of decrease in the similarity index caused by each unit deviation in cross-sectional dimensions. This indicates the minimum cross-sectional dimension of the structure in historical records. Indicates the minimum dimension of the target engineering structure cross-section. The durability lifespan similarity decay coefficient characterizes the rate of similarity index decay caused by each unit design life deviation. Indicates the design lifespan of historical records. Indicates the design service life of the target project. The similarity decay coefficient of the cementitious material system characterizes the rate of decay of the similarity index caused by the deviation in the total amount of mineral admixture per unit. This indicates the total amount of mineral admixtures used in historical records. This indicates the target value for the total amount of mineral admixtures used in the target project. The weights representing the similarity of cross-sectional features The weights representing the similarity of durability design levels The weights representing the similarity of cementitious material systems.

[0016] Based on the above technical solution, preferably, the historical concrete mix design record also includes four measured performance values, namely, measured compressive strength, measured slump, measured adiabatic temperature rise, and measured chloride ion diffusion coefficient. Step S3 specifically includes: All of the data in the matching dataset Using historical records as a sample, for each historical record... Constructing the independent variable vector from the proportioning parameters and intercept term ; Establish a weighted least squares objective function and to Solving the system yields a set of normal equations. The unique analytical solution to this set is the response value vector. Response vectors are constructed using measured performance values, including the regression coefficient vector of the compressive strength prediction equation. The regression coefficient vector of the slump prediction equation The regression coefficient vector of the adiabatic temperature rise prediction equation and the regression coefficient vector of the chloride ion diffusion coefficient prediction equation :

[0017]

[0018] in To augment the design matrix, its first... Behavior ; This is a weight matrix with the weights of each sample as diagonal elements; For the response value vector, Indicates the first Records of historical concrete mix proportions and measured performance values; For the ratio parameter field to be predicted, construct the independent variable vector and perform the inner product with the response vector to obtain four prediction equations, including the compressive strength prediction equation, the slump prediction equation, the adiabatic temperature rise prediction equation, and the chloride ion diffusion coefficient prediction equation, thus forming a performance prediction model.

[0019] Based on the above technical solutions, preferably, step S4 specifically includes: Based on the proportioning parameter fields, uniform discretization is performed within their respective value ranges to obtain sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio. Enumerating all finite quinary parameter combinations on the Cartesian product of the sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio yields the complete set of finite quinary parameter combinations. ; The total amount of cementitious materials is calculated based on the cement content, mineral powder content, and fly ash content. The water consumption per unit volume is then calculated based on the product of the total amount of cementitious materials and the water-cement ratio. ; Based on cement usage, mineral powder content, fly ash content, and water consumption per unit volume. Calculate the amount of sand used Dosage of He Shi ; Based on the complete set of finite five-element parameter combinations Water consumption per unit volume, sand consumption Dosage of He Shi Forming a set of candidate concrete mix design schemes ; Set of candidate concrete mix proportions The proportioning parameters of each candidate mix design are substituted into the performance prediction model to obtain the predicted values ​​of compressive strength, slump, adiabatic temperature rise, and chloride ion diffusion coefficient. Based on the heat dissipation correction coefficient of the target project, the predicted internal and external temperature difference of the candidate mix design is calculated.

[0020] Based on the above technical solutions, the preferred formula for calculating the predicted internal and external temperature difference is:

[0021]

[0022] in, This indicates the predicted temperature difference between inside and outside. This represents the heat dissipation correction factor. This indicates the predicted peak adiabatic temperature rise. This represents the combined thermal conductivity related to the thermal diffusivity of concrete and the geometry of the structure. This represents the minimum structural cross-sectional dimension of the target project. This represents the overall thermal resistance of the insulation and maintenance measures for the target project. Indicates the ambient temperature during the construction period. This indicates the correction amount for the temperature calculation baseline.

[0023] Based on the above technical solutions, preferably, step S5 specifically includes: Set of candidate concrete mix proportions Each candidate formulation was screened, and the predicted compressive strength was retained. Not lower than the formulation strength The candidate solutions are then used to obtain the first subset of candidate solutions; The subset of first candidate solutions is filtered by workability constraints, retaining the predicted slump. Falling below the lower limit of the slump required for construction and upper limit Among the candidate solutions, a subset of the second candidate solutions is obtained; The subset of second candidate schemes was screened using temperature control constraints, retaining the predicted internal and external temperature differences. The candidate schemes that do not exceed the temperature control limit of 25℃ for large-volume concrete are used to obtain the third candidate subset. The third candidate subset was screened using durability constraints, retaining the predicted chloride ion diffusion coefficient. Not exceeding the upper limit of the durability design of the target project The candidate schemes were analyzed to obtain the fourth candidate subset. ; If the fourth candidate subset If not empty, then start from the fourth candidate subset. The recommended concrete mix design is determined in the middle; if the fourth candidate subset If the value is empty, it is determined that there is no feasible allocation scheme that simultaneously satisfies all four constraints within the current range of design variable values. A prompt message is output, and the range of design variable values ​​is adjusted or the constraints are relaxed.

[0024] More preferably, the fourth candidate subset The recommended concrete mix design is determined in the middle, specifically including: With the optimization objective of minimizing the total cost of concrete per cubic meter, the fourth candidate subset is selected. The material parameters for each candidate scheme are combined with the market unit price of each component to establish a total material cost objective function. :

[0025] in, This indicates the unit price of cement. This indicates the unit price of mineral powder. This indicates the unit price of fly ash. This indicates the unit price of water. This indicates the unit price of sand. Indicates the unit price of the stone; For the fourth candidate subset The objective function value is calculated for each of the candidate solutions, and the global enumeration method is used to select the fourth candidate subset. The minimum value of the objective function is obtained, and the candidate mix design corresponding to the minimum value of the objective function is taken as the recommended concrete mix design.

[0026] In addition, the present invention also provides a data analysis-based concrete mix design optimization system for wind turbine foundations to implement the aforementioned data analysis-based concrete mix design optimization method for wind turbine foundations, comprising: The data acquisition module is used to collect historical concrete mix ratio records from the historical concrete mix ratio database of wind turbine foundation projects, preprocess the historical concrete mix ratio records, and obtain the mix ratio dataset. The weight calculation module is used to evaluate the scene similarity between each historical concrete mix ratio record in the mix ratio dataset and the scene of the target wind turbine foundation project, based on the scene similarity, and to calculate the weight of each historical concrete mix ratio record to obtain the sample weight of each historical concrete mix ratio record. The model building module is used to perform weighted modeling of the mix proportion dataset based on the sample weights of each historical concrete mix proportion record, and obtain a performance prediction model with mix proportion parameters as independent variables. The temperature difference calculation module is used to discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Each candidate scheme in the set of candidate concrete mix design schemes is input into the performance prediction model to obtain the performance prediction value of each candidate scheme. Based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project, the predicted internal and external temperature difference of each candidate scheme is calculated. The scheme selection module is used to filter candidate concrete mix design schemes based on predicted internal and external temperature differences, constrained by strength, workability, durability, and temperature control, to obtain recommended concrete mix design schemes.

[0027] The present invention has the following advantages over the prior art: By applying adiabatic temperature rise prediction, heat dissipation correction, and strength, workability, durability, and temperature control constraints in parallel during the selection of concrete mix design schemes for wind turbine foundations, the output recommended mix design achieves optimal material cost while satisfying all constraints. This eliminates the inefficient iteration problem caused by placing temperature control verification after mix design in existing methods, and enables parallel solutions for temperature control compliance verification and cost optimization during the mix design stage, thereby improving the efficiency and compliance reliability of mix design. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0029] Figure 1 This is a flowchart of a data analysis-based method for optimizing the concrete mix proportion of wind turbine foundations according to the present invention. Figure 2 This is a schematic diagram illustrating the multi-constraint screening method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis, as proposed in this invention. Detailed Implementation

[0030] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0031] like Figure 1 As shown, this invention provides a data analysis-based method for optimizing the concrete mix design of wind turbine foundations, comprising: S1. Collect historical concrete mix proportion records from the historical concrete mix proportion database of wind turbine foundation projects, preprocess the historical concrete mix proportion records to obtain a mix proportion dataset; wherein, the historical concrete mix proportion records include concrete mix proportion parameters and scene features, wherein the concrete mix proportion parameters include cement dosage. Mineral powder content fly ash content Water-to-glue ratio and sand ratio The scene characteristics include the minimum size of the structural cross section, the design service life, and the total amount of mineral admixtures.

[0032] Specifically, preprocessing includes outlier detection and data normalization. Outlier detection employs the interquartile range method, including: Perform interquartile range outlier detection independently on each consecutive field in the historical concrete mix design records, and calculate the first quartile of the field. With the third quartile The interquartile range is obtained. Where j represents the field's ordinal index, used to distinguish different continuous fields in historical concrete mix design records. This represents the interquartile range of field j; The interquartile range falls within the interval Historical concrete mix design records other than those specified are identified as abnormal records and removed. The records retained after outlier detection of all fields constitute the clean dataset. The total number of records in the clean dataset is denoted as N. The original values ​​of each record in the cleaning dataset are processed using the min-max normalization method for each continuous field to obtain the matching dataset:

[0033] in, Indicates the first Fields in historical records The normalized values ​​range from 0 to 0. , Indicates the first Fields in historical records The original value, This represents the minimum value of field j in the clean dataset. This represents the maximum value of field j in the clean dataset.

[0034] S2. Based on the scenario characteristics of the target wind turbine foundation project, evaluate the scenario similarity between each historical concrete mix ratio record in the mix ratio dataset and the target project, and calculate the weight of each historical concrete mix ratio record based on the scenario similarity to obtain the sample weight of each historical concrete mix ratio record. Specifically, step S2 includes: For the first in the proportion dataset Each historical concrete mix design record was compared with the scene feature parameters corresponding to the target project to obtain the cross-sectional feature similarity. Durability design level similarity Similarity to cementitious material systems ; Based on cross-sectional feature similarity Durability design level similarity Similarity to cementitious material systems By merging the data, a comprehensive scene similarity score can be obtained. :

[0035]

[0036]

[0037]

[0038]

[0039] in, The similarity decay coefficient for cross-sectional dimensions represents the rate of decrease in the similarity index caused by each unit deviation in cross-sectional dimensions. This indicates the minimum cross-sectional dimension of the structure in historical records. Indicates the minimum dimension of the target engineering structure cross-section. The durability lifespan similarity decay coefficient characterizes the rate of similarity index decay caused by each unit design life deviation. Indicates the design lifespan of historical records. Indicates the design service life of the target project. The similarity decay coefficient of the cementitious material system characterizes the rate of decay of the similarity index caused by the deviation in the total amount of mineral admixture per unit. This indicates the total amount of mineral admixtures used in historical records. This indicates the target value for the total amount of mineral admixtures used in the target project. The weights representing the similarity of cross-sectional features The weights representing the similarity of durability design levels Weights representing the similarity of cementitious material systems; Combine scene similarity Transformed into the first exponential form Sample weights of historical records :

[0040] in It is a sensitivity coefficient for similarity-weight transformation, controlling the rate at which the weight distribution shifts towards the high-similarity sample set.

[0041] Understandable. and . The range of values ​​is The closer the value is to 1, the more similar the historical record is to the scenario of the target project. When the values ​​are small, the weight distribution is more uniform; When the value is larger, the advantage of high similarity samples over low similarity samples increases significantly.

[0042] This invention introduces a scenario similarity calculation mechanism with three evaluation dimensions: structural cross-sectional characteristics, durability level, and cementitious material system. The scenario similarity between each historical concrete mix ratio record and the target project is transformed into sample weights in weighted least squares modeling through exponential transformation. This enables the performance prediction model to achieve directional convergence within the wind turbine foundation mix ratio parameter space, thus improving the problem of decreased prediction accuracy under the condition of equal-weighted mixed modeling of heterogeneous historical data in multiple scenarios.

[0043] S3. Weighted modeling of the mix design dataset is performed based on the sample weights of each historical concrete mix design record to obtain a performance prediction model with mix design parameters as independent variables. The historical concrete mix design records also include four measured performance values, namely the measured compressive strength values. Measured slump value Measured values ​​of adiabatic temperature rise Measured value of chloride ion diffusion coefficient ; Specifically, step S3 includes: All of the data in the matching dataset Using historical records as a sample, for each historical record... Constructing the independent variable vector from the proportioning parameters and intercept term ; where 1 is the intercept term; Establish a weighted least squares objective function and to Solving the system yields a set of normal equations. The unique analytical solution to this set is the response value vector. Response vectors are constructed using measured performance values, including the regression coefficient vector of the compressive strength prediction equation. The regression coefficient vector of the slump prediction equation The regression coefficient vector of the adiabatic temperature rise prediction equation and the regression coefficient vector of the chloride ion diffusion coefficient prediction equation :

[0044]

[0045] in To augment the design matrix, its first... Behavior ; This is a weight matrix with the weights of each sample as diagonal elements; For the response value vector, This represents the vector of regression coefficients to be estimated. Indicates the first Records of historical concrete mix proportions and measured performance values; This represents the intercept term of the compressive strength prediction equation. The amount of cement used in the equation for predicting compressive strength The regression coefficients, The amount of mineral powder in the compressive strength prediction equation The regression coefficients, The fly ash content in the compressive strength prediction equation The regression coefficients, The water-cement ratio in the compressive strength prediction equation The regression coefficients, The sand ratio in the compressive strength prediction equation The regression coefficients, This represents the intercept term of the slump prediction equation. The amount of cement used in the slump prediction equation The regression coefficients, Indicates the amount of mineral powder in the slump prediction equation The regression coefficients, The fly ash content in the slump prediction equation The regression coefficients, The water-cement ratio in the slump prediction equation The regression coefficients, The sand ratio in the slump prediction equation The regression coefficients, This represents the intercept term of the adiabatic temperature rise prediction equation. The amount of cement used in the adiabatic temperature rise prediction equation The regression coefficients, The amount of mineral powder in the adiabatic temperature rise prediction equation The regression coefficients, The fly ash content in the adiabatic temperature rise prediction equation The regression coefficients, The water-cement ratio in the adiabatic temperature rise prediction equation The regression coefficients, The sand ratio in the adiabatic temperature rise prediction equation The regression coefficients, This represents the intercept term in the equation for predicting the chloride ion diffusion coefficient. The amount of cement used in the equation for predicting the chloride ion diffusion coefficient The regression coefficients, The amount of mineral powder in the equation for predicting the chloride ion diffusion coefficient The regression coefficients, The fly ash content in the equation for predicting the chloride ion diffusion coefficient The regression coefficients, The water-to-gel ratio in the chloride ion diffusion coefficient prediction equation The regression coefficients, The sand ratio in the chloride ion diffusion coefficient prediction equation The regression coefficients; For the ratio parameter fields to be predicted, construct the independent variable vector and perform the inner product with the response vector to obtain four prediction equations, including the compressive strength prediction equation, the slump prediction equation, the adiabatic temperature rise prediction equation, and the chloride ion diffusion coefficient prediction equation, forming a performance prediction model:

[0046]

[0047]

[0048]

[0049] in, This represents the predicted compressive strength of the candidate mix design. This represents the predicted slump value of the candidate mix design. This represents the predicted adiabatic temperature rise of the candidate mix design. This represents the predicted chloride ion diffusion coefficient of the candidate formulation.

[0050] Understandably, intercept term 1 is in the independent variable vector This means that regardless of the value of the matching parameter, this component and its corresponding intercept regression coefficient... The product of is always equal to It itself implements the inclusion of the intercept term into the vector dot product operation. The goal is to allow prediction equations to be written in a uniform matrix form without requiring separate treatment of the intercept term. Based on the sum of squared residuals in ordinary least squares, the squared residuals of each historical record are assigned their corresponding sample weights. The proportional weighting coefficients cause the objective function to impose a greater penalty on the fitting error of high-weight (i.e., high-similarity) samples during the minimization process, thereby causing the regression coefficient vector obtained by the solution to shift in the fitting direction towards the data pattern of high-similarity samples, achieving the purpose of directional modeling.

[0051] By assigning nonlinear enhancement weights to historical records that are highly similar to the target engineering scenario, the regression coefficients of the performance prediction model are given greater fitting weights to samples within the thermal parameter space of the wind turbine foundation when solving the model. This reduces the interference of heterogeneous scenario data on the regression coefficients of the prediction equation and improves the applicability of the adiabatic temperature rise prediction equation within the wind turbine foundation proportioning parameter space.

[0052] S4. Discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Input each candidate scheme in the set of candidate concrete mix design schemes into the performance prediction model to obtain the performance prediction value of each candidate scheme. Calculate the predicted internal and external temperature difference of each candidate scheme based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project. Specifically, step S4 includes: Based on the proportioning parameter fields, uniform discretization is performed within their respective value ranges to obtain sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio. Enumerating all finite quinary parameter combinations on the Cartesian product of the sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio yields the complete set of finite quinary parameter combinations. ; The total amount of cementitious materials is calculated based on the cement content, mineral powder content, and fly ash content. The water consumption per unit volume is then calculated based on the product of the total amount of cementitious materials and the water-cement ratio. ; Based on cement usage, mineral powder content, fly ash content, and water consumption per unit volume. Calculate the amount of sand used Dosage of He Shi ; Based on the complete set of finite five-element parameter combinations Water consumption per unit volume, sand consumption Dosage of He Shi Forming a set of candidate concrete mix design schemes ; Set of candidate concrete mix proportions The mixing parameters of each candidate mix design are substituted into the performance prediction model to obtain predicted values ​​for compressive strength, slump, adiabatic temperature rise, and chloride ion diffusion coefficient. Based on the heat dissipation correction coefficient of the target project, the predicted internal and external temperature difference of the candidate mix design is calculated.

[0053]

[0054] in, This indicates the predicted temperature difference between inside and outside. This represents the heat dissipation correction factor. This indicates the predicted peak adiabatic temperature rise. This represents the combined thermal conductivity related to the thermal diffusivity of concrete and the geometry of the structure. This represents the minimum structural cross-sectional dimension of the target project. This represents the overall thermal resistance of the insulation and maintenance measures for the target project. Indicates the ambient temperature during the construction period. This indicates the correction amount for the temperature calculation baseline.

[0055] Understandably, in actual engineering structures, concrete releases heat during hydration while simultaneously dissipating heat to the outside, resulting in a lower actual temperature difference between the inside and outside compared to the peak temperature rise under adiabatic conditions. (Heat dissipation correction factor) This represents an estimated ratio of the actual internal and external temperature difference to the peak adiabatic temperature rise, used to convert the predicted adiabatic temperature rise into the predicted internal and external temperature difference under actual heat dissipation conditions. Minimum structural cross-sectional dimensions. The square of the coefficient reflects the square effect of the path length of heat conduction from the structural core to the surface on the heat dissipation rate. The larger the cross-section, the more difficult it is for internal heat to dissipate. The closer to 1; the greater the overall thermal resistance. This reflects the effect of thermal insulation measures on heat dissipation; the greater the thermal resistance, the slower the heat dissipation. The closer to 1; ambient temperature correction term This reflects the influence of external temperature boundary conditions on the heat dissipation driving force. The exponential term approaches 0 in extreme cases with large cross-sections and high thermal resistance, making... The exponent term approaches 1, corresponding to a perfectly adiabatic thermal shape; in cases of small cross-section and low thermal resistance, the exponent term is larger, making... A value significantly less than 1 indicates good heat dissipation, which is consistent with physical laws.

[0056] This invention incorporates engineering characteristic parameters such as the minimum structural cross-sectional size, the comprehensive thermal resistance of insulation measures, and the ambient temperature during construction into the calculation of the heat dissipation correction coefficient. It converts the predicted adiabatic temperature rise into a predicted internal and external temperature difference corresponding to the thermal conditions of the target project. This transforms the basis for applying temperature control constraints from general adiabatic temperature rise results to temperature difference estimates based on the actual heat dissipation conditions of the target project, thereby enhancing the correspondence between temperature control compliance judgments and actual engineering conditions.

[0057] In one embodiment of the present invention, the amount of sand used... Dosage of He Shi The calculation method is as follows: Candidate formulation Calculate the total amount of cementitious materials and the water consumption per unit volume based on the cementitious material components:

[0058]

[0059] in, This indicates the amount of water used per unit volume of concrete. Indicates the water-to-glue ratio; Under the absolute volume method, the sum of the absolute volumes of the six components (cement, mineral powder, fly ash, water, sand, and aggregate) in one cubic meter of concrete equals 1. The sum of the absolute volumes of sand and aggregate is:

[0060] in , , , These are the apparent densities of cement, mineral powder, fly ash, and the density of water, respectively. This represents the sum of the absolute volumes of sand and stone in each cubic meter of concrete. Due to sand ratio (Defined as the proportion of sand's mass to the total mass of sand and gravel) (Simultaneous equations) The absolute volumes of sand and stone are obtained as follows:

[0061]

[0062] in, This indicates the absolute volume of sand per cubic meter of concrete. This indicates the absolute volume of stone in each cubic meter of concrete. , These are the apparent densities of sand and stone, respectively. Sand dosage Dosage of stone They are respectively:

[0063]

[0064] In one embodiment of the present invention, in the wind turbine foundation cylinder structure, the cross-sectional temperature distribution is not a simple linear gradient, but rather exhibits a nonlinear amplification due to the accumulation of friction impedance along the conduction path as heat is conducted outward from the central region. Based on the heat dissipation correction coefficient, the predicted internal and external temperature difference is corrected as follows:

[0065]

[0066] in, This represents the predicted maximum internal and external temperature difference after fractal boundary correction. This indicates the predicted temperature difference between inside and outside. This represents the amplification factor of thermal accumulation at the fractal boundary. , This represents the heat accumulation sensitivity coefficient, with a value range of... , Indicates the minimum dimension of the structural cross-section. The fractal dimension of the heat dissipation boundary of the cross section, with a range of values. Characterizes the complexity of the cross-sectional geometric boundary. Indicates the thermal diffusivity of concrete. Indicates the moment when the peak temperature of concrete hydration occurs; Understandable. The term couples the thermal accumulation effect of the cross-sectional dimensions with the geometric complexity of the heat dissipation boundary. Time degenerates into (i.e., flat boundaries without additional magnification), when As the cross-section size increases, the thermal accumulation effect of the large cross-section is nonlinearly amplified; The term corresponds in dimensions to the characteristic length of thermal diffusion at its peak (i.e., (unit: m), normalizing the degree of thermal accumulation amplification of the cross-sectional dimension to the material's own thermal diffusivity, so that Maintain dimensional consistency and physical comparability for concretes with different thermal conductivity. In temperature-controlled screening, [the following criteria are used]. (Typically 25℃) is used as a criterion for temperature control compliance, replacing the original This allows for the introduction of a conservative non-uniform temperature difference correction during the formulation screening stage, reducing the risk of on-site rework after solution delivery due to underestimation of local temperature differences in macro heat dissipation estimation.

[0067] Furthermore, Characterizing the complexity of the cross-sectional geometric boundary, for circular or regular rectangular cross-sections, the heat dissipation boundary is smooth, and we take... For structures with pre-reserved holes, varying thicknesses, or irregular cross-sections, where the heat dissipation path is tortuous, take... It can be estimated using the box count method of the cross-sectional profile; for conventional wind turbine foundation cylindrical cross-sections, it is recommended to take... .

[0068] S5. Based on the predicted internal and external temperature difference, the candidate concrete mix design schemes are screened according to strength constraints, workability constraints, durability constraints and temperature control constraints to obtain the recommended concrete mix design scheme.

[0069] like Figure 2 As shown, specifically, step S5 includes: Set of candidate concrete mix proportions Each candidate formulation was screened, and the predicted compressive strength was retained. Not lower than the formulation strength The candidate solutions are then used to obtain the first subset of candidate solutions; The subset of first candidate solutions is filtered by workability constraints, retaining the predicted slump. Falling below the lower limit of the slump required for construction and upper limit Among the candidate solutions, a subset of the second candidate solutions is obtained; The subset of second candidate schemes was screened using temperature control constraints, retaining the predicted internal and external temperature differences. The candidate schemes that do not exceed the temperature control limit of 25℃ for large-volume concrete are used to obtain the third candidate subset. The third candidate subset was screened using durability constraints, retaining the predicted chloride ion diffusion coefficient. Not exceeding the upper limit of the durability design of the target project The candidate schemes were analyzed to obtain the fourth candidate subset. ; If the fourth candidate subset If not empty, then start from the fourth candidate subset. The recommended concrete mix design is determined in the middle; if the fourth candidate subset If the value is empty, it is determined that there is no feasible allocation scheme that simultaneously satisfies all four constraints within the current range of design variable values. A prompt message is output, and the range of design variable values ​​is adjusted or the constraints are relaxed.

[0070] In one embodiment of the present invention, from the fourth candidate subset The recommended concrete mix design is determined in the middle, specifically including: With the optimization objective of minimizing the total cost of concrete per cubic meter, the fourth candidate subset is selected. The material parameters for each candidate scheme are combined with the market unit price of each component to establish a total material cost objective function. :

[0071] in, This indicates the unit price of cement. This indicates the unit price of mineral powder. This indicates the unit price of fly ash. This indicates the unit price of water. This indicates the unit price of sand. Indicates the unit price of the stone; For the fourth candidate subset The objective function value is calculated for each of the candidate solutions, and the global enumeration method is used to select the fourth candidate subset. The minimum value of the objective function is obtained, and the candidate mix design corresponding to the minimum value of the objective function is taken as the recommended concrete mix design.

[0072] This invention applies adiabatic temperature rise prediction, heat dissipation correction, and strength, workability, durability, and temperature control constraints in parallel during the selection stage of wind turbine foundation concrete mix design. The output recommended mix design achieves optimal material cost while satisfying all constraints. This eliminates the inefficient iteration problem caused by placing temperature control verification after mix design in existing methods. It enables parallel solutions for temperature control compliance verification and cost optimization during the mix design stage, improving the efficiency and compliance reliability of mix design.

[0073] In one embodiment of the present invention, the method for optimizing the concrete mix proportion of wind turbine foundations further includes: Prepare specimens according to the recommended mix proportions and relevant standards, cure them, and determine the actual compressive strength. and measured chloride ion diffusion coefficient The verification pass criteria are: and ; When conditions permit, the measured adiabatic temperature rise can be determined simultaneously. and verify ; Once the verification is successful, all fields of the recommended formulation scheme will be recorded, including the formulation parameters. , , , , Actual performance values , , and the scene feature fields of the target project , , This creates a complete mix proportion record and stores it in the historical concrete mix proportion database for wind turbine foundations. , , , , The proportioning parameters are used to recommend a concrete mix design.

[0074] This invention also provides a data analysis-based concrete mix design optimization system for wind turbine foundations, used to implement the data analysis-based concrete mix design optimization method for wind turbine foundations as described above, comprising: The data acquisition module is used to collect historical concrete mix ratio records from the historical concrete mix ratio database of wind turbine foundation projects, preprocess the historical concrete mix ratio records, and obtain the mix ratio dataset. The weight calculation module is used to evaluate the scene similarity between each historical concrete mix ratio record in the mix ratio dataset and the scene of the target wind turbine foundation project, based on the scene similarity, and to calculate the weight of each historical concrete mix ratio record to obtain the sample weight of each historical concrete mix ratio record. The model building module is used to perform weighted modeling of the mix proportion dataset based on the sample weights of each historical concrete mix proportion record, and obtain a performance prediction model with mix proportion parameters as independent variables. The temperature difference calculation module is used to discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Each candidate scheme in the set of candidate concrete mix design schemes is input into the performance prediction model to obtain the performance prediction value of each candidate scheme. Based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project, the predicted internal and external temperature difference of each candidate scheme is calculated. The scheme selection module is used to select candidate concrete mix design schemes based on the predicted internal and external temperature differences, and based on strength constraints, workability constraints, durability constraints, and temperature control constraints, to obtain recommended concrete mix design schemes.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the concrete mix design of wind turbine foundations based on data analysis, characterized in that, include: S1. Collect historical concrete mix proportion records from the historical concrete mix proportion database of wind turbine foundation projects, preprocess the historical concrete mix proportion records, and obtain the mix proportion dataset. S2. Based on the scenario characteristics of the target wind turbine foundation project, evaluate the scenario similarity between each historical concrete mix ratio record in the mix ratio dataset and the target project, and calculate the weight of each historical concrete mix ratio record based on the scenario similarity to obtain the sample weight of each historical concrete mix ratio record. S3. Based on the sample weights of each historical concrete mix design record, a weighted model is constructed for the mix design dataset to obtain a performance prediction model with mix design parameters as independent variables. S4. Discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Input each candidate scheme in the set of candidate concrete mix design schemes into the performance prediction model to obtain the performance prediction value of each candidate scheme. Calculate the predicted internal and external temperature difference of each candidate scheme based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project. S5. Based on the predicted internal and external temperature difference, the candidate concrete mix design schemes are screened according to strength constraints, workability constraints, durability constraints and temperature control constraints to obtain the recommended concrete mix design scheme.

2. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 1, characterized in that: The historical concrete mix design records include concrete mix design parameters and scene characteristics, wherein the concrete mix design parameters include cement content. Mineral powder content fly ash content Water-to-glue ratio and sand ratio The scene characteristics include the minimum size of the structural cross section, the design service life, and the total amount of mineral admixtures.

3. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 2, characterized in that: Step S2 specifically includes: For the first in the proportion dataset Each historical concrete mix design record was compared with the scene feature parameters corresponding to the target project to obtain the cross-sectional feature similarity. Durability design level similarity Similarity to cementitious material systems ; Based on cross-sectional feature similarity Durability design level similarity Similarity to cementitious material systems By merging the data, a comprehensive scene similarity score can be obtained. : Combine scene similarity Transformed into the first exponential form Sample weights of historical records : in This is the sensitivity coefficient for similarity-weight transformation.

4. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 3, characterized in that: The formula for calculating the overall scene similarity is: in, The similarity decay coefficient for cross-sectional dimensions represents the rate of decrease in the similarity index caused by each unit deviation in cross-sectional dimensions. This indicates the minimum cross-sectional dimension of the structure in historical records. Indicates the minimum dimension of the target engineering structure cross-section. The durability lifespan similarity decay coefficient characterizes the rate of similarity index decay caused by each unit design life deviation. Indicates the design lifespan of historical records. Indicates the design service life of the target project. The similarity decay coefficient of the cementitious material system characterizes the rate of decay of the similarity index caused by the deviation in the total amount of mineral admixture per unit. This indicates the total amount of mineral admixtures used in historical records. This indicates the target value for the total amount of mineral admixtures used in the target project. The weights representing the similarity of cross-sectional features The weights representing the similarity of durability design levels The weights representing the similarity of cementitious material systems.

5. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 4, characterized in that: The historical concrete mix design record also includes four measured performance values: measured compressive strength, measured slump, measured adiabatic temperature rise, and measured chloride ion diffusion coefficient. Step S3 specifically includes: All of the data in the matching dataset Using historical records as a sample, for each historical record... Constructing the independent variable vector from the proportioning parameters and intercept term ; Establish a weighted least squares objective function and to Solving the system yields a set of normal equations. The unique analytical solution to this set is the response value vector. Response vectors are constructed using measured performance values, including the regression coefficient vector of the compressive strength prediction equation. The regression coefficient vector of the slump prediction equation The regression coefficient vector of the adiabatic temperature rise prediction equation and the regression coefficient vector of the chloride ion diffusion coefficient prediction equation : in To augment the design matrix, its first... Behavior ; This is a weight matrix with the weights of each sample as diagonal elements; For the response value vector, Indicates the first Records of historical concrete mix proportions and measured performance values; For the ratio parameter field to be predicted, construct the independent variable vector and perform the inner product with the response vector to obtain four prediction equations, including the compressive strength prediction equation, the slump prediction equation, the adiabatic temperature rise prediction equation, and the chloride ion diffusion coefficient prediction equation, thus forming a performance prediction model.

6. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 5, characterized in that: Step S4 specifically includes: Based on the proportioning parameter fields, uniform discretization is performed within their respective value ranges to obtain sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio. Enumerating all finite quinary parameter combinations on the Cartesian product of the sets of cement dosage, mineral powder dosage, fly ash dosage, water-cement ratio, and sand ratio yields the complete set of finite quinary parameter combinations. ; The total amount of cementitious materials is calculated based on the cement content, mineral powder content, and fly ash content. The water consumption per unit volume is then calculated based on the product of the total amount of cementitious materials and the water-cement ratio. ; Based on cement usage, mineral powder content, fly ash content, and water consumption per unit volume. Calculate the amount of sand used Dosage of He Shi ; Based on the complete set of finite five-element parameter combinations Water consumption per unit volume, sand consumption Dosage of He Shi Forming a set of candidate concrete mix design schemes ; Set of candidate concrete mix proportions The proportioning parameters of each candidate mix design are substituted into the performance prediction model to obtain the predicted values ​​of compressive strength, slump, adiabatic temperature rise, and chloride ion diffusion coefficient. Based on the heat dissipation correction coefficient of the target project, the predicted internal and external temperature difference of the candidate mix design is calculated.

7. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 6, characterized in that: The formula for calculating the predicted internal and external temperature difference is as follows: in, This indicates the predicted temperature difference between inside and outside. This represents the heat dissipation correction factor. This indicates the predicted peak adiabatic temperature rise. This represents the combined thermal conductivity related to the thermal diffusivity of concrete and the geometry of the structure. This represents the minimum structural cross-sectional dimension of the target project. This represents the overall thermal resistance of the insulation and maintenance measures for the target project. Indicates the ambient temperature during the construction period. This indicates the correction amount for the temperature calculation baseline.

8. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 7, characterized in that: Step S5 specifically includes: Set of candidate concrete mix proportions Each candidate formulation was screened, and the predicted compressive strength was retained. Not lower than the formulation strength The candidate solutions are then used to obtain the first subset of candidate solutions; The subset of first candidate solutions is filtered by workability constraints, retaining the predicted slump. Falling below the lower limit of the slump required for construction and upper limit Among the candidate solutions, a subset of the second candidate solutions is obtained; The subset of second candidate schemes was screened using temperature control constraints, retaining the predicted internal and external temperature differences. The candidate schemes that do not exceed the temperature control limit of 25℃ for large-volume concrete are used to obtain the third candidate subset. The third candidate subset was screened using durability constraints, retaining the predicted chloride ion diffusion coefficient. Not exceeding the upper limit of the durability design of the target project The candidate schemes were analyzed to obtain the fourth candidate subset. ; If the fourth candidate subset If not empty, then start from the fourth candidate subset. The recommended concrete mix design is determined in the middle; if the fourth candidate subset If the value is empty, it is determined that there is no feasible allocation scheme that simultaneously satisfies all four constraints within the current range of design variable values. A prompt message is output, and the range of design variable values ​​is adjusted or the constraints are relaxed.

9. The method for optimizing the concrete mix proportion of wind turbine foundations based on data analysis as described in claim 8, characterized in that: The fourth candidate subset The recommended concrete mix design is determined in the middle, specifically including: With the optimization objective of minimizing the total cost of concrete per cubic meter, the fourth candidate subset is selected. The material parameters for each candidate scheme are combined with the market unit price of each component to establish a total material cost objective function. : in, This indicates the unit price of cement. This indicates the unit price of mineral powder. This indicates the unit price of fly ash. This indicates the unit price of water. This indicates the unit price of sand. Indicates the unit price of the stone; For the fourth candidate subset The objective function value is calculated for each of the candidate solutions, and the global enumeration method is used to select the fourth candidate subset. The minimum value of the objective function is obtained, and the candidate mix design corresponding to the minimum value of the objective function is taken as the recommended concrete mix design.

10. A data analysis-based concrete mix design optimization system for wind turbine foundations, characterized in that, The system is used to implement the data analysis-based concrete mix design optimization method for wind turbine foundations as described in any one of claims 1-9, including: The data acquisition module is used to collect historical concrete mix ratio records from the historical concrete mix ratio database of wind turbine foundation projects, preprocess the historical concrete mix ratio records, and obtain the mix ratio dataset. The weight calculation module is used to evaluate the scene similarity between each historical concrete mix ratio record in the mix ratio dataset and the scene of the target wind turbine foundation project, based on the scene similarity, and to calculate the weight of each historical concrete mix ratio record to obtain the sample weight of each historical concrete mix ratio record. The model building module is used to perform weighted modeling of the mix proportion dataset based on the sample weights of each historical concrete mix proportion record, and obtain a performance prediction model with mix proportion parameters as independent variables. The temperature difference calculation module is used to discretize and generate a set of candidate concrete mix design schemes within the range of values ​​of each mix design variable. Each candidate scheme in the set of candidate concrete mix design schemes is input into the performance prediction model to obtain the performance prediction value of each candidate scheme. Based on the heat dissipation correction coefficient corresponding to the target wind turbine foundation project, the predicted internal and external temperature difference of each candidate scheme is calculated. The scheme selection module is used to select candidate concrete mix design schemes based on the predicted internal and external temperature differences, and based on strength constraints, workability constraints, durability constraints, and temperature control constraints, to obtain recommended concrete mix design schemes.

Images