A method and system for intensity conversion using micro-core samples in UHPC
By constructing a mathematical model for predictive conversion coefficients that considers multiple feature parameters, and using the differential evolution algorithm and the finite memory quasi-Newton method for parameter inversion, the problem of low intensity detection accuracy of UHPC micro core samples was solved, and high-precision intensity conversion was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TIANJIN PORT ENG INST LTD OF CCCC FIRST HARBOR ENG
- Filing Date
- 2026-04-28
- Publication Date
- 2026-06-30
Smart Images
Figure CN122108774B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of concrete strength testing technology, specifically relating to a method and system for strength conversion using a micro-core sample method for UHPC. Background Technology
[0002] In-situ strength testing of concrete structures is an important method for assessing engineering quality and a direct basis for determining the actual strength of the structure. Core drilling, a commonly used method for in-situ strength testing, involves directly drilling a concrete cylindrical core sample for compressive strength testing. The compressive strength of the core sample is then converted to the compressive strength of a standard cubic specimen, thereby evaluating the strength of the concrete structure.
[0003] With the development of construction engineering, people have higher requirements for structural safety and seismic resistance. On the one hand, this leads to larger reinforcement ratios and smaller rebar spacing in concrete structures. To avoid touching the rebars during strength testing, core sampling with micro-cores is necessary. On the other hand, ultra-high performance concrete (Ultra High Performance Concrete)... High Performance Concrete (UHPC) is increasingly widely used in modern construction engineering due to its excellent strength, toughness, durability, self-compacting properties, and environmental friendliness and energy saving. In practical applications, UHPC structures are designed to be relatively thin and light. In order to reduce the impact of strength testing on the UHPC structure, it is necessary to take micro-core samples. The diameter of the micro-core samples is usually no more than 50mm.
[0004] However, current standards and specifications for concrete strength testing lack a specific strength conversion method for the UHPC micro-core sample method. Existing in-situ strength testing techniques using core drilling are mostly suitable for ordinary concrete with strength grades of C100 and below, and are not applicable to UHPC strength testing. Furthermore, the core samples obtained using current core drilling methods are relatively large (typically 70mm–100mm), which can significantly damage the UHPC structure. Moreover, existing strength testing methods generally use fixed conversion factors, establishing a linear relationship between the core sample compressive strength and the standard specimen compressive strength. While this is simple to operate, it fails to consider the coupling effects of multiple characteristic parameters of the core sample (core sample size and material parameters, etc.), making it impossible to accurately establish the conversion relationship between the UHPC micro-core sample compressive strength and the standard specimen compressive strength, thus resulting in excessively low accuracy in UHPC structural strength testing.
[0005] In addition, Chinese patent CN120741145A discloses a technical method for testing the strength of ultra-high performance concrete using a 52mm diameter core drilling method. The method corrects the compressive strength of the core sample by multiplying the measured compressive strength by a conversion factor. However, the conversion factor only includes the steel fiber content conversion factor and the size effect conversion factor, and these two conversion factors are also taken as fixed values. Therefore, it does not truly take into account the nonlinear relationship of multi-factor coupling, and cannot meet the high-precision strength testing requirements of ultra-high performance material structures such as UHPC. Summary of the Invention
[0006] To address the shortcomings of related technologies, this invention provides a method and system for strength conversion using the micro-core sample method for UHPC, aiming to accurately establish the conversion relationship between the compressive strength of micro-core samples and the compressive strength of standard specimens, thereby improving the accuracy of UHPC solid structure strength testing.
[0007] This invention provides a method for intensity conversion using a small core sample method for UHPC, comprising the following steps:
[0008] S1. Prepare multiple different micro-core samples, and cast multiple standard specimens adapted to them at the same time, under the same preparation process and environmental conditions; the feature vector of the micro-core sample. ,in, This is the sample number for the micro-core sample. The diameter of the core sample. The aspect ratio is height to diameter. This refers to the water-to-glue ratio. This refers to the fiber volume content. The aspect ratio of the fiber;
[0009] S2. Perform compressive strength tests on all micro-core samples and standard specimens, and calculate the measured conversion factor between the compressive strength of each micro-core sample and the compressive strength of the standard specimen. ;
[0010] S3. Constructing Predictive Conversion Factors The mathematical model is expressed as equation (1); where, , , , These are the diameter factor, aspect ratio geometric constraint factor, material factor, and fiber reinforcement factor, respectively, and are expressed as equations (2) to (5). This is the diameter effect coefficient. The preset reference core sample diameter, These are the limit constraint coefficients. To constrain the attenuation coefficient, The preset reference height-to-diameter ratio, For feature size, The preset baseline water-to-glue ratio, This is the size effect coefficient. The water-cement ratio influence coefficient is... The maximum fiber reinforcement factor, Determine the critical fiber parameters; determine the parameter vector to be calibrated in the mathematical model. and the range of values for each parameter;
[0011] (1);
[0012] (2);
[0013] (3);
[0014] (4);
[0015] (5);
[0016] S4. Parameter vectors in the mathematical model Calibration includes,
[0017] Construct the objective function , expressed as equation (6); where, For the fitness function, This represents the total number of microcore samples. The preset penalty coefficient, This is a physical constraint penalty term;
[0018] (6);
[0019] The parameter inversion is performed using the differential evolution algorithm and the finite memory quasi-Newton method to determine the parameter vector. The calibration values are used to update the prediction conversion factors. Mathematical model;
[0020] S5. Take small core samples from the on-site concrete structure and conduct compressive strength tests on them to obtain compressive strength. Substitute the characteristics of the tiny core sample into the updated mathematical model to calculate the prediction conversion factor. The compressive strength of the standard specimen is calculated according to formula (7). ;
[0021] (7).
[0022] In some embodiments, in step S1, the microcore sample is prepared by laboratory casting or core sampling from a concrete structure.
[0023] In some embodiments, in step S2, the measured conversion factor is... The calculation includes the following steps:
[0024] S21. Perform a compressive strength test on each micro core sample and each standard specimen sample to obtain a set of compressive strength data, each set of compressive strength data including at least 4 compressive strength values.
[0025] S22. Use the outlier detection method to test each group of compressive strength data; if there are no outliers, take the average value of the compressive strength data of the group as the representative value of compressive strength; if there is one outlier, take the average value of the compressive strength data of the group after removing the outlier as the representative value of compressive strength; if there are two or more outliers, conduct the compressive strength test again to supplement the data volume of the group of compressive strength data, and re-detect outliers and recalculate the representative value of compressive strength.
[0026] S23. Calculate the measured conversion factor according to formula (8). ,in, This represents the compressive strength of a small core sample. This represents the compressive strength of a standard specimen.
[0027] (8).
[0028] In some embodiments, in step S4, the physical constraint penalty term The calculation is performed according to equation (9), where, This is the preset lower limit of the prediction conversion factor. This is the preset upper limit of the prediction conversion factor;
[0029] (9).
[0030] In some embodiments, in step S4, the differential evolution algorithm is used to process the parameter vector. Perform global optimization to output parameter vector The initial optimization values specifically include the following operations:
[0031] Population initialization, in the parameter vector Randomly generated in the eight-dimensional parameter space Individual of generation 0 , ,in, For individual identification numbers, ;
[0032] The mutation operation uses the best / 1 / bin strategy for the first... For each individual Generate mutation vectors ;
[0033] Crossover operation, on the mutated vector and the original individual Perform a binary crossover to generate the test vector. ,in, For dimension indexing;
[0034] Select operation, compare the original individuals With test vector The fitness function value is used to select the optimal individual for the next generation. ;
[0035] The iteration terminates when the number of iterations reaches the preset maximum or the improvement in optimal fitness over multiple consecutive generations is less than the preset convergence threshold. The current optimal individual, i.e., the parameter vector, is then output. The initial optimization results.
[0036] In some embodiments, in step S4, the finite memory quasi-Newton method is used to process the parameter vector. Local optimization is performed on the initial optimization results, specifically including using the parameter vector. The initial optimization result is taken as the initial iteration point. The parameters are iteratively updated using a quadratic approximation model of the objective function near the current initial iteration point, and the gradient of the objective function at the current point is calculated. Optimization terminates when the gradient norm is less than a preset threshold, the parameter change is less than a preset tolerance, or the optimization algebra reaches a preset maximum number of iterations, and the parameter vector is output. The final optimization result is determined as a parameter vector. The calibration value.
[0037] This invention also provides a micro-core sample strength conversion system for UHPC, used to execute the aforementioned micro-core sample strength conversion method for UHPC. The micro-core sample strength conversion system for UHPC includes:
[0038] The test module is used to perform compressive tests on micro core samples, standard specimen samples, and micro core samples obtained from field sampling in order to obtain compressive strength.
[0039] The model building module is used to construct a mathematical model based on the characteristics of the micro core sample, including the prediction conversion coefficients such as diameter factor, aspect ratio geometric constraint factor, material factor and fiber reinforcement factor.
[0040] The first calculation module is used to calculate the measured conversion factor based on the compressive strength of each micro core sample and the compressive strength of the standard specimen sample that is compatible with it.
[0041] The second calculation module is used to perform parameter inversion based on the objective function, using the differential evolution algorithm and the finite memory quasi-Newton method, to determine the calibration value of the parameter vector, and update the mathematical model of the prediction conversion coefficient accordingly.
[0042] The third calculation module is used to calculate the prediction conversion factor based on the characteristics of the micro core sample obtained from the field sampling and the updated mathematical model, and then combine it with the compressive strength of the micro core sample to calculate the compressive strength of the standard specimen.
[0043] Based on the above technical solution, the micro-core sample strength conversion method and system for UHPC in this embodiment of the invention considers the coupled influence of multiple characteristic parameters of micro-core samples (core sample diameter, height-to-diameter ratio, water-to-binder ratio, fiber volume content, and fiber length-to-diameter ratio) on the strength conversion coefficient. It constructs a mathematical model for predicting conversion coefficients, including diameter factor, height-to-diameter ratio geometric constraint factor, material factor, and fiber reinforcement factor. Based on the measured conversion coefficients of the samples, it calibrates the parameters in the mathematical model using a global optimization and local optimization approach, thereby accurately establishing the conversion relationship between the compressive strength of the micro-core sample and the compressive strength of the standard specimen. This enables accurate conversion from the actual compressive strength of a cylindrical micro-core sample to the compressive strength of a cubic standard specimen, improving the accuracy of UHPC solid structure strength testing. Attached Figure Description
[0044] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:
[0045] Figure 1 This is a flowchart of the micro-core sample strength conversion method for UHPC according to the present invention;
[0046] Figure 2 This is a schematic diagram illustrating the conversion factor obtained using existing strength testing methods.
[0047] Figure 3 This is a comparison chart of the fixed conversion factor obtained using existing strength testing methods and the measured conversion factor for different micro-core samples.
[0048] Figure 4 This is a comparison chart of the predicted conversion coefficients and the measured conversion coefficients obtained using this invention. Detailed Implementation
[0049] The technical solutions in 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 a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0050] In the description of this invention, it should be understood that the terms "center", "lateral", "longitudinal", "upper", "lower", "top", "bottom", "inner", "outer", "left", "right", "front", "rear", "vertical", "horizontal", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.
[0051] The terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature.
[0052] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0053] refer to Figure 1 As shown, the present invention provides a method for intensity conversion of micro-core samples for UHPC, including the following steps S1 to S5.
[0054] Step S1: Prepare multiple different micro-core samples, and simultaneously cast multiple matching standard specimens under the same preparation process and environmental conditions. Feature vector of the micro-core sample. ,in, This is the sample number for the micro-core sample. The diameter of the core sample. The height-to-diameter ratio (the ratio of the core sample's height to its diameter). This refers to the water-to-glue ratio. This refers to the fiber volume content. The aspect ratio is the fiber length-to-diameter ratio; it should be noted that each feature includes multiple variables to form multiple different micro-core samples.
[0055] Specifically, micro-core samples can be prepared using laboratory casting or core sampling from concrete structures; the casting and drilling of core samples should be carried out in accordance with the standard operating procedures. Micro-core samples are cylindrical, while standard specimens are cubic, with dimensions of 100x100x100mm or 150x150x150mm; the curing conditions for standard specimens and micro-core samples are the same, or both are cured under standard conditions.
[0056] Step S2: Perform compressive strength tests on all micro-core samples and standard specimen samples, and calculate the measured conversion factor between the compressive strength of each micro-core sample and the compressive strength of the standard specimen sample. Further explanation: The measured conversion factor... The calculation includes the following steps S21 to S23.
[0057] Step S21: Perform a compressive strength test on each micro core sample and each standard specimen sample to obtain a set of compressive strength data, each set of compressive strength data including at least 4 compressive strength values.
[0058] Step S22: Use the outlier detection method to detect each group of compressive strength data; if there are no outliers, the average value of the compressive strength data in the group is taken as the representative value of compressive strength; if there is one outlier, the average value of the compressive strength data in the group after removing the outlier is taken as the representative value of compressive strength; if there are two or more outliers, the compressive strength test is performed again to supplement the amount of data in the group of compressive strength data, and the outlier detection and the calculation of the representative value of compressive strength are performed again.
[0059] Briefly explain the basic principle of outlier testing: First, calculate the statistic according to the following formula. ;in, To remove suspicious values After, the remaining The average of the data; These are suspected outliers in the data set, typically the maximum or minimum value. To remove suspicious values After, the remaining The standard deviation of each data point The number of data items in the data group;
[0060]
[0061] The calculated statistics and the corresponding critical value Compare; if Then it is believed These are outliers and should be removed; if Then it is believed Not an outlier, should be retained; critical value You can refer to Table 1 for selection.
[0062] Table 1 Critical Values
[0063]
[0064] Step S23: Calculate the measured conversion factor according to formula (8). ,in, This represents the compressive strength of a small core sample. The representative value of the compressive strength of a standard specimen that is compatible with this micro-core sample;
[0065] (8).
[0066] Step S3: Construct the prediction conversion coefficients The mathematical model is expressed as equation (1); where, , , , These are, respectively, diameter factor, aspect ratio geometric constraint factor, material factor, and fiber reinforcement factor;
[0067] (1).
[0068] diameter factor Represented as equation (2); where, The core sample diameter (in mm) In this embodiment, the preset reference core sample diameter (in mm) is used. Take 30mm; The diameter effect coefficient is usually negative, indicating that the strength decreases as the diameter increases. The diameter factor expression is constructed considering that the diameter effect mainly stems from the defect statistical effect and stress gradient effect. Larger specimens have a higher probability of containing more severe defects, and the internal stress distribution is more complex.
[0069] (2).
[0070] Aspect Ratio Geometric Constraint Factor Represented as equation (3); where, The aspect ratio is height to diameter. In this embodiment, the reference height-to-diameter ratio is set as a preset value. Set the value to 1.0; The limit constraint coefficient represents the relative strength at infinite height-to-diameter ratio; The constraint attenuation coefficient describes the rate attenuation of constraint as the height-to-diameter ratio increases. The expression for the height-to-diameter ratio geometric constraint factor is constructed considering that the end constraint of the concrete specimen weakens as the height-to-diameter ratio increases, leading to a decrease in strength. The exponential function form conforms to the physical law that the constraint effect decreases with distance.
[0071] (3).
[0072] Material factor Represented as equation (4); where, This refers to the water-to-glue ratio. In this embodiment, the preset baseline water-cement ratio is used. Take 0.18; For feature size, This is the size effect coefficient. The water-binder ratio influence coefficient describes the nonlinear effect of the water-binder ratio on strength; the material factor expression is constructed considering the Bažant size effect law and the nonlinear effect of the water-binder ratio on strength.
[0073] (4).
[0074] Fiber reinforcing factor Represented as equation (5); where, Fiber volume content (%) The aspect ratio of the fiber. The maximum fiber reinforcement coefficient represents the upper limit of strength improvement when the fiber content is infinite; The critical fiber parameter controls the saturation rate of fiber reinforcement; the expression for the fiber reinforcement factor is constructed by comprehensively considering the coupling effect of fiber content and aspect ratio.
[0075] (5).
[0076] Furthermore, determine the prediction conversion factor. The parameter vector to be calibrated in the mathematical model ,in, The transpose sign is used; the parameter vector is determined. The range and initial values of each parameter can be found in Table 2. The initial values can be given based on experience or taken as the midpoint of the range.
[0077] Table 2 Prediction Conversion Factors The range of values and initial values of each parameter in the mathematical model.
[0078]
[0079] Step S4: Adjust the prediction conversion coefficients Parameter vector in the mathematical model The calibration process includes the following steps S41 to S42.
[0080] Step S41: Construct the objective function , expressed as equation (6); where, For the fitness function, This represents the total number of microcore samples. The preset penalty coefficient, This is a physical constraint penalty term;
[0081] (6);
[0082] objective function The construction aims to find the optimal parameter vector. Make the fitness function Minimize; specifically, the first term on the right-hand side of the equation is the data fitting term, which measures the prediction conversion factor. Conversion factor with actual measurement The root mean square error (RMSE) represents the model's ability to fit the data; the second term on the right-hand side of the equation is a penalty term, which introduces physical constraints to prevent the parameters from having solutions that do not conform to engineering principles. This is a preset penalty coefficient, which can be set to 100, used to balance the weights of data fitting and physical constraints.
[0083] To further explain, physical constraint penalty items The calculation is performed according to equation (9), where, This is the preset lower limit of the prediction conversion factor. This is a preset upper limit for the prediction conversion factor; in this embodiment, it can be set to... , ;
[0084] (9).
[0085] Step S42: Perform parameter inversion using the Differential Evolution (DE) algorithm and the Finite Memory Quasi-Newton method (L-BFGS-B) to determine the parameter vector. The calibration values are used to update the prediction conversion factors. The mathematical model.
[0086] To further explain, the differential evolution algorithm (DE) is used to process the parameter vector. Perform global optimization to output parameter vector The initial optimization value is determined by global optimization, which specifically includes population initialization, mutation operation, crossover operation, selection operation, etc.
[0087] Population initialization, in the parameter vector Randomly generated in the eight-dimensional parameter space Individual of generation 0 , , It is generated according to formula (10); where, For individual identification numbers, ; For interval Independent uniform random numbers within the range, and The parameter boundary vector is used to limit the physical range of each parameter; the core logic of equation (10) is to define the upper and lower bounds of the parameters. and Initial solutions are generated randomly using a uniform distribution between them;
[0088] (10).
[0089] The mutation operation employs the best / 1 / bin strategy, and is performed according to equation (11) on the first... For each individual Generate mutation vectors ;in, This refers to the individual with the best fitness in the current population. and They are distinct and not equal to Random indexes are used to ensure the randomness of the difference direction. ; and They are distinct and not equal to Random individuals; To control the scaling factor of the differential perturbation amplitude, To avoid step size being too large or too small; the core logic of equation (11) is to use the current optimal individual Based on this, a new mutation direction is generated by adding differential perturbations from two random individuals.
[0090] (11).
[0091] Crossover operation, according to equation (12) on the mutation vector and the original individual Perform a binary crossover to generate the test vector. To enhance population diversity; among them, For crossover probability, ; For dimensional indexing, The dimension index is randomly selected to ensure that at least one dimension of the experimental vector comes from the mutation vector, thus avoiding the experimental vector being completely identical to the original individual. The core logic of Equation (12) is to inherit the dimension from the mutation vector with a certain probability, retain some characteristics of the original individual, and enhance the diversity of the population.
[0092] (12).
[0093] Select the operation and compare the original individuals according to equation (13). With test vector The fitness function value is used to select the optimal individual for the next generation. The core logic of equation (13) is that only when the fitness function value of the experimental vector is better will the original individual be replaced and enter the next generation, so as to ensure that the overall fitness of the population continues to improve.
[0094] (13).
[0095] Iteration terminates when the number of iterations reaches a preset maximum (e.g., 500) or the optimal fitness improvement over multiple consecutive iterations is less than a preset convergence threshold (e.g., 10). -6 When the iteration terminates, output the current best individual. That is, parameter vector The initial optimization results.
[0096] To further explain, the finite memory quasi-Newton method (L-BFGS-B) is used for the parameter vector. The initial optimization results are used for local optimization, specifically including the parameter vector output by the differential evolution algorithm (DE). The initial optimization result is used as the initial iteration point. The parameters are iteratively updated using a quadratic approximation model of the objective function near the current initial iteration point, and the gradient of the objective function at the current point is calculated. It can be understood that L-BFGS-B only stores the most recent... The correction in the next iteration (usually) Specifically, this includes operations such as iterating vectors and calculating search directions.
[0097] Iteration vector, let For the first The parameter vector for the next iteration is updated according to equation (14), where, For the search direction, The step size is determined by line search;
[0098] (14).
[0099] The search direction is calculated according to equation (15). ,in, It is an approximation of the Hessian inverse matrix, calculated using a stored vector; The gradient of the objective function at the current point is calculated numerically using the central difference method, and is expressed as equation (16), where, It is a unit vector. For minute perturbations, ;
[0100] (15);
[0101] (16).
[0102] The calculation terminates when any one of the following three conditions is met: the gradient norm is less than a preset threshold, the parameter change is less than a preset tolerance, or the number of optimization algebras reaches a preset maximum number of iterations (e.g., 1000). The parameter vector is then output. The final optimization result is determined as a parameter vector. The calibration value.
[0103] Specifically, a gradient norm less than a preset threshold is expressed as: , Can be taken as This indicates that the gradient is small enough and the parameters are close to the extreme point; the parameter change is less than the preset tolerance, which is expressed as... , Can be taken as This indicates that the parameter updates are minimal and the solution has stabilized.
[0104] Step S5: Take a small core sample from the concrete structure on site and conduct a compressive strength test on the core sample to obtain its compressive strength. The characteristics of this tiny core sample (i.e., the core sample diameter) Height-to-diameter ratio Water-to-glue ratio Fiber volume content Fiber aspect ratio Substitute these values into the updated mathematical model to calculate the prediction conversion coefficients. ;Will and Combined, the compressive strength of the standard specimen is calculated according to equation (7). This allows for the assessment of the strength of the concrete structure on site.
[0105] (7).
[0106] The above illustrative embodiment considers the coupled influence of multiple characteristic parameters of the micro-core sample (core sample diameter, aspect ratio, water-cement ratio, fiber volume content, and fiber length-to-diameter ratio) on the strength conversion coefficient. A mathematical model for predicting conversion coefficients is constructed, including diameter factor, aspect ratio geometric constraint factor, material factor, and fiber reinforcement factor. Based on the measured conversion coefficients of the samples, the parameters in the mathematical model for predicting conversion coefficients are calibrated by using a global optimization superposition with local optimization. This makes the predicted conversion coefficients directly related to the multiple characteristic parameters of the micro-core sample, thereby accurately establishing the conversion relationship between the compressive strength of the micro-core sample and the compressive strength of the standard specimen. This enables accurate conversion from the actual compressive strength of the cylindrical micro-core sample to the compressive strength of the cubic standard specimen, improving the accuracy of UHPC solid structure strength testing.
[0107] The present invention also provides a micro-core sample strength conversion system for UHPC, for performing the aforementioned micro-core sample strength conversion method for UHPC. The strength conversion system includes an experimental module, a model building module, a first calculation module, a second calculation module, and a third calculation module. The experimental module is used to conduct compressive strength tests on micro-core samples, standard specimen samples, and micro-core samples obtained from field sampling to obtain compressive strength. The model building module is used to construct a mathematical model for predicting conversion coefficients based on the characteristics of the micro-core samples, including diameter factor, aspect ratio geometric constraint factor, material factor, and fiber reinforcement factor. The first calculation module is used to calculate the measured conversion coefficients based on the compressive strength of each micro-core sample and the compressive strength of the standard specimen sample that matches it. The second calculation module is used to perform parameter inversion based on the objective function, using the differential evolution algorithm and the finite memory quasi-Newton method to determine the calibration values of the parameter vector, and update the mathematical model for predicting conversion coefficients accordingly. The third calculation module is used to calculate the predicted conversion coefficients based on the characteristics of the micro-core samples obtained from field sampling and the updated mathematical model, and then combine them with the compressive strength of the micro-core sample to calculate the compressive strength of the standard specimen.
[0108] The technical effects of the present invention are described below based on specific embodiments:
[0109] Multiple different micro-core samples and multiple standard samples adapted to them were prepared; the characteristic variables of the micro-core samples are as follows: core diameter... Choose 30mm, 40mm, and 50mm, with a height-to-diameter ratio. Use water-to-glue ratios of 1, 1.5, 2, 2.5, and 3. Fiber volume fractions of 0.22, 0.2, 0.18, 0.17, and 0.16 were used. Take 0.5%, 1%, 1.5%, 2%, and 2.5% fiber aspect ratios. Take 65; perform compressive strength tests on all micro-core samples and standard specimens, and obtain the representative compressive strength value of each micro-core sample. Representative compressive strength of a standard specimen compatible with this micro-core sample Calculate the measured conversion factor .
[0110] Based on the mathematical model and objective function for predicting conversion coefficients constructed in this invention, the parameter vector is determined after parameter inversion using the differential evolution algorithm and the finite memory quasi-Newton method. The calibration values are [-0.1370, 0.85534, 0.6153, 11.6595, 0.1440, 0.5615, 0.30285, 1.3609], based on which a new mathematical model predicts the conversion factors; the specific characteristics of each micro-core sample are substituted into the updated mathematical model, combined with its representative compressive strength value. The prediction conversion factor for each microcore sample was calculated. As shown in Table 3 (only a portion of the data is shown here).
[0111] Table 3. Prediction conversion coefficients obtained using the present invention Conversion factor with actual measurement Comparison
[0112]
[0113] As can be seen from Table 3, the prediction conversion coefficients based on this specific embodiment... Conversion factor with actual measurement The deviation is between 8% and -6%, which meets the general requirement of ±10% for prediction deviation. Understandably, if the number of microcore samples increases, the prediction deviation will further decrease. (Reference) Figure 4 As shown in the figure, the purple line represents the case where the predicted conversion factor equals the measured conversion factor, and the red dots represent the conversion factors obtained based on different micro-core samples. The horizontal axis of each red dot is the measured conversion factor. The vertical axis represents the prediction conversion factor. The closer the red dot is to the purple line, the closer the predicted conversion factor is to the actual conversion factor, and the higher the accuracy of the prediction result.
[0114] However, if the existing strength testing method described in the background art is used to obtain the representative value of the compressive strength of a small core sample... and representative value of compressive strength of standard specimen samples Then, directly through simple linear fitting methods (such as...) Figure 2(As shown) to obtain the conversion factor, which is a fixed value of 0.8667; Reference Figure 3 As shown in the figure, the purple line represents the fixed conversion factor of 0.8667 obtained by the existing method, and the red dots represent the measured conversion factors for different micro-core samples. It is evident that the consistency between the fixed conversion factor and the measured conversion factor in existing strength testing methods is very poor, with a deviation as high as 29% to -16%. Therefore, while the existing strength testing methods, which establish a linear relationship between the core sample compressive strength and the standard specimen compressive strength by using a fixed conversion factor, are relatively simple, they do not consider the nonlinearity of multiple characteristic parameter couplings, resulting in excessively low strength testing accuracy. Furthermore, using commonly used evaluation indicators in the industry... (Determination coefficient) and The root mean square error (RMSE) is used to evaluate the results of existing strength testing methods and the method of this invention; as is well known to those skilled in the art, This indicates how well the model fits the data. , The closer the value is to 1, the better the model fits the data. Used to measure the prediction accuracy of the model. The smaller the value, the higher the model's prediction accuracy; existing intensity detection methods , (like Figure 2 (as shown), and the present invention , (like Figure 4 As shown), it can be seen that although the present invention Slightly lower than existing strength testing methods, but The conversion factor is significantly reduced, thus the present invention can significantly improve the prediction accuracy of the conversion factor, and thus improve the detection accuracy of the UHPC solid structure strength.
[0115] Through the description of several embodiments of the micro-core sample strength conversion method and system for UHPC of the present invention, it can be seen that the present invention has at least one or more of the following advantages:
[0116] 1) By preparing and testing multiple different micro-core samples and their matching standard specimens, the measured conversion coefficient between each micro-core sample and the standard specimen is obtained, providing a basis for the parameter inversion and calibration of the subsequent prediction conversion system mathematical model.
[0117] 2) A mathematical model for predicting conversion coefficients, including diameter factor, aspect ratio geometric constraint factor, material factor and fiber reinforcement factor, is constructed. Considering the coupled influence of multiple characteristic parameters of micro core samples (core sample diameter, aspect ratio, water-binder ratio, fiber volume content and fiber length-to-diameter ratio) on the strength conversion coefficients, it is beneficial to accurately establish the conversion relationship between the compressive strength of micro core samples and the compressive strength of standard specimens.
[0118] 3) Construct an objective function and use a global optimization superimposed with a local optimization method to perform parameter inversion on the parameters to be calibrated in the mathematical model, determine the parameter calibration values, and obtain the updated prediction conversion coefficient mathematical model. This makes the prediction conversion coefficient directly related to the multiple characteristic parameters of the micro core sample rather than a fixed value, thereby accurately establishing the conversion relationship between the compressive strength of the micro core sample and the compressive strength of the standard specimen. Therefore, when a micro core sample is taken from the concrete structure on site, its characteristics can be substituted into the mathematical model to accurately obtain the prediction conversion coefficient. Then, by combining it with the compressive strength of the micro core sample, the compressive strength of the standard specimen can be accurately obtained, thus improving the accuracy of the UHPC solid structure strength test.
[0119] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0120] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.
Claims
1. A method for intensity conversion using micro-core samples in UHPC, characterized in that, Includes the following steps: S1. Prepare multiple different micro-core samples, and cast multiple standard specimens adapted to them at the same time, under the same preparation process and environmental conditions; the feature vector of the micro-core sample. ,in, This is the sample number for the micro-core sample. The diameter of the core sample. The aspect ratio is height to diameter. This refers to the water-to-glue ratio. This refers to the fiber volume content. The aspect ratio of the fiber; S2. Perform compressive strength tests on all micro-core samples and standard specimens, and calculate the measured conversion factor between the compressive strength of each micro-core sample and the compressive strength of the standard specimen. ; S3. Constructing Predictive Conversion Factors The mathematical model is expressed as equation (1); where, , , , These are the diameter factor, aspect ratio geometric constraint factor, material factor, and fiber reinforcement factor, respectively, and are expressed as equations (2) to (5). This is the diameter effect coefficient. The preset reference core sample diameter, These are the limit constraint coefficients. To constrain the attenuation coefficient, The preset reference height-to-diameter ratio, For feature size, The preset baseline water-to-glue ratio, This is the size effect coefficient. The water-cement ratio influence coefficient is... The maximum fiber reinforcement factor, Determine the critical fiber parameters; determine the parameter vector to be calibrated in the mathematical model. and the range of values for each parameter; (1)； (2)； (3)； (4)； (5)； S4. Parameter vectors in the mathematical model Calibration is performed, including, Construct the objective function , expressed as equation (6); where, For the fitness function, This represents the total number of microcore samples. The preset penalty coefficient, This is a physical constraint penalty term; (6)； The parameter inversion is performed using the differential evolution algorithm and the finite memory quasi-Newton method to determine the parameter vector. The calibration values are used to update the prediction conversion factors. Mathematical model; S5. Take small core samples from the on-site concrete structure and conduct compressive strength tests on them to obtain compressive strength. Substitute the characteristics of the tiny core sample into the updated mathematical model to calculate the prediction conversion factor. The compressive strength of the standard specimen is calculated according to formula (7). ; (7)。 2. The method for intensity conversion using micro-core samples for UHPC according to claim 1, characterized in that, In step S1, the micro-core samples are prepared by laboratory casting or core sampling from concrete structures.
3. The method for intensity conversion using micro-core samples for UHPC according to claim 1, characterized in that, In step S2, the measured conversion factor The calculation includes the following steps: S21. Perform a compressive strength test on each micro core sample and each standard specimen sample to obtain a set of compressive strength data, each set of compressive strength data including at least 4 compressive strength values. S22. Use the outlier detection method to test each group of compressive strength data; if there are no outliers, take the average value of the compressive strength data of the group as the representative value of compressive strength; if there is one outlier, take the average value of the compressive strength data of the group after removing the outlier as the representative value of compressive strength; if there are two or more outliers, conduct the compressive strength test again to supplement the data volume of the group of compressive strength data, and re-detect outliers and recalculate the representative value of compressive strength. S23. Calculate the measured conversion factor according to formula (8). ,in, This represents the compressive strength of a small core sample. This represents the compressive strength of a standard specimen. (8)。 4. The method for intensity conversion using micro-core samples for UHPC according to claim 1, characterized in that, In step S4, the physical constraint penalty term The calculation is performed according to equation (9), where, This is the preset lower limit of the prediction conversion factor. This is the preset upper limit of the prediction conversion factor; (9)。 5. The method for intensity conversion using micro-core samples for UHPC according to claim 1, characterized in that, In step S4, the differential evolution algorithm is used to process the parameter vector. Perform global optimization to output parameter vector The initial optimization values specifically include the following operations: Population initialization, in parameter vector Randomly generated in the eight-dimensional parameter space Individual of generation 0 , ,in, For individual identification numbers, ; The mutation operation uses the best / 1 / bin strategy for the first... For each individual Generate mutation vectors ; Crossover operation, on the mutated vector and the original individual Perform a binary crossover to generate the test vector. ,in, For dimension indexing; Select operation, compare the original individuals With test vector The fitness function value is used to select the optimal individual for the next generation. ; The iteration terminates when the number of iterations reaches the preset maximum or the improvement in optimal fitness over multiple consecutive generations is less than the preset convergence threshold. The current optimal individual, i.e., the parameter vector, is then output. The initial optimization results.
6. The method for intensity conversion using micro-core samples for UHPC according to claim 5, characterized in that, In step S4, the finite memory quasi-Newton method is used to process the parameter vector. Local optimization is performed on the initial optimization results, specifically including using the parameter vector. The initial optimization result is taken as the initial iteration point. The parameters are iteratively updated using a quadratic approximation model of the objective function near the current initial iteration point, and the gradient of the objective function at the current point is calculated. Optimization terminates when the gradient norm is less than a preset threshold, the parameter change is less than a preset tolerance, or the optimization algebra reaches a preset maximum number of iterations, and the parameter vector is output. The final optimization result is determined as a parameter vector. The calibration value.
7. A micro-core sample strength conversion system for UHPC, characterized in that, For performing the micro-core sample strength conversion method for UHPC as described in any one of claims 1 to 6, the micro-core sample strength conversion system for UHPC comprises: The test module is used to perform compressive strength tests on micro core samples, standard specimen samples, and micro core samples obtained from field sampling in order to obtain compressive strength. The model building module is used to construct a mathematical model based on the characteristics of the micro core sample, including the prediction conversion coefficients such as diameter factor, aspect ratio geometric constraint factor, material factor and fiber reinforcement factor. The first calculation module is used to calculate the measured conversion factor based on the compressive strength of each micro core sample and the compressive strength of the standard specimen sample that is compatible with it. The second calculation module is used to perform parameter inversion based on the objective function, using the differential evolution algorithm and the finite memory quasi-Newton method, to determine the calibration value of the parameter vector, and update the mathematical model of the prediction conversion coefficient accordingly. The third calculation module is used to calculate the prediction conversion factor based on the characteristics of the micro core sample obtained from the field sampling and the updated mathematical model, and then combine it with the compressive strength of the micro core sample to calculate the compressive strength of the standard specimen.