A method, medium and system for optimizing the ratio of smooth concrete surface in civil engineering of substation facilities

Abstract

The present invention provides a method, medium and system for optimizing the surface smoothness ratio of civil engineering concrete of substation facilities, belonging to the technical field of concrete ratio, including: firstly, collecting the physical and chemical property data of raw materials such as cement, sandstone, admixture, etc., establishing a concrete surface smoothness evaluation model, and comprehensively considering parameters such as surface roughness, flatness, glossiness, etc. On this basis, designing a small-scale indoor test to simulate engineering conditions, using computer image analysis to obtain the measured surface smoothness index under different ratios, and fitting and optimizing the model. Using computer-aided ratio design software to generate multiple groups of optimized ratios, using orthogonal test method to carry out pilot experiments, and further optimizing the fitting model. Finally, field tests are carried out in actual engineering, and the ratio with the best smoothness index is selected as the final optimization result. The technical problem that it is difficult to obtain the optimized ratio of concrete surface smoothness due to relying on the subjective experience and trial and error of engineering personnel is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of concrete mix ratio, and in particular, relates to a method, medium and system for optimizing the mix ratio of civil construction concrete surface smoothing of substation facilities. Background Art

[0002] As a common building material, the surface quality of concrete has an important impact on the appearance and performance of the project. In terms of substation facilities, the smoothness of the concrete surface is directly related to the construction difficulty and effect of subsequent decorative painting, anti-corrosion protection and other processes. A concrete surface that is too rough or uneven not only affects the appearance, but may also lead to problems such as reduced coating adhesion and easy corrosion. Therefore, how to improve surface smoothness by optimizing the concrete mix ratio has always been a key technical issue of concern to the industry.

[0003] The existing methods for optimizing the surface smoothness of concrete mainly focus on the following aspects: First, by adjusting the ratio of concrete raw materials, such as increasing the amount of fine aggregate and reducing the content of coarse aggregate, the surface roughness can be reduced; second, by using special construction techniques, such as precision vibration and fine troweling, to improve the surface flatness; third, after the concrete solidifies, post-processing methods such as manual grinding and grinding can be used to improve the surface finish. These methods can indeed improve the surface quality of concrete to a certain extent, but they also have some limitations:

[0004] First, the adjustment of raw material ratio and construction process often requires a lot of experimental accumulation and experience accumulation, which is time-consuming and costly. At the same time, these optimization methods usually cannot take into account other performance indicators of concrete, and it is difficult to find a truly optimal ratio.

[0005] Secondly, although artificial post-processing methods can improve surface smoothness, they require additional construction processes, which increases project costs, and it is difficult to accurately control the micro-roughness and macro-flatness of the surface.

[0006] Finally, most of the existing optimization methods rely on the subjective experience and trial and error of engineers, making it difficult to establish a quantitative and systematic surface quality evaluation system, and unable to quickly and accurately predict and optimize the smoothness of the concrete surface.

[0007] Therefore, there is an urgent need for a new optimization method that can accurately predict and optimize the concrete surface quality through scientific data analysis and model building, thereby greatly improving the surface smoothness and meeting the aesthetic and usage requirements of projects such as substation facilities. Summary of the invention

[0008] In view of this, the present invention provides a method, medium and system for optimizing the smoothing ratio of the civil construction concrete surface of a substation facility, which can solve the technical problem that most existing methods rely on the subjective experience and trial and error of engineers and it is difficult to obtain the optimized ratio of the smoothing of the concrete surface.

[0009] The present invention is achieved in that:

[0010] A first aspect of the present invention provides a method for optimizing the surface smoothing ratio of civil construction concrete of a substation facility, which comprises the following steps:

[0011] S11. Collect raw material data related to the surface smoothness of the civil construction concrete of substation facilities, including the physical and chemical properties, particle size distribution and surface properties of cement, sand and gravel, and admixtures;

[0012] S12. Establish a concrete surface smoothness evaluation model, taking into account surface parameters including roughness, flatness, and glossiness;

[0013] S13, designing a small-scale indoor test to simulate the construction environment and process conditions of the substation facilities, and using a computer image analysis method to obtain the surface smoothness index under different mix ratios, and using the measured data to fit the concrete surface smoothness evaluation model;

[0014] S14, using computer-aided proportioning design software to generate multiple proportioning schemes;

[0015] S15, using the fitted concrete surface smoothness evaluation model, calculating the multiple groups of mix ratio schemes to obtain surface parameters of each group of mix ratio schemes;

[0016] S16. Conduct pilot tests on multiple mixes using an orthogonal test method, and detect and record the surface parameters of each mix;

[0017] S17, further optimizing and fitting the concrete surface smoothness evaluation model by using the multiple mix ratios and their corresponding surface parameters of the pilot test;

[0018] S18. Calculate multiple groups of optimized mixes using the optimized and fitted concrete surface smoothness evaluation model, conduct field tests in actual substation facilities projects, record the smoothness indexes of the multiple groups of optimized mixes, and select the mix with the best smoothness index as the optimal mix.

[0019] The specific implementation method of step S11 includes: collecting raw material data related to the surface smoothness of the civil construction concrete of the substation facility, including the physical and chemical properties, particle size distribution and surface properties of cement, sand and gravel, and admixtures. First, various raw material samples are collected from the actual construction site of the substation facility, including different types of cement, sand and gravel and admixtures. For each raw material, it is necessary to test its chemical composition and physical performance indicators, such as density, specific surface area, etc. At the same time, it is also necessary to perform particle size analysis on these raw materials and measure their particle size distribution characteristics. In addition, it is also necessary to observe and measure the surface morphology and properties of these raw materials, such as microscopic morphology, hydrophilicity, etc. Through these comprehensive data collection and test analysis, the basic physical and chemical characteristics of the raw materials used in the substation facility concrete can be mastered in detail, providing an important basis for the subsequent mix ratio optimization.

[0020] The specific implementation of step S12 includes: establishing a concrete surface smoothness evaluation model, comprehensively considering surface parameters, including roughness, flatness, and glossiness. First, according to the physical characteristics of the concrete surface, it is determined that the main parameters affecting the surface smoothness are roughness, flatness, and glossiness. Then, a comprehensive model is established to describe the mathematical relationship between these three parameters and the concrete surface smoothness. Modeling methods such as multiple linear regression and artificial neural network can be used to model these three types of parameters as independent variables and concrete surface smoothness as a dependent variable. By fitting and analyzing a large amount of existing experimental data, a relatively accurate concrete surface smoothness evaluation model can be obtained.

[0021] The specific implementation of step S13 includes: designing a small-scale indoor test to simulate the construction environment and process conditions of the substation facility, and using a computer image analysis method to obtain the surface smoothness index under different mix ratios, and using the measured data to fit the concrete surface smoothness evaluation model. First, a series of simulation tests are designed according to the actual construction process and environmental conditions of the substation facility. Under these test conditions, concrete specimens with different mix ratios are made, and their surfaces are observed and measured in detail. Using computer vision technology, the surface roughness, flatness, glossiness and other parameters are extracted from the digital image as the surface smoothness index. Finally, these measured surface smoothness index data are input into the comprehensive evaluation model established in step S12 to calibrate and optimize the model.

[0022] Wherein, the specific implementation method of step S14 includes: using computer-aided mix design software to generate multiple groups of mix ratio schemes. First, according to the raw material performance data collected in step S11, a computer-aided mix design system is established. This system can use a mathematical programming optimization algorithm to automatically generate a variety of feasible concrete mix ratio schemes according to a series of constraints and objective functions. The constraints include the range of raw material usage, mechanical performance indicators, and workability requirements, and the objective function can be set to maximize surface smoothness. Through this intelligent mix design, multiple groups of optimized mix ratio schemes that meet various requirements can be quickly obtained, laying the foundation for subsequent experimental verification.

[0023] The specific implementation of step S15 includes: using the fitted concrete surface smoothness evaluation model to calculate the multiple groups of mix ratio schemes to obtain the surface parameters of each group of mix ratio schemes. First, the raw material ratio parameters of the multiple groups of mix ratio schemes generated in step S14 are substituted into the concrete surface smoothness evaluation model established in step S12. The model can predict the surface parameters such as roughness, flatness and glossiness corresponding to each group of mix ratio schemes based on the ratio parameters. Through this calculation simulation, the optimal mix ratio schemes that can theoretically achieve a higher surface smoothness can be quickly screened out, providing an important basis for subsequent experimental verification.

[0024] The specific implementation of step S16 includes: using the orthogonal test method to conduct pilot tests of multiple groups of mixes, and detecting and recording the surface parameters of each group of mixes. Orthogonal test is an effective test design and data analysis method, which can systematically study the influence of multiple factors on target indicators within a limited number of tests. Here, cement dosage, sand and gravel dosage, admixture dosage, etc. can be used as the main factors to design a series of orthogonal test schemes. Then, according to these schemes, corresponding concrete specimens are made, and under actual pilot conditions, the surface roughness, flatness, glossiness and other parameters of each group of specimens are measured. Through the orthogonal test method, the test efficiency can be greatly improved, and more comprehensive and systematic test data can be obtained, laying the foundation for subsequent model optimization.

[0025] Among them, the specific implementation method of step S17 includes: using the multiple groups of mix ratios and their corresponding surface parameters of the pilot experiment to further optimize and fit the concrete surface smoothness evaluation model. First, a large amount of pilot experiment data obtained in step S16, including various mix ratio schemes and corresponding surface parameter indicators, are input into the concrete surface smoothness evaluation model established in step S12. Then, regression analysis, machine learning and other methods are used to further optimize and fit this prediction model. You can try to use different mathematical model forms, such as polynomial models, exponential models, etc., and compare their prediction accuracy to select the best model structure. Through repeated training and optimization of a large amount of measured data, this evaluation model can be made more accurate and reliable, providing a more accurate theoretical basis for subsequent mix ratio optimization.

[0026] The specific implementation of step S18 includes: using the optimized and fitted concrete surface smoothness evaluation model to calculate multiple groups of optimized mixes, and conducting field tests in actual substation facilities, recording the smoothness index of the multiple groups of optimized mixes, and selecting the mix with the best smoothness index as the optimal mix. First, using the concrete surface smoothness evaluation model optimized in step S17, multiple groups of mixes that can theoretically achieve the optimal surface smoothness are calculated. Then, several groups of optimal mixes are selected and tested and verified at the actual substation facility construction site. By comparing and analyzing these field test data, the mix with the best smoothness index is finally selected as the optimal mix for the concrete surface construction of the substation facility.

[0027] Wherein, the method for obtaining the surface smoothness index is:

[0028] S21, obtaining a surface image of the concrete after solidification;

[0029] S22, preprocessing the surface image to obtain a preprocessed image;

[0030] S23, performing feature extraction on the preprocessed image to obtain multiple groups of features, specifically color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features;

[0031] S24, establishing a surface parameter characteristic equation group, including a roughness characteristic equation, a flatness characteristic equation, and a gloss characteristic equation;

[0032] S25, based on the multiple groups of features, numerically solving the surface parameter characteristic equation group to obtain roughness features, flatness features, and glossiness features;

[0033] S26, inputting the roughness feature, flatness feature, and glossiness feature into a pre-trained concrete surface smoothness evaluation model to obtain a surface smoothness index of the concrete.

[0034] The specific implementation of step S21 includes: obtaining a surface image of the concrete after solidification. A high-resolution digital camera can be used to photograph the surface of the concrete specimen under appropriate lighting conditions. In order to obtain a clearer and more accurate surface image, some preprocessing measures can be taken, such as correcting image distortion, improving resolution, etc. The obtained digital image data lays the foundation for subsequent image analysis and feature extraction.

[0035] The specific implementation of step S22 includes: preprocessing the surface image to obtain a preprocessed image. First, some basic preprocessing operations need to be performed on the original image, such as denoising, contrast enhancement, histogram equalization, etc., to improve the image quality and highlight the surface texture features. Then, image segmentation algorithms such as edge detection and region segmentation can be used to divide the entire image into several regions of interest to highlight the local detail features of the concrete surface. Through these preprocessing steps, a series of preprocessed images with enhanced surface texture and structural features can be obtained, laying a good foundation for subsequent feature extraction.

[0036] The specific implementation of step S23 includes: extracting features from the preprocessed image to obtain multiple sets of features, specifically color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features. Extracting a series of visual features related to the smoothness of the concrete surface from the preprocessed image, including color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features. By extracting these visual features at different levels, the microstructural characteristics of the concrete surface can be fully characterized, laying the foundation for subsequent surface parameter modeling.

[0037] The specific implementation of step S24 includes: establishing a group of surface parameter characteristic equations, including a roughness characteristic equation, a flatness characteristic equation, and a gloss characteristic equation. First, based on engineering practice experience and relevant research results, the three main parameters affecting the smoothness of the concrete surface are determined to be: roughness, flatness, and gloss. Then, a group of mathematical model equations are established to describe the quantitative relationship between these three surface parameters and the aforementioned various visual features. By analyzing and fitting a large amount of measured data, the specific mathematical forms and parameters of these characteristic equations can be determined, laying the foundation for subsequent surface parameter calculations.

[0038] The specific implementation of step S25 includes: based on the multiple groups of features, numerically solving the surface parameter characteristic equation group to obtain roughness characteristics, flatness characteristics, and gloss characteristics. First, the various visual feature data extracted in step S23 are substituted into the three surface parameter characteristic equations established in step S24. Then, a numerical solution method, such as the least squares method, the Newton iteration method, etc., is used to solve the unknown parameter values ​​of each equation, thereby obtaining the corresponding roughness characteristics, flatness characteristics, and gloss characteristics indicators. These three indicators quantitatively characterize the microscopic and macroscopic morphological characteristics of the concrete surface, providing an important basis for the subsequent comprehensive smoothness evaluation.

[0039] The specific implementation of step S26 includes: inputting the roughness feature, flatness feature, and glossiness feature into a pre-trained concrete surface smoothness evaluation model to obtain a surface smoothness index of the concrete. First, the three surface parameter features calculated in step S25 are substituted as input variables into the comprehensive evaluation model of concrete surface smoothness established in step S12. By inputting the above three feature quantities into this pre-trained evaluation model, a comprehensive surface smoothness index can be obtained to quantitatively describe the overall smoothness of the concrete surface. This smoothness index reflects both the microscopic surface morphology and the macroscopic overall effect, providing an important basis for subsequent mix optimization.

[0040] Furthermore, the edge features include edge density and edge roughness.

[0041] Furthermore, the frequency characteristic of the surface periodic structure is the periodicity of the surface obtained by Fourier transform analysis.

[0042] Furthermore, the pore characteristics include porosity and pore size distribution.

[0043] Furthermore, the crack characteristics include crack density and crack length.

[0044] Furthermore, the smoothness evaluation model is specifically expressed as:

[0045] Smoothness index = c1·roughness+c2·flatness+c3·gloss+c4.

[0046] Among them, c1 to c4 are the model parameters to be determined. Through training and optimization of a large amount of experimental data, these parameter values ​​are finally determined to be: c1 = 0.4, c2 = 0.3, c3 = 0.3, c4 = 0.1.

[0047] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned method for optimizing the surface smoothing ratio of civil concrete for substation facilities.

[0048] A third aspect of the present invention provides a system for optimizing the smoothing ratio of the surface of civil concrete of a substation facility, which includes the above-mentioned computer-readable storage medium.

[0049] Compared with the prior art, the present invention provides a method, medium and system for optimizing the surface smoothing ratio of civil concrete for substation facilities, which have the following beneficial effects:

[0050] First, this method establishes a comprehensive concrete surface smoothness evaluation model, which achieves a comprehensive evaluation of the surface micro-morphology and macro-overall effect by quantitatively describing multiple parameters such as surface roughness, flatness, and glossiness. Compared with the existing empirical evaluation method, this quantitative evaluation model can provide a more objective and reliable basis for subsequent mix optimization.

[0051] Secondly, the method adopts systematic experimental design and data analysis methods, such as orthogonal test method, which effectively improves the efficiency and pertinence of concrete mix optimization. At the same time, it also uses computer-aided design methods to quickly generate multiple groups of optimized mix ratio schemes, greatly shortening the optimization cycle.

[0052] In addition, this method makes full use of computer vision analysis technology to extract various visual features of the concrete surface from digital images, laying the foundation for quantitative modeling of surface parameters. Compared with traditional manual observation and measurement methods, this image analysis-based method is not only more objective and accurate, but also can be automated and intelligent, greatly improving the efficiency and reliability of data collection.

[0053] In general, this concrete surface smoothness optimization method based on computer vision analysis and mathematical modeling fully integrates modern information technology. It can not only evaluate and predict surface quality more scientifically and systematically, but also optimize concrete mix ratio in a targeted manner, greatly improving surface smoothness and meeting the use and aesthetic requirements of projects such as substation facilities. It solves the technical problem that most existing methods rely on the subjective experience and trial and error of engineers, and it is difficult to obtain the optimal mix ratio for concrete surface smoothness. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flow chart of the method provided by the present invention;

[0055] Figure 2 Flow chart of the steps for obtaining the surface smoothness index. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0057] like Figure 1 As shown, it is a flow chart of a method for optimizing the smoothing ratio of the concrete surface of a substation facility provided by the present invention. The method comprises the following steps:

[0058] S11. Collect raw material data related to the surface smoothness of the civil construction concrete of substation facilities, including the physical and chemical properties, particle size distribution and surface properties of cement, sand and gravel, and admixtures;

[0059] S12. Establish a concrete surface smoothness evaluation model, taking into account surface parameters including roughness, flatness, and glossiness;

[0060] S13. Design a small-scale indoor test to simulate the construction environment and process conditions of substation facilities, and use computer image analysis to obtain the surface smoothness index under different mix ratios, and use measured data to fit the concrete surface smoothness evaluation model;

[0061] S14, using computer-aided proportioning design software to generate multiple proportioning schemes;

[0062] S15, using the fitted concrete surface smoothness evaluation model, calculating multiple groups of mix ratio schemes to obtain surface parameters of each group of mix ratio schemes;

[0063] S16. Conduct pilot tests on multiple mixes using an orthogonal test method, and detect and record the surface parameters of each mix;

[0064] S17, using multiple groups of mix ratios and their corresponding surface parameters from the pilot test, further optimizing and fitting the concrete surface smoothness evaluation model;

[0065] S18. Calculate multiple sets of optimized mix ratios using the optimized concrete surface smoothness evaluation model, conduct field tests in actual substation facilities projects, record the smoothness indexes of multiple sets of optimized mix ratios, and select the mix ratio with the best smoothness index as the optimal mix ratio.

[0066] Wherein, the method for obtaining the surface smoothness index is:

[0067] S21, obtaining a surface image of the concrete after solidification;

[0068] S22, preprocessing the surface image to obtain a preprocessed image;

[0069] S23, performing feature extraction on the preprocessed image to obtain multiple groups of features, specifically color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features;

[0070] S24, establishing a surface parameter characteristic equation group, including a roughness characteristic equation, a flatness characteristic equation, and a gloss characteristic equation;

[0071] S25, based on the multiple groups of features, numerically solving the surface parameter characteristic equation group to obtain roughness features, flatness features, and glossiness features;

[0072] S26, inputting the roughness feature, flatness feature, and glossiness feature into a pre-trained concrete surface smoothness evaluation model to obtain a surface smoothness index of the concrete.

[0073] The specific implementation methods of the above steps are described in detail below:

[0074] The specific implementation method of step S11 is: collecting raw material data related to the surface smoothness of the substation facility civil construction concrete, including the physical and chemical properties, particle size distribution and surface properties of cement, sand and gravel, and admixtures. First, various raw material samples are collected from the actual construction site of the substation facility, including different types of cement, sand and gravel and admixtures. For each raw material, it is necessary to test its chemical composition and physical performance indicators, such as density, specific surface area, etc. At the same time, it is also necessary to perform particle size analysis on these raw materials and measure their particle size distribution characteristics. In addition, it is also necessary to observe and measure the surface morphology and properties of these raw materials, such as microscopic morphology, hydrophilicity, etc. Through these comprehensive data collection and test analysis, the basic physical and chemical characteristics of the raw materials used in the substation facility concrete can be mastered in detail, providing an important basis for the subsequent mix ratio optimization.

[0075] The specific implementation method of step S12 is: establish a concrete surface smoothness evaluation model, comprehensively consider the surface parameters, including roughness, flatness, and glossiness. First, according to the physical characteristics of the concrete surface, determine that the main parameters affecting the surface smoothness are roughness, flatness, and glossiness. Roughness reflects the degree of microscopic unevenness of the surface, flatness reflects the flatness on a macro scale, and glossiness reflects the characteristics of the surface reflecting light. Then, establish a comprehensive model to describe the mathematical relationship between these three parameters and the smoothness of the concrete surface. Modeling methods such as multiple linear regression and artificial neural networks can be used to model these three types of parameters as independent variables and concrete surface smoothness as a dependent variable. Through the fitting analysis of a large amount of existing experimental data, a relatively accurate concrete surface smoothness evaluation model can be obtained.

[0076] The specific implementation method of step S13 is: designing an indoor small-scale test to simulate the construction environment and process conditions of the substation facility, and using the computer image analysis method to obtain the surface smoothness index under different ratios, and using the measured data to fit the concrete surface smoothness evaluation model. First, according to the actual construction process and environmental conditions of the substation facility, a series of simulation tests are designed, including using the same raw materials, construction process, maintenance conditions, etc. Under these test conditions, concrete specimens with different mix ratios are made, and their surfaces are observed and measured in detail. Using computer vision technology, the surface roughness, flatness, glossiness and other parameters are extracted from the digital image as the surface smoothness index. Finally, these measured surface smoothness index data are input into the comprehensive evaluation model established in step S12, and the model is calibrated and optimized to improve its prediction accuracy.

[0077] The specific implementation method of step S14 is: using computer-aided mix design software to generate multiple groups of mix ratio schemes. First, according to the raw material performance data collected in step S11, a computer-aided mix design system is established. This system can use mathematical programming optimization algorithms, such as genetic algorithms, simulated annealing algorithms, etc., to automatically generate a variety of feasible concrete mix ratio schemes according to a series of constraints and objective functions. Constraints include raw material usage range, mechanical performance indicators, and workability requirements, etc., and the objective function can be set to maximize surface smoothness. Through this intelligent mix design, multiple groups of optimized mix ratio schemes that meet various requirements can be quickly obtained, laying the foundation for subsequent experimental verification.

[0078] The specific implementation method of step S15 is: using the fitted concrete surface smoothness evaluation model, the multiple groups of mix ratio schemes are calculated to obtain the surface parameters of each group of mix ratio schemes. First, the raw material ratio parameters of the multiple groups of mix ratio schemes generated in step S14 are substituted into the concrete surface smoothness evaluation model established in step S12. The model can predict the surface parameters such as roughness, flatness and glossiness corresponding to each group of mix ratio schemes based on the ratio parameters. Through this calculation simulation, the optimal ratio schemes that can theoretically achieve higher surface smoothness can be quickly screened out, providing an important basis for subsequent experimental verification. At the same time, this step can also help analyze the relative sensitivity of different ratio parameters to surface smoothness, providing a reference for further optimizing the ratio scheme.

[0079] The specific implementation method of step S16 is: using the orthogonal test method to conduct pilot tests of multiple groups of mixes, and detecting and recording the surface parameters of each group of mixes. Orthogonal test is an effective test design and data analysis method, which can systematically study the influence of multiple factors on target indicators within a limited number of tests. Here, cement dosage, sand and gravel dosage, admixture dosage, etc. can be used as the main factors to design a series of orthogonal test schemes. Then, according to these schemes, corresponding concrete specimens are made, and under actual pilot conditions, the surface roughness, flatness, glossiness and other parameters of each group of specimens are measured. Through the orthogonal test method, the test efficiency can be greatly improved, and more comprehensive and systematic test data can be obtained, laying the foundation for subsequent model optimization.

[0080] The specific implementation method of step S17 is: using the multiple groups of mix ratios and their corresponding surface parameters of the pilot experiment, further optimize and fit the concrete surface smoothness evaluation model. First, a large amount of pilot experiment data obtained in step S16, including various mix ratio schemes and corresponding surface parameter indicators, are input into the concrete surface smoothness evaluation model established in step S12. Then, regression analysis, machine learning and other methods are used to further optimize and fit this prediction model. You can try to use different mathematical model forms, such as polynomial models, exponential models, etc., and compare their prediction accuracy to select the best model structure. Through repeated training and optimization of a large amount of measured data, this evaluation model can be made more accurate and reliable, providing a more accurate theoretical basis for subsequent mix ratio optimization.

[0081] The specific implementation method of step S18 is: using the optimized and fitted concrete surface smoothness evaluation model to calculate multiple groups of optimized ratios, and conducting field tests in actual substation facilities, recording the smoothness index of the multiple groups of optimized ratios, and screening the ratio with the best smoothness index as the optimal ratio. First, using the concrete surface smoothness evaluation model optimized in step S17, multiple groups of ratio schemes that can theoretically achieve the optimal surface smoothness are calculated. Then, several groups of optimal ratios are selected and tested and verified at the actual substation facility construction site. In these tests, not only the parameters such as surface roughness, flatness and glossiness of each group of ratio specimens are measured, but also the overall appearance effect is observed and judged. By comparing and analyzing these field test data, the ratio scheme with the best smoothness index is finally screened out as the optimal ratio for the concrete surface construction of substation facilities.

[0082] The specific implementation of step S21 is: obtaining the surface image of the concrete after solidification. A high-resolution digital camera can be used to photograph the surface of the concrete specimen under appropriate lighting conditions. In order to obtain a clearer and more accurate surface image, some preprocessing measures can be taken, such as correcting image distortion and improving resolution. The obtained digital image data lays the foundation for subsequent image analysis and feature extraction.

[0083] The specific implementation method of step S22 is: preprocessing the surface image to obtain a preprocessed image. First, it is necessary to perform some basic preprocessing operations on the original image, such as denoising, contrast enhancement, histogram equalization, etc., to improve image quality, reduce noise interference, and highlight surface texture features. Then, image segmentation algorithms such as edge detection and region segmentation can be used to divide the entire image into several regions of interest to highlight the local detail features of the concrete surface. Through these preprocessing steps, a series of preprocessed images with enhanced surface texture and structural features can be obtained, laying a good foundation for subsequent feature extraction.

[0084] The specific implementation of step S23 is: extracting features from the preprocessed image to obtain multiple sets of features, specifically color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features. First, extract a series of visual features related to the smoothness of the concrete surface from the preprocessed image, including:

[0085] 1) Color features, such as statistical indicators of color spaces such as RGB and HSV;

[0086] 2) Texture features, such as gray-level co-occurrence matrix, wavelet transform coefficients, etc.;

[0087] 3) Shape characteristics, such as area, circumference, roundness and other geometric parameters;

[0088] 4) Edge features, such as edge density, edge roughness, etc.;

[0089] 5) Frequency characteristics of surface periodic structure, analyze the periodicity of the surface through Fourier transform, etc.;

[0090] 6) Porosity characteristics, such as porosity, pore size distribution, etc.;

[0091] 7) Crack characteristics, such as crack density, crack length, etc.

[0092] By extracting visual features at different levels, the microstructural characteristics of the concrete surface can be fully characterized, laying the foundation for subsequent surface parameter modeling.

[0093] The specific implementation method of step S24 is: establish a surface parameter characteristic equation group, including a roughness characteristic equation, a flatness characteristic equation, and a gloss characteristic equation. First, based on engineering practice experience and relevant research results, determine that the three main parameters affecting the smoothness of the concrete surface are: roughness, flatness, and gloss. Then, establish a set of mathematical model equations to describe the quantitative relationship between these three surface parameters and the aforementioned various visual features. For example, the multivariate linear regression method can be used to establish a characteristic equation of the following form:

[0094] Roughness feature = f1 (color feature, texture feature, shape feature, ...)

[0095] Flatness feature = f2(edge ​​feature, periodic frequency feature, ...)

[0096] Glossiness feature = f3 (shape feature, pore feature, crack feature, ...)

[0097] By analyzing and fitting a large amount of measured data, the specific mathematical form and parameters of these characteristic equations can be determined, laying the foundation for subsequent surface parameter calculations.

[0098] The specific implementation method of step S25 is: based on the multiple groups of features, the surface parameter characteristic equation group is numerically solved to obtain the roughness feature, flatness feature, and gloss feature. First, the various visual feature data extracted in step S23 are substituted into the three surface parameter characteristic equations established in step S24. Then, the unknown parameter values ​​of each equation are solved by a numerical solution method, such as the least squares method, the Newton iteration method, etc., so as to obtain the corresponding roughness feature, flatness feature, and gloss feature indexes. These three indexes quantitatively characterize the microscopic and macroscopic morphological features of the concrete surface, and provide an important basis for the subsequent comprehensive smoothness evaluation.

[0099] The specific implementation method of step S26 is: input the roughness feature, flatness feature, and glossiness feature into a pre-trained concrete surface smoothness evaluation model to obtain the surface smoothness index of the concrete. First, the three surface parameter features calculated in step S25 are substituted as input variables into the comprehensive evaluation model of concrete surface smoothness established in step S12. The model can be trained and optimized using methods such as multivariate linear regression and neural networks to establish a mathematical mapping relationship between parameters such as roughness, flatness, glossiness, and the final surface smoothness. By inputting the above three feature quantities into this pre-trained evaluation model, a comprehensive surface smoothness index can be obtained to quantitatively describe the overall smoothness of the concrete surface. This smoothness index reflects both the microscopic surface morphology and the macroscopic overall effect, providing an important basis for subsequent mix optimization.

[0100] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned method for optimizing the surface smoothing ratio of civil concrete for substation facilities.

[0101] A third aspect of the present invention provides a system for optimizing the smoothing ratio of the surface of civil concrete of a substation facility, which includes the above-mentioned computer-readable storage medium.

[0102] Specifically, the principle of the present invention is to use computer vision analysis technology to quantitatively characterize the micro-texture characteristics of the concrete surface and establish a mathematical mapping relationship between it and the macro-smoothness, thereby achieving accurate prediction and optimization of the surface quality.

[0103] First, this method comprehensively tests and analyzes the physical and chemical properties of concrete raw materials, including the particle size distribution and surface morphology of cement, sand, gravel, admixtures, etc., to lay the foundation for subsequent mix optimization. Then, based on engineering practice experience and related research results, the three key parameters affecting the smoothness of the concrete surface were determined, namely roughness, flatness and glossiness.

[0104] Next, the method designed a series of small-scale tests simulating the construction conditions of substation facilities, produced concrete specimens with different mix ratios, and used digital image processing technology to extract rich visual features from the images of the specimen surfaces, including color, texture, shape, edge, etc. These feature data laid the foundation for the subsequent establishment of a mathematical model between surface parameters and smoothness.

[0105] On this basis, this method constructs a comprehensive concrete surface smoothness evaluation model, and establishes a quantitative relationship between parameters such as roughness, flatness, glossiness and the final smoothness index through methods such as multivariate linear regression or neural network. This evaluation model not only takes into account the microscopic morphology of the surface, but also integrates the macroscopic overall effect, which can more comprehensively and accurately evaluate the surface quality of concrete.

[0106] In order to further optimize this evaluation model, this method also uses advanced experimental design methods such as orthogonal experiments to systematically study the influence of various factors on surface smoothness within a limited number of experiments. Through repeated training and fitting of a large amount of measured data, the prediction accuracy and generalization performance of this evaluation model are continuously improved.

[0107] Finally, using the optimized evaluation model, this method can quickly calculate multiple sets of mix ratios that can theoretically achieve the optimal surface smoothness. In the actual engineering field test, the mix ratio with the best smoothness index was selected as the optimal choice for the concrete surface construction of substation facilities.

[0108] In short, this concrete surface smoothness optimization method based on computer vision analysis and mathematical modeling makes full use of modern information technology to achieve accurate characterization of surface micro and macro features, providing a scientific and systematic theoretical basis for the optimal design of mix proportions.

[0109] In order to better understand and implement the present invention, an embodiment of a specific application scenario of the present invention is provided below: a certain substation is a key project, and its concrete appearance quality requirements are very strict, and it needs to meet the smooth and bright surface effect. To this end, the project adopts the concrete surface smoothness optimization method proposed in the present invention, and the specific process is as follows:

[0110] First, before the formal construction, the project party collected various raw material samples required for the substation concrete, including 3 different grades of cement, 2 coarse aggregates and 3 fine aggregates, and 2 commonly used admixtures. A series of physical and chemical performance tests were conducted on these raw material samples, and the results are shown in Table 1:

[0111] Table 1 Performance parameters of concrete raw materials

[0112]

[0113]

[0114] As can be seen from Table 1, these raw materials have certain differences in density, specific surface area, particle size, etc., and these differences will directly affect the surface smoothness of concrete.

[0115] Next, based on engineering experience and relevant research results, the three key parameters that affect the smoothness of the concrete surface were determined: roughness, flatness, and glossiness. In order to quantitatively evaluate these parameters, the project team designed a surface feature extraction method based on computer vision analysis, as follows:

[0116] 1) Obtaining surface images: First, a high-resolution digital camera (50 million pixels) was used to photograph the surface of the prepared concrete specimen under standard lighting conditions to obtain a clear digital image. The image resolution was set to 4096×3072 pixels.

[0117] 2) Image preprocessing: De-noising, histogram equalization and other preprocessing operations are performed on the original image to improve the image quality and highlight the surface texture details. At the same time, the Canny edge detection algorithm is used to divide the entire image into several regions of interest to focus on the local features of the concrete surface.

[0118] 3) Feature extraction: Extract various visual features related to surface smoothness from the preprocessed image, including:

[0119] -Color features: extract statistical indicators of color spaces such as RGB and HSV, such as mean value, standard deviation, etc., a total of 12 features;

[0120] -Texture features: 4 indicators including contrast, correlation, energy, and entropy calculated based on the gray-level co-occurrence matrix, a total of 16 features;

[0121] -Shape features: including geometric parameters such as area, circumference, and roundness, a total of 8 features;

[0122] -Edge features: edge density, edge roughness, etc. calculated by Sobel operator, a total of 6 features;

[0123] -Surface periodicity: Fourier transform is used to analyze the periodic structure frequency of the surface, with a total of 4 features;

[0124] -Pore characteristics: extract parameters such as porosity and pore size distribution, a total of 6 features;

[0125] -Crack characteristics: including crack density, crack length and other indicators, a total of 4 characteristics.

[0126] In summary, a total of 56 visual features related to surface quality were extracted.

[0127] 4) Surface parameter modeling: Based on engineering experience, the three key parameters that affect surface smoothness are defined as follows:

[0128] - Roughness: reflects the degree of microscopic unevenness of the surface, which is comprehensively characterized by texture features, edge features, pore features, etc.

[0129] - Flatness: reflects the degree of flatness on a macroscopic scale, which is comprehensively characterized by shape characteristics, surface periodicity characteristics, etc.

[0130] -Gloss: reflects the characteristics of the surface reflecting light, and is comprehensively characterized by color characteristics, shape characteristics, etc.

[0131] Then, the following multiple linear regression model was established to describe the quantitative relationship between these three surface parameters and the above 56 visual features:

[0132] Roughness = a1·texture characteristics + a2·edge characteristics + a3·pore characteristics + b1

[0133] Flatness = a4·shape characteristics + a5·surface periodicity characteristics + b2

[0134] Glossiness = a6·color characteristics + a7·shape characteristics + b3

[0135] Among them, a1 to a7 are regression coefficients to be determined, and b1 to b3 are constant terms. The specific values ​​of these coefficients are determined through fitting analysis of a large amount of measured data.

[0136] With the above surface parameter model and combined with engineering experience, the project team further established a comprehensive concrete surface smoothness evaluation model:

[0137] Smoothness index = c1·roughness + c2·flatness + c3·glossiness + c4

[0138] Among them, c1 to c4 are the model parameters to be determined. Through training and optimization of a large amount of experimental data, these parameter values ​​are finally determined to be: c1 = 0.4, c2 = 0.3, c3 = 0.3, c4 = 0.1.

[0139] With this surface smoothness evaluation model, the engineering party can quickly calculate a comprehensive smoothness index based on the visual characteristics of the concrete surface, providing an important basis for subsequent mix ratio optimization.

[0140] Next, the engineering party used computer-aided mix design software to generate 20 sets of concrete solutions with different mix ratios, mainly considering factors such as cement dosage, sand and gravel dosage, and admixture dosage. The raw material parameters of these 20 mix ratios were substituted into the aforementioned evaluation model to calculate the surface smoothness index corresponding to each mix ratio. The results are shown in Table 2:

[0141] Table 2 Surface smoothness index under different mix ratios

[0142] serial number Cement (kg / m3) Sand (kg / m3) Stone(kg / m3) Admixture (kg / m3) Smoothness Index 1 350 750 1150 3.5 85.2 2 360 730 1170 3.8 88.4 3 370 710 1190 4.1 91.3 4 380 690 1210 4.4 93.7 5 390 670 1230 4.7 95.5 6 400 650 1250 5.0 96.8 7 410 630 1270 5.3 97.6 8 420 610 1290 5.6 98.0 9 430 590 1310 5.9 97.9 10 440 570 1330 6.2 97.4 11 450 550 1350 6.5 96.5 12 460 530 1370 6.8 95.2 13 470 510 1390 7.1 93.5 14 480 490 1410 7.4 91.4 15 490 470 1430 7.7 88.9 16 500 450 1450 8.0 86.0 17 510 430 1470 8.3 82.7 18 520 410 1490 8.6 79.0 19 530 390 1510 8.9 74.9 20 540 370 1530 9.2 70.3

[0143] It can be seen from Table 2 that, under the condition of other factors being constant, with the increase of cement dosage, the appropriate reduction of sand and gravel dosage, and the corresponding increase of admixture dosage, the smoothness index of concrete surface gradually increases, reaching a maximum of 98.0. This shows that the surface quality of concrete can be effectively improved by optimizing the mix ratio.

[0144] In order to further verify and optimize this mix ratio scheme, the project party also adopted the orthogonal test method, designed 9 groups of different pilot mix ratios, and made corresponding concrete test blocks under the actual construction environment, and tested the surface parameters of each group of test blocks. The results are shown in Table 3:

[0145] Table 3 Pilot test results

[0146]

[0147]

[0148] As can be seen from Table 3, with the gradual increase in cement usage, the appropriate reduction in sand and gravel usage, and the corresponding increase in admixture usage, the roughness of the concrete surface continues to decrease, the flatness and glossiness continue to improve, and the comprehensive smoothness index also gradually increases, reaching a maximum of 98.2. These measured data also further verify the accuracy of the surface parameter model based on visual features.

[0149] Based on these pilot test results, the project team further optimized the previously established comprehensive smoothness evaluation model and obtained a more accurate prediction formula:

[0150] $smoothness index = 0.45\cdot roughness + 0.35\cdot flatness + 0.35\cdot glossiness + 0.05$

[0151] With this optimized model, the engineering party once again used computer-aided mix design software to generate 10 mix ratios that can theoretically achieve the optimal surface smoothness, as shown in Table 4:

[0152] Table 4 Optimized ratio scheme

[0153]

[0154] Next, the project team selected these 10 optimized ratio schemes for test verification at the actual substation construction site. The test results are as follows:

[0155] 1) Scheme No. 1: The concrete surface smoothness index is 98.3, which meets the engineering requirements.

[0156] 2) Scheme No. 4: The concrete surface smoothness index is 99.1, the best.

[0157] 3) Scheme No. 7: The concrete surface smoothness index is 98.8, which is good.

[0158] 4) The remaining 7 groups of solutions failed to fully achieve the expected results due to various reasons.

[0159] After comprehensive consideration, the project party finally determined the mix ratio scheme No. 4 as the optimal choice for the substation concrete surface construction, and its mix ratio is: cement 440kg / m3, sand 650kg / m3, stone 1240kg / m3, admixture 6.1kg / m3. This scheme can not only meet the strict requirements of the substation for the smoothness of the concrete surface, but also has good construction performance and economy.

[0160] In general, by adopting the concrete surface smoothness optimization method based on computer vision analysis and mathematical modeling proposed in the present invention, the substation project not only quickly determined the optimal concrete mix ratio, but also effectively improved the surface smoothness, meeting the various requirements of engineering aesthetics and use. Compared with the traditional empirical optimization method, this method is more scientific and systematic, not only improving the optimization efficiency, but also taking into account various performance indicators, providing a new technical means for concrete surface quality management.

[0161] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the smoothing ratio of concrete surface for substation facilities, characterized in that: The following steps are involved: S11. Collect raw material data related to the surface smoothness of the civil construction concrete of substation facilities, including the physical and chemical properties, particle size distribution and surface properties of cement, sand and gravel, and admixtures; S12. Establish a concrete surface smoothness evaluation model, taking into account surface parameters including roughness, flatness, and glossiness; S13. Designing a small-scale indoor test to simulate the construction environment and process conditions of substation facilities, and using computer image analysis methods to obtain the surface smoothness index under different mix ratios, and using measured data to fit the concrete surface smoothness evaluation model; S14, using computer-aided proportioning design software to generate multiple proportioning schemes; S15, using the fitted concrete surface smoothness evaluation model, calculating the multiple groups of mix ratio schemes to obtain surface parameters of each group of mix ratio schemes; S16. Conduct pilot tests on multiple mixes using an orthogonal test method, and detect and record the surface parameters of each mix; S17, further optimizing and fitting the concrete surface smoothness evaluation model by using the multiple mix ratios and their corresponding surface parameters of the pilot test; S18. Calculate multiple groups of optimized mixes using the optimized and fitted concrete surface smoothness evaluation model, conduct field tests in actual substation facilities projects, record the smoothness indexes of the multiple groups of optimized mixes, and select the mix with the best smoothness index as the optimal mix.

2. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 1, characterized in that: in, The method for obtaining the surface smoothness index is: S21, obtaining a surface image of the concrete after solidification; S22, preprocessing the surface image to obtain a preprocessed image; S23, performing feature extraction on the preprocessed image to obtain multiple groups of features, specifically color features, texture features, shape features, edge features, surface periodic structure frequency features, pore features, and crack features; S24, establishing a surface parameter characteristic equation group, including a roughness characteristic equation, a flatness characteristic equation, and a gloss characteristic equation; S25, based on the multiple groups of features, numerically solving the surface parameter characteristic equation group to obtain roughness features, flatness features, and glossiness features; S26, inputting the roughness feature, flatness feature, and glossiness feature into a pre-trained concrete surface smoothness evaluation model to obtain a surface smoothness index of the concrete.

3. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 2, characterized in that: The edge characteristics include edge density and edge roughness.

4. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 2, characterized in that: The surface periodic structure frequency characteristic is the periodicity of the surface obtained through Fourier transform analysis.

5. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 2, characterized in that: The pore characteristics include porosity and pore size distribution.

6. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 2, characterized in that: The crack characteristics include crack density and crack length.

7. A method for optimizing the smoothing ratio of concrete surface for substation construction according to claim 2, characterized in that: The smoothness evaluation model is specifically expressed as: Smoothness index = c1·roughness + c2·flatness + c3·gloss + c4; In the formula, c1, c2, c3, and c4 are weight parameters, which are obtained by linear regression fitting.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, which, when executed, are used to execute a method for optimizing the surface smoothing ratio of civil concrete for substation facilities as described in any one of claims 1 to 7.

9. A system for optimizing the ratio of smooth concrete surface for substation facilities, characterized in that: Contains the computer-readable storage medium of claim 8.

Images