Earth and rockfill dam construction feedback control optimization method considering grading spatial variability of damming rockfill material

Abstract

The invention discloses an earth and rockfill dam construction feedback control optimization method considering the grading space variability of damming rockfill materials, and the method comprises the steps: obtaining an image of the surface of an earth and rockfill dam through an unmanned plane, and carrying out the preprocessing of the image; the fractal dimension of the rockfill material filling bin surface image is obtained through a box dimension method, and information such as the potential dense stacking void ratio of the in-situ grading rockfill material is predicted according to a stacking algorithm; repeating the steps at different positions and different elevations of the filling bin surface of the earth and rockfill dam, and obtaining the overall mechanical property distribution of the earth and rockfill dam by adopting a state-related constitutive theory of rockfill materials; according to the overall mechanical property distribution, the deformation rule of the dam body in the construction period is rapidly predicted, and a follow-up construction feedback control optimization scheme is made according to a set deformation threshold value. According to the method, the grading information of the rockfill material can be rapidly and nondestructively obtained, construction optimization is carried out, the economic and time cost is reduced, and the construction quality and the stability of the earth and rockfill dam are improved.

Description

Technical Field

[0001] The invention belongs to the field of civil engineering, and in particular relates to an earth-rock dam construction feed-control optimization method taking into account the spatial variability of dam-building rockfill material grading. Background Art

[0002] Earth-rock dams are a widely used hydraulic engineering structure, mainly used for water storage, flood control, power generation, etc. They are widely used around the world for their economy, strong adaptability and simple construction.

[0003] Construction control is a key link to ensure the quality of earth-rock dam projects. Existing construction control technologies mainly include quality inspection of construction materials, real-time monitoring of the construction process, and strict management of construction technology. However, due to the large scale of earth-rock dam projects and the complex and changeable construction environment, existing technologies still face many challenges in actual operation, such as the unevenness of construction materials and uncertain factors in the construction process.

[0004] During the construction of earth-rock dams, the gradation and spatial variability of rockfill materials have an important impact on the stability and safety of earth-rock dams. Spatial variability may lead to uneven stress distribution, affecting the long-term stability and safety of the dam body. Traditional design methods often assume that the material is evenly distributed, but this assumption is not always true in actual engineering. In addition, traditional methods for detecting the gradation of rockfill materials mainly rely on on-site sampling and laboratory screening. However, the on-site sampling and laboratory screening processes are complicated, time-consuming, and labor-intensive; traditional detection methods usually require destructive sampling and cannot achieve large-scale, continuous detection; due to the long detection cycle, traditional detection and construction quality control methods are difficult to provide real-time feedback on the impact of gradation changes during construction on dam deformation and stability.

[0005] With the rapid development of drone technology and image analysis algorithms, it is possible to use drones to collect images of the earth-rock dam surface and obtain rockfill gradation information in combination with intelligent image analysis algorithms. Drones can quickly cover the surface of earth-rock dams and collect high-resolution images, significantly improving detection efficiency; image analysis technology does not require destructive sampling and can achieve large-scale, continuous non-destructive testing; combined with intelligent image analysis algorithms, it can obtain rockfill gradation information in real time and provide timely feedback for the construction process. Summary of the invention

[0006] In view of the shortcomings of the existing technology, the present invention proposes a feedback control optimization method for earth-rock dam construction taking into account the spatial variability of the gradation of dam-building rockfill materials. Through unmanned aerial vehicle image acquisition, intelligent image analysis, rockfill material state-related constitutive theory and finite element numerical simulation, the rockfill material gradation information is obtained quickly and non-destructively, and construction optimization is carried out to ensure the construction quality and long-term stability of the earth-rock dam.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A feed-back optimization method for earth-rock dam construction considering spatial variability of dam-building rockfill material gradation comprises the following steps:

[0009] S1: using a drone to take images of the surface of the earth-rock dam, and preprocessing the images;

[0010] S2: Analyze the pre-processed image through intelligent image analysis algorithm to obtain the grading curve and fractal dimension of the rockfill material;

[0011] S3: Through the gradation curve and fractal dimension of the rockfill material, the potential dense stacking porosity and fine particle content information of the rockfill material are obtained;

[0012] S4: obtaining a constitutive relationship related to the state of the rockfill material through information related to the dense stacking porosity and fine particle content and triaxial test results of the calibrated graded rockfill material sample, and predicting the mechanical properties of the rockfill material in the image acquisition area based on the constitutive relationship related to the state of the rockfill material;

[0013] S5: By repeating S1 to S3 at different positions of the earth-rock dam, the overall mechanical property distribution of the earth-rock dam is obtained;

[0014] S6: Carry out corresponding finite element numerical simulation according to the distribution law of the overall mechanical properties of the earth-rock dam, predict the deformation law of the dam body during the construction period, and realize the optimization of the feedback control of the earth-rock dam construction based on the deformation law.

[0015] Preferably, in S2, the method of analyzing the preprocessed image by an intelligent image analysis algorithm to obtain the grading curve and fractal dimension of the rockfill material includes:

[0016] The collected RGB color image of the earth-rock dam surface is converted into a grayscale image by floating point method;

[0017] Applying the Otsu binarization method to binarize the grayscale image and extracting the boundary texture features of the rockfill particles;

[0018] The binarized image is analyzed using the "box" dimension method to extract the image fractal dimension D i , as an indicator to characterize the fine material content of rockfill materials;

[0019] Using the image fractal dimension D i , the grading curve of rockfill materials is evaluated and calculated.

[0020] Preferably, the binarized image is analyzed using the "box" dimension method to extract the image fractal dimension D i The methods include:

[0021] According to the aspect ratio of the image, a rectangular "box" with a characteristic size of Δ is selected to cover the image. The covered pixels must satisfy the pixel aspect ratio of the box and the overall image, and the length and width of the selected image must be greater than 1000 pixels.

[0022] Count the number of boundaries of particles contained in a "box" with a characteristic size of Δ, N(Δ), and change the value of Δ to obtain the number N(Δ) covered by several boxes;

[0023] The relationship curve between lgN and lgΔ is plotted in the double logarithmic coordinate system, and the fractal dimension of the filling bin surface image is taken as the negative number of the slope of the curve D i =-k.

[0024] Preferably, the image fractal dimension D is used i , the methods for evaluating and calculating the grading curve of rockfill materials include:

[0025]

[0026] Where d is the particle size, d M is the maximum particle size of rockfill materials; D is the fractal dimension of rockfill material grading.

[0027] Preferably, in S4, the method for predicting the mechanical properties of the rockfill material in the image acquisition area based on the constitutive relationship related to the state of the rockfill material includes:

[0028]

[0029] η c =M g (ξ c Ψ+1)

[0030]

[0031] η g =M g (ξ g Ψ-1)

[0032] Where p and q are the mean effective stress and generalized shear stress, respectively, and η is the stress ratio η=p / q; η c , η p and η g are the stress ratios of zero expansion, peak value and critical state respectively; N is the material parameter related to the normal consolidation line, k f and n f They are the material parameters, and are plastic volume strain and plastic shear strain, m g and M g is the material parameter related to the plastic potential function, ξ c andg are the state-dependent material parameters and constitutive model parameters respectively; H is the yield surface hardening parameter of the elastoplastic model; the expression of the state parameter of the model is Ψ=ee cs , where e is the current porosity of the material, e cs is the void ratio under critical state; p r and p' r are the reference stresses on the loading and unloading curves, respectively, according to the known porosity ratio e at a certain moment. L and the corresponding normal stress p L get and λ and κ are parameters related to the plastic and elastic moduli of the normal consolidation line, respectively.

[0033] Preferably, in S5, the method of obtaining the overall mechanical property distribution of the earth-rock dam by repeating S1 to S3 at different positions of the earth-rock dam comprises:

[0034] Perform image recognition, gradation analysis and porosity ratio prediction on rockfill materials at different locations of earth-rock dams to obtain mechanical property data of rockfill materials at each location;

[0035] The mechanical property data of each location are integrated through Kriging interpolation to obtain the overall mechanical property distribution of the earth-rock dam.

[0036] Preferably, S6: carrying out corresponding finite element numerical simulation according to the distribution law of the overall mechanical properties of the earth-rock dam to predict the deformation law of the dam body during the construction period, and based on the deformation law, the method for realizing the optimization of the feedback control of the earth-rock dam construction includes:

[0037] Collect and analyze the overall mechanical properties distribution data of earth-rock dams, and use the critical state porosity ratio e of a dam-building rockfill material to determine the critical state porosity ratio e of a dam-building rockfill material. cs The area with a probability of 5% greater than the set threshold is used as a filling warning area;

[0038] According to the distribution law of overall mechanical properties and finite element numerical simulation, the deformation law of the dam body during construction is predicted, and the ratio of the extreme settlement of the dam body to the current dam height s / H greater than the set threshold is used as an early warning indicator of deformation during dam body construction;

[0039] According to the filling warning area and the deformation warning indicators during the dam construction period, the construction quality feedback information is returned, and the dam material source and construction parameters during the construction process are dynamically adjusted to achieve feedback control optimization of earth-rock dam construction.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] Non-destructive and rapid acquisition of rockfill material gradation information: Using drones and intelligent image analysis technology, the present invention can quickly and non-destructively acquire rockfill material gradation information. Compared with traditional on-site sampling and laboratory screening methods, the detection efficiency is greatly improved and the detection time is reduced. The method of the present invention realizes non-destructive detection of rockfill material gradation through image analysis technology, avoiding the impact of destructive sampling in traditional methods on earth-rock dam structures, and is suitable for large-scale, continuous detection tasks.

[0042] Real-time feedback and optimization: Combining real-time monitoring data and intelligent analysis algorithms, the present invention can timely feedback the grading information of rockfill materials and dynamically adjust the construction plan according to the detection results to ensure the construction quality. Compared with traditional methods, the present invention can perform real-time quality control and optimization during the construction process, significantly improving the construction efficiency and quality.

[0043] Accurately predict deformation laws during construction: Through finite element numerical simulation and overall mechanical property distribution analysis, the present invention can accurately predict the deformation laws of the dam body during construction, timely identify potential mechanical weaknesses and risk areas, and formulate scientific construction feedback control optimization plans to ensure the safety and stability of earth-rock dams.

[0044] Dynamically adjust construction parameters: According to the filling warning area and deformation warning indicators, the present invention can provide construction quality feedback information to the design and construction team, dynamically adjust the dam material source and construction parameters during the construction process, ensure the construction quality and deformation stability of the earth-rock dam, and further improve the safety and reliability of the construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0046] Figure 1 It is a general flow chart of a construction feedback control optimization method emphasizing gradation spatial variability according to an embodiment of the present invention;

[0047] Figure 2 A flow chart for predicting mechanical behavior related to soil-rock filling material gradation according to an embodiment of the present invention;

[0048] Figure 3 This is a surface image of a rockfill material compaction bin surface obtained by using a drone in an embodiment of the present invention;

[0049] Figure 4 This is an apparent image of the rockfill material after the grayscale image is binarized and then processed by the "box" dimension method according to an embodiment of the present invention;

[0050] Figure 5 A schematic diagram of predicting the critical state porosity according to the linear relationship between the minimum porosity and the critical state porosity in an embodiment of the present invention;

[0051] Figure 6 It is a schematic diagram of carrying out corresponding finite element numerical simulation according to the overall mechanical property distribution of the dam in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Embodiment 1

[0055] This embodiment provides a feedback control optimization method for earth-rock dam construction considering the spatial variability of dam rockfill material grading. The overall process is as follows: Figure 1 As shown, the following steps are included:

[0056] Step 1: Use a drone to take images of the surface of the earth-rock dam, and pre-process the images;

[0057] Step 2: Analyze the preprocessed image through intelligent image analysis algorithm to obtain the grading curve and fractal dimension of the rockfill material;

[0058] Step 3: Obtain the potential dense stacking porosity and fine particle content information of the rockfill material through the gradation curve and fractal dimension of the rockfill material;

[0059] Step 4: Obtain the relevant constitutive relationship of the rockfill material state through the information related to the dense stacking porosity and fine particle content and the triaxial test results of the calibrated graded rockfill material sample, and predict the mechanical properties of the rockfill material in the image acquisition area based on the relevant constitutive relationship of the rockfill material state;

[0060] Step 5: By repeating S1 to S3 at different positions of the earth-rock dam, the overall mechanical property distribution of the earth-rock dam is obtained;

[0061] Step 6: Carry out corresponding finite element numerical simulation according to the distribution law of the overall mechanical properties of the earth-rock dam, predict the deformation law of the dam body during the construction period, and realize the optimization of the feedback control of the earth-rock dam construction based on the deformation law.

[0062] Specifically, step 1: drone image capture of the dam surface image

[0063] Choose a drone with a high-resolution camera and stable flight performance, such as the DJI Phantom4RTK or a similar model.

[0064] Make sure the drone battery is sufficient, the camera is clean, and the memory card has sufficient capacity.

[0065] Set the flight parameters, including flight altitude, speed, angle and shooting interval, to obtain clear and high-resolution images. The flight altitude is usually set to 5 meters, and the shooting angle is perpendicular to the dam surface.

[0066] According to the specific shape and area of ​​the earth-rock dam, a reasonable flight path is planned to ensure the comprehensiveness and coverage of image acquisition. The flight path should cover the entire earth-rock dam filling area, especially the primary and secondary rockfill areas. The "sliding window" method is used for multiple shooting to ensure that the image overlap rate is above 60% for subsequent image stitching.

[0067] The drone was flown according to the predetermined path and parameters, and RGB color images of the earth-rock dam surface were collected. During the shooting process, the drone was ensured to fly stably to avoid image blur. The images of each warehouse surface were spliced ​​into a large warehouse surface image using the "sliding window" method.

[0068] Step 2: Obtain information such as rockfill material gradation and fractal dimension through intelligent image analysis algorithm.

[0069] The collected RGB color image (such as Figure 3 The image is converted into a grayscale image by floating point method to simplify the subsequent processing flow.

[0070] Image preprocessing includes operations such as image denoising and contrast enhancement to ensure image quality.

[0071] Apply Otsu binarization method or equivalent grayscale method to binarize the grayscale image, such as Figure 4 As shown in the figure, the boundary texture features of rockfill particles are extracted. The Otsu method calculates the inter-class variance of the image and automatically determines the optimal threshold for binarization to ensure clear particle boundaries.

[0072] The binarized image is analyzed using the "box" dimension method to extract the image fractal dimension D i , as an indicator of the fine material content of rockfill material. The specific steps include: selecting a rectangular "box" with a characteristic size of Δ to cover the image according to the aspect ratio of the image, such as Figure 4As shown, the covered pixels must satisfy the pixel aspect ratio of the box and the overall image, and the length and width of the selected image must be greater than 1000 pixels. The number of particle boundaries contained in the "box" with a characteristic size of Δ is counted N(Δ), and the number of boxes covered N(Δ) is obtained by changing the value of Δ. The relationship curve between lgN and lgΔ is plotted in a double logarithmic coordinate system. The fractal dimension of the filling bin surface image is the negative number of the slope of the curve D i =-k.

[0073] Using fractal dimension D i Calculate the grading curve of rockfill materials using the following formula:

[0074]

[0075] Where d is the particle size, d M is the maximum particle size of the rockfill material; D is the fractal dimension of the rockfill material gradation. M The maximum value of the distance between two adjacent black pixels in the binary image is selected; the value of D is related to the fractal dimension D of the image. i The linear relationship is shown in Figure 2, and the linear parameters are determined through the particle analysis test results and image analysis results of several groups of rockfill materials.

[0076] Step 3: Obtain the potential dense stacking porosity of rockfill materials through image information

[0077] The grading curve of the rockfill material is determined by the above method, and the dimensionality reduction mapping algorithm is used to map the three-dimensional particle stacking to a one-dimensional line segment, thereby obtaining the minimum void ratio, and the potential dense stacking density of the rockfill material in the image area is obtained according to the particle specific gravity of the rockfill material:

[0078]

[0079] Among them, G s is the particle density of rockfill material, e min is the minimum void ratio.

[0080] Step 4: Obtain mechanical properties of rockfill materials through information such as porosity ratio

[0081] The steps to predict the mechanical properties of rockfill materials are as follows: Figure 2 As shown in the figure, a set of maximum dry density and triaxial tests of typical graded rockfill materials are carried out, and the parameters of the state-dependent elastic-plastic constitutive model are determined based on the test results. The constitutive model is in the following form:

[0082]

[0083] η c =M g (ξ c Ψ+1)

[0084]

[0085] η g =M g (ξ g Ψ-1)

[0086] Where p and q are the mean effective stress and generalized shear stress, respectively, and η is the stress ratio η=p / q; η c , η p and η g are the stress ratios of zero expansion, peak value and critical state respectively; N is the material parameter related to the normal consolidation line, k f and n f They are the material parameters, and are plastic volume strain and plastic shear strain, m g and M g is the material parameter related to the plastic potential function, ξ c and g are state-dependent material parameters and constitutive model parameters, respectively; H is the yield surface hardening parameter of the elastic-plastic model; p r and p' r are the reference stresses on the loading and unloading curves, respectively, according to the known porosity ratio e at a certain moment. L and the corresponding normal stress p L get and λ and κ are parameters related to the plastic and elastic moduli of the normal consolidation line, respectively.

[0087] Calculate the material state parameter Ψ:

[0088] Ψ=ee cs

[0089] Among them, e is the current porosity of the material, e cs is the void ratio under critical state (sufficiently large shear strain).

[0090] Determine the critical state line position:

[0091] e cs =Γ(p+p cs ) -λ

[0092]

[0093] Where Γ is the material parameter, p cs According to the porosity ratio e on a known critical state line cs0 and its corresponding average stress p cs0 get

[0094] Where: critical porosity ratio e cs The relationship with the minimum porosity is as follows Figure 5 As shown, therefore, the minimum porosity can be obtained using step 3, and the critical porosity can be predicted based on the linear relationship between the minimum porosity and the critical porosity.

[0095] For any graded rockfill material that has not been subjected to triaxial tests, the above method is used to obtain the fractal dimension of the image and predict its maximum dry density and critical state line position. cs The new mechanical response of the rockfill is obtained.

[0096] Step 5: Obtain the overall mechanical properties distribution of the dam by repeating the above steps at different locations of the dam

[0097] According to the above method, image recognition, gradation analysis and porosity prediction are carried out on the rockfill materials at different positions of the earth-rock dam to obtain the mechanical properties of the rockfill materials at each position.

[0098] The finite element method was used to model the earth-rock dam during construction and fill, and the construction process of layered filling was simulated. The constitutive model and parameters of the dam rockfill were obtained using the above method.

[0099] The mechanical property data of each location are integrated through Kriging interpolation to obtain the overall mechanical property distribution of the earth-rock dam.

[0100] Step 6: Carry out corresponding finite element numerical simulation to predict the deformation law of the dam during construction

[0101] According to the geometric characteristics and construction technology of the earth-rock dam, a finite element model is established. The model should include the dam body, foundation and soil layer around the dam body. Figure 6 As shown in the figure, the dam body is divided by an appropriate meshing method, and the mechanical property parameters of the rockfill at each location are input into the finite element model. The material parameters are mainly the void ratio, but also include elastic modulus, Poisson's ratio, density, etc. The Kriging interpolation method is used to integrate the mechanical property data at each location to ensure the accuracy of the model parameters. The layered filling process of the earth-rock dam is simulated. After each layer is filled, compaction and consolidation analysis is performed. The stress and deformation distribution during construction is calculated to predict the settlement and lateral displacement of the dam body.

[0102] According to the predicted deformation law of the dam during construction, the construction feedback control optimization plan is made

[0103] According to the finite element calculation results, the deformation law of the dam body during construction is analyzed. The deformation concentration area and potential risk points are identified. The critical state porosity ratio e of a dam rockfill material is determined. csThe area with a probability of 5% greater than the set threshold is used as the filling warning area; the ratio of the extreme settlement of the dam body to the current dam height (s / H) is calculated as the deformation warning indicator of the dam body during construction. According to the deformation warning indicators and filling warning areas, the areas that need to be monitored and optimized are determined, and then according to the mechanical property distribution data and deformation laws, a specific construction optimization plan is designed to dynamically adjust the dam construction parameters during the construction process, such as adjusting the filling sequence and compaction method of the rockfill material to ensure the uniformity and density of the rockfill material, and to ensure the construction quality and deformation stability of the earth-rock dam. Finally, numerical simulation and optimization algorithms can be used to verify the feasibility and effectiveness of the optimization plan.

[0104] Specifically, the optimization scheme for dam material parameters includes:

[0105] 1) For the filling warning area, improve the fractal dimension of the gradation curve of the rockfill material in the material yard. For the gradation with a fractal dimension greater than 2.6, reduce the content of fine material in the rockfill material in the subsequent construction plan; for the gradation with a fractal dimension less than 2.3, increase the content of fine material in the rockfill material in the subsequent construction plan;

[0106] 2) If the deformation warning index exceeds the threshold, the number of times the rockfill layer is rolled will be increased in the subsequent construction plan.

[0107] The present invention proposes a feedback control optimization method for earth-rock dam construction taking into account the spatial variability of the gradation of dam-building rockfill materials. It combines unmanned aerial vehicle image acquisition, intelligent image analysis, void ratio analysis and mechanical property prediction with finite element numerical simulation to achieve real-time feedback control of the deformation law of the earth-rock dam during the construction period of the earth-rock dam, providing a scientific basis for the construction quality and deformation control of the earth-rock dam.

[0108] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A feed-back optimization method for earth-rock dam construction considering the spatial variability of dam rockfill material gradation, characterized in that: The following steps are involved: S1: using a drone to take images of the surface of the earth-rock dam, and preprocessing the images; S2: Analyze the pre-processed image through intelligent image analysis algorithm to obtain the grading curve and fractal dimension of the rockfill material; S3: Through the gradation curve and fractal dimension of the rockfill material, the potential dense stacking porosity and fine particle content information of the rockfill material are obtained; S4: obtaining a constitutive relationship related to the state of the rockfill material through information related to the dense stacking porosity and fine particle content and triaxial test results of the calibrated graded rockfill material sample, and predicting the mechanical properties of the rockfill material in the image acquisition area based on the constitutive relationship related to the state of the rockfill material; S5: By repeating S1 to S3 at different positions of the earth-rock dam, the overall mechanical property distribution of the earth-rock dam is obtained; S6: Carry out corresponding finite element numerical simulation according to the distribution law of the overall mechanical properties of the earth-rock dam, predict the deformation law of the dam body during the construction period, and realize the optimization of the feedback control of the earth-rock dam construction based on the deformation law.

2. The method for optimizing the construction of earth-rock dams by feed-back control considering the spatial variability of gradation of dam-building rockfill materials according to claim 1, characterized in that: In S2, the method of analyzing the pre-processed image by an intelligent image analysis algorithm to obtain the grading curve and fractal dimension of the rockfill material includes: The collected RGB color image of the earth-rock dam surface is converted into a grayscale image by floating point method; Applying the Otsu binarization method to binarize the grayscale image and extracting the boundary texture features of the rockfill particles; The "box" dimension method is used to analyze the binary image and extract the image fractal dimension D i , as an indicator to characterize the fine material content of rockfill materials; Using the image fractal dimension D i , the grading curve of rockfill materials is evaluated and calculated.

3. The method for optimizing the construction of earth-rock dams by feed-back control considering the spatial variability of gradation of dam-building rockfill materials according to claim 2, characterized in that: The "box" dimension method is used to analyze the binary image and extract the image fractal dimension D i The methods include: According to the aspect ratio of the image, a rectangular "box" with a characteristic size of Δ is selected to cover the image. The covered pixels must satisfy the pixel aspect ratio of the box and the overall image, and the length and width of the selected image must be greater than 1000 pixels. Count the number of boundaries of particles contained in a "box" with a characteristic size of Δ, N(Δ), and change the value of Δ to obtain the number N(Δ) covered by several boxes; The relationship curve between lgN and lgΔ is plotted in the double logarithmic coordinate system, and the fractal dimension of the filling bin surface image is taken as the negative number of the slope of the curve D i =-k.

4. The method for optimizing the construction of earth-rock dams by feed-back control considering the spatial variability of gradation of dam-building rockfill materials according to claim 2, characterized in that: Using the image fractal dimension D i , the methods for evaluating and calculating the grading curve of rockfill materials include: Where d is the particle size, d M is the maximum particle size of rockfill materials; D is the fractal dimension of rockfill material grading.

5. The method for optimizing the construction of earth-rock dams by feed-back control considering the spatial variability of gradation of rockfill materials for dam construction according to claim 1, characterized in that: In S4, based on the constitutive relationship related to the state of the rockfill material, the method for predicting the mechanical properties of the rockfill material in the image acquisition area includes: or c =M g (x) c (P+1) or g =M g (x) g P-1) Where p and q are the mean effective stress and generalized shear stress, respectively, and η is the stress ratio η=p / q; η c , η p and ηg are the stress ratios of zero expansion, peak value and critical state respectively; N is the material parameter related to the normal consolidation line, k f and n f They are the material parameters, and are plastic volume strain and plastic shear strain, m g and M g is the material parameter related to the plastic potential function, ξ c and g are the state-dependent material parameters and constitutive model parameters respectively; H is the yield surface hardening parameter of the elastoplastic model; the expression of the state parameter of the model is Ψ=ee cs , where e is the current porosity of the material, e cs is the void ratio under critical state; p r and p' r are the reference stresses on the loading and unloading curves, respectively, according to the known porosity ratio e at a certain moment. L and the corresponding normal stress p L get and λ and κ are parameters related to the plastic and elastic moduli of the normal consolidation line, respectively.

6. The method for optimizing the construction of earth-rock dams by feed-back control considering the spatial variability of gradation of dam-building rockfill materials according to claim 1, characterized in that: In S5, by repeating S1 to S3 at different positions of the earth-rock dam, the method for obtaining the overall mechanical property distribution of the earth-rock dam includes: Perform image recognition, gradation analysis and porosity ratio prediction on rockfill materials at different locations of earth-rock dams to obtain mechanical property data of rockfill materials at each location; The mechanical property data of each location are integrated through Kriging interpolation to obtain the overall mechanical property distribution of the earth-rock dam.

7. The feed-back optimization method for earth-rock dam construction considering the spatial variability of dam rockfill material gradation according to claim 1, The method is characterized in that, S6: carrying out corresponding finite element numerical simulation according to the distribution law of the overall mechanical characteristics of the earth-rock dam, predicting the deformation law of the dam body during the construction period, and realizing the method of optimizing the feedback control of the earth-rock dam construction based on the deformation law, including: Collect and analyze the overall mechanical properties distribution data of earth-rock dams, and use the critical state porosity ratio e of a dam-building rockfill material to determine the critical state porosity ratio e of a dam-building rockfill material. cs The area with a probability of 5% greater than the set threshold is used as a filling warning area; According to the distribution law of overall mechanical properties and finite element numerical simulation, the deformation law of the dam body during construction is predicted, and the ratio of the extreme settlement of the dam body to the current dam height s / H greater than the set threshold is used as an early warning indicator of deformation during dam body construction; According to the filling warning area and the deformation warning indicators during the dam construction period, the construction quality feedback information is returned, and the dam material source and construction parameters during the construction process are dynamically adjusted to achieve feedback control optimization of earth-rock dam construction.

Images