Search

CN-121977957-A - Hydraulic engineering-oriented soil curing agent curing effect detection device and method

CN121977957ACN 121977957 ACN121977957 ACN 121977957ACN-121977957-A

Abstract

The invention relates to the technical field of detection of the solidification effect of a soil solidifying agent, in particular to a device and a method for detecting the solidification effect of the soil solidifying agent for hydraulic engineering. The method comprises the steps of firstly, carrying out physical detection on a cured soil sample, specifically comprising hardness detection, fracture resistance detection, water permeability detection, water flow resistance detection and erosion resistance detection, further constructing a three-dimensional image on the cured soil sample, carrying out image analysis on the three-dimensional image by utilizing an image processing technology to obtain a cured effect analysis result of the cured soil sample, and finally analyzing the cured soil sample by using a prediction model to obtain the cured effect analysis result of the cured soil sample.

Inventors

  • CAO XINGYA
  • GUO SHUYU

Assignees

  • 曹兴亚

Dates

Publication Date
20260505
Application Date
20240420

Claims (10)

  1. 1. Soil stabilizer solidification effect detection device towards hydraulic engineering, its characterized in that includes: The physical detection module comprises a shore hardness tester, a universal material testing machine and a variable head penetration device, wherein the shore hardness tester is used for carrying out hardness detection on the surface of a cured soil sample to obtain a first detection value, the universal material testing machine is used for carrying out fracture resistance detection on the surface of the cured soil sample to obtain a second detection value, and the variable head penetration device comprises a variable head soil penetration instrument and is used for carrying out water penetration detection on the surface of the cured soil sample to obtain a third detection value; The water flow scouring resistance detection module is used for carrying out water flow scouring resistance detection on the solidified soil sample, and calculating to obtain a fourth detection value; The erosion resistance detection module is used for detecting erosion resistance of the solidified soil sample and calculating a fifth detection value; the imaging comprehensive detection module of the three-dimensional soil structure comprises a high-resolution industrial CT scanner, wherein the high-resolution industrial CT scanner is used for carrying out omnibearing scanning on the solidified soil sample to obtain three-dimensional image data of an internal structure; The model detection module comprises data set construction, model design, model training and model prediction and is used for obtaining the predicted solidification effect value of the solidified soil sample.
  2. 2. The hydraulic engineering-oriented soil stabilizer solidification effect detection device according to claim 1, wherein the hardness detection comprises selecting 5 test points on the surface of the solidified soil sample, wherein the test points are uniformly distributed on the surface of the solidified soil sample, the pressure head of the shore durometer slowly presses down and applies a specific force to the surface of the solidified soil sample, and when the shore durometer reaches a preset pressure, the pressure is maintained for 3s for indentation formation, and the calculation formula of the first detection value is as follows: Wherein σ is the first detection value, F i is the force applied by the ith test point, A i is the indentation volume of the ith test point, and i represents the number of the test point; The anti-fracture detection comprises the steps of fixing the solidified soil sample on the universal material testing machine, gradually increasing the force until the solidified soil sample breaks, recording the maximum bearing force, and calculating the second detection value, wherein the calculation formula of the second detection value is as follows: Wherein, the Is the second detection value, F max is the maximum force applied, L is the support distance, and b and d are the width and thickness of the sample, respectively; the water permeability detection comprises the step of detecting the water permeability of the solidified soil sample by using the variable head infiltration device to obtain a third detection value, wherein the calculation formula is as follows: wherein k is the third detection value, Q is the water amount passing through the soil sample, A is the cross-sectional area, h is the head loss, L is the sample length, and Δt is the experimental duration.
  3. 3. The hydraulic engineering oriented soil stabilizer solidification effect detection device according to claim 1, wherein the water flow scouring resistance detection module comprises a step of placing the solidified soil sample in a real hydraulic engineering environment according to a set holding time, wherein the holding time is 1440 minutes, and the calculation formula of the fourth detection value is as follows: Wherein E 1 is the fourth detection value, M 1 ,M 2 is the mass value before and after water flow flushing, V 1 ,V 2 is the volume value before and after water flow flushing, V flow is the water flow speed, and T is the flushing time.
  4. 4. The hydraulic engineering-oriented soil stabilizer solidification effect detection device according to claim 1, wherein the erosion resistance detection module comprises three identical solidified soil samples, wherein the solidified soil samples are respectively soaked in various water chemistry environments of a real hydraulic engineering for 1440 minutes, the water chemistry environments comprise a river, a river and a sea, and the calculation formula of the fifth detection value is as follows: wherein E 2 is the fifth detection value, n is 3, the number of experiments is represented, Δm j ,ΔV j is the difference between the mass and the volume before and after the jth soaking, M c ,V c is the initial mass and the volume of the cured soil sample, pH j is the pH of the aqueous chemical environment for the jth soaking, t soaking time, and k 1 ,k 2 is the coefficient to be determined by regression analysis.
  5. 5. The hydraulic engineering-oriented soil stabilizer solidification effect detection device according to claim 1 is characterized in that the image processing technology comprises a V-Net model, wherein the V-Net model comprises model training and model analysis; The model training comprises data set construction and data set division training; the data set construction comprises data acquisition, data preprocessing, data labeling and data enhancement; the data acquisition comprises the steps of collecting 500 original images, wherein the acquisition of the original images scans the cured soil sample by using a high-resolution industrial CT scanner, the cured soil sample covers different curing time and soil types, the data preprocessing comprises the steps of converting the original image data into a PNG format, cutting, contrast enhancement and denoising operation, the data labeling comprises imageJ and is used for labeling the pore structure of the images, and the data enhancement comprises the steps of rotating, scaling and mirror inversion of the original image data by applying an image enhancement technology; the data set division training comprises the steps of dividing a data set into a training set and a testing set according to the proportion of 7:3, training the V-Net model by using the training set, and testing the effect of the V-Net model by using the testing set.
  6. 6. The hydraulic engineering-oriented soil stabilizer solidification effect detection device according to claim 5, wherein the model analysis comprises data preprocessing and image analysis, wherein the data preprocessing comprises inputting three-dimensional image data of the solidified soil sample obtained by the high-resolution industrial CT scanner into the V-Net model, converting the three-dimensional image data into a PNG format, cutting and carrying out contrast enhancement and denoising operations, the image analysis comprises carrying out feature analysis on the three-dimensional image data by the V-Net model, outputting the probability that each three-dimensional pixel belongs to different components of solidified soil, converting the probability into a binary label for determining the category of each three-dimensional pixel and extracting information of a pore structure, wherein the components comprise pores and solidified materials, the pore structure comprises the porosity and pore size distribution, the structural analysis comprises calculation of the porosity and the pore size distribution, the sixth detection value is a porosity value, and the seventh detection value is a pore size distribution value; the calculation formula of the sixth detection value is as follows: Wherein P V is the sixth detection value, X is the number of voxels of the aperture, V X is the volume of a single voxel, and V total is the total volume of the cured soil sample; The calculation formula of the seventh detection value is as follows: Wherein d is the seventh detection value and V hole is the pore volume.
  7. 7. The hydraulic engineering oriented soil curing agent curing effect detection device is characterized in that data set construction comprises data collection and data arrangement, wherein the data collection comprises collection of 1000 pieces of experimental data, curing time, soil type and first detection results of the cured soil samples in the experimental data are recorded, the soil type comprises clay and sand, the first detection results comprise a first detection value, a second detection value, a third detection value, a fourth detection value, a fifth detection value, a sixth detection value and a seventh detection value, the data arrangement comprises arrangement of the collected data into a unified Excel table format, each row represents one sample, and each column represents one feature; The model design comprises a model framework and model realization, wherein the model framework comprises an input layer, a plurality of GRU layers, an attention layer and an output layer, the input layer receives all characteristic data, the GRU layers are used for capturing the dependency relationship of time sequence data, the attention layer is used for strengthening the attention of the model to key information, and the output layer gives a prediction result; the model implementation includes implementing the model using TensorFlow framework, cross entropy loss function, and Adam optimizer; The model training comprises the steps of carrying out normalization processing on the experimental data, dividing the data set into a training set, a verification set and a test set according to the proportion of 7:1:2, inputting the training set into a GRU-Attention model for batch training, evaluating the performance of the model by using the verification set and adjusting model parameters after each training round is finished, and finally testing the GRU-Attention model by using the test set; the model prediction comprises the step of inputting the new curing time, soil type and first detection result of the cured soil sample into the trained GRU-Attention model to obtain a prediction result, wherein the prediction result is the curing effect of the cured soil sample.
  8. 8. A method for detecting the curing effect of a soil curing agent for hydraulic engineering is characterized by comprising the following steps: Carrying out comprehensive detection on the solidified soil sample, wherein the comprehensive detection comprises hardness detection, folding resistance detection, water permeability detection, water flow scouring resistance detection and erosion resistance detection; Constructing a three-dimensional image on the cured soil sample, and performing image analysis on the three-dimensional image by utilizing an image processing technology to obtain a curing effect analysis result of the cured soil sample; And analyzing the cured soil sample by adopting a prediction model to obtain a curing effect analysis result of the cured soil sample.
  9. 9. The method for detecting the curing effect of the soil curing agent for hydraulic engineering according to claim 8, wherein the image processing technology comprises a V-Net model, wherein the V-Net model comprises model training and model analysis; The model training comprises data set construction and data set division training; the data set construction comprises data acquisition, data preprocessing, data labeling and data enhancement; the data acquisition comprises the steps of collecting 500 original images, wherein the acquisition of the original images scans the cured soil sample by using a high-resolution industrial CT scanner, the cured soil sample covers different curing time and soil types, the data preprocessing comprises the steps of converting the original image data into a PNG format, cutting, contrast enhancement and denoising operation, the data labeling comprises imageJ and is used for labeling the pore structure of the images, and the data enhancement comprises the steps of rotating, scaling and mirror inversion of the original image data by applying an image enhancement technology; the data set division training comprises the steps of dividing a data set into a training set and a testing set according to the proportion of 7:3, training the V-Net model by using the training set, and testing the effect of the V-Net model by using the testing set.
  10. 10. The hydraulic engineering-oriented soil curing agent curing effect detection method according to claim 9 is characterized in that the model analysis comprises data preprocessing, image analysis and result analysis, wherein the data preprocessing comprises the steps of inputting three-dimensional image data of a cured soil sample obtained by the high-resolution industrial CT scanner into the V-Net model, converting the three-dimensional image data into a PNG format, cutting and carrying out contrast enhancement and denoising operation, the image analysis comprises the steps of carrying out feature analysis on the three-dimensional image data by the V-Net model, outputting the probability that each three-dimensional pixel belongs to different components of cured soil, converting the probability into a binary label, determining the category of each three-dimensional pixel and extracting information of a pore structure, wherein the components comprise pores and cured materials, and the structural analysis comprises calculation of the porosity and the pore size distribution; The calculation formula of the porosity is as follows: Wherein P V is the porosity, X is the number of voxels of the aperture, V X is the volume of a single voxel, and V total is the total volume of the cured soil sample; The calculation formula of the pore diameter distribution value is as follows: where d is the pore size distribution and V hole is the pore volume.

Description

Hydraulic engineering-oriented soil curing agent curing effect detection device and method Technical Field The invention relates to the technical field of detection of the solidification effect of a soil solidifying agent, in particular to a device and a method for detecting the solidification effect of the soil solidifying agent for hydraulic engineering. Background In the construction and maintenance of hydraulic engineering, it is important to ensure the stability and durability of the infrastructure such as dams, river levees and irrigation systems. Soil is used as one of basic construction materials of hydraulic engineering, and physical and chemical properties of the soil directly influence the safety, stability and efficiency of the engineering. Because hydraulic engineering is often faced with complicated and changeable environmental conditions, such as water flow erosion, temperature and humidity change, chemical corrosion and the like, soil is easy to erode, compress, displace and the like, and the safe operation of the engineering is further affected. Therefore, soil curing agents are used for improving soil performance and enhancing erosion resistance and bearing capacity of the soil, and the soil curing agents become an important technical means in hydraulic engineering. Currently, the use effect of soil solidifying agents is affected by many factors including the kind of solidifying agent, the type of soil, environmental conditions, etc., and thus an accurate and effective detection method is required to evaluate the solidifying effect. Traditional soil solidification effect detection methods focus on basic physical property tests in laboratories, such as compressive strength and permeability coefficient. In the prior study, scholars Liu Xiaojin are introduced into the current state simple analysis of durability study of curing agent and curing soil, summarize the study methods for improving the durability of curing soil such as water stability, freezing resistance, dry and wet cycle resistance, shrinkage cracking resistance and the like of the curing soil, and put forward partial study suggestions, and scholars ' scholars use the scholars in the detection and evaluation of the in-situ curing effect of a shallow soft soil roadbed in the ' detection and evaluation of the in-situ curing effect of a shallow soft soil roadbed ' of a certain engineering, and detect and evaluate the curing effect of the in-situ curing soft foundation of a test area by adopting a core drilling method and a standard penetration test. At present, a plurality of methods for detecting the curing effect of a soil curing agent still have some defects, the Chinese patent of the prior art application number CN202110503994.9 discloses a test device and a test method for detecting the curing effect of MICP on polluted soil, the device is used for carrying out microorganism-induced calcium carbonate precipitation and solidification on polluted soil to adsorb metal ions so as to achieve the aim of solidifying the soil, and carrying out seepage test, water head measurement, stress strain test, calcium carbonate content detection and ion concentration detection on a solidified sample through the device so as to detect the effect of MICP on solidifying the polluted soil. The Chinese patent of the prior art application number CN201711088180.3 discloses a device and a method for rapidly testing the long-term stability of a solidified body under a polluted soil well, which adopt similar principles and simplified means according to the conditions under a metal ore or a coal mine well, the device mainly comprises a gas sample chamber, a test body box, a water sample tube, a pressure stabilizing system and a monitoring system, and can be used for rapidly detecting the stability of the polluted soil solidified body and the influence on the underground environment in a laboratory. When the solidification effect of the soil solidifying agent is detected in the prior art, the water permeability and the hardness of soil are detected mainly by means of physical and chemical testing methods in a laboratory, comprehensive evaluation of the stability and the durability of the solidified soil under the special environmental conditions of hydraulic engineering is lacked, such as long-term water flow flushing and change of the water chemical environment, so that in the actual hydraulic engineering application scene, the performance display of the soil solidifying body may deviate from the testing result in the laboratory, and the detection accuracy and the detection reliability are affected. Aiming at the problems in the prior art, the invention provides a device and a method for detecting the curing effect of a soil curing agent for hydraulic engineering. Disclosure of Invention The invention aims to provide a device and a method for detecting the solidifying effect of a soil solidifying agent for hydraulic engineering, which are used for firstly c