Automatic proportioning method and device for conductive grouting tracing material

Abstract

The invention relates to the technical field of pretreatment of forming materials, and discloses an automatic proportioning method and device for a conductive grouting tracing material. The technical problem that in the prior art, in the grouting reinforcement process of a foundation pit, the design of the proportion of a conductive tracing material lacks an automatic means, and consequently the electrical resistivity tomography (ERT) effect is poor is solved. The method comprises the following steps: data acquisition and collection; setting parameters and formulating a construction scheme; the grouting ratio is controlled, and construction measures are determined; automatic matching generation and output; and automatic real-time feedback and correction are realized. Through the automatic matching design method, the construction efficiency and the monitoring precision of the foundation pit grouting reinforcement project are remarkably improved, and the necessity of manual intervention and the potential construction risk are reduced. A high-precision sensor and an advanced data analysis tool are combined, dynamic monitoring and intelligent feedback adjustment of the tomography effect in the whole construction process are achieved, and service is provided for engineering quality and construction safety.

Description

Technical Field

[0001] The invention relates to the technical field of pretreatment of molding materials, and in particular to an automatic proportioning method and device for conductive grouting tracing materials. Background Art

[0002] Grouting reinforcement of foundation pits is a common soil reinforcement method that increases the stability and bearing capacity of the soil by injecting cement slurry or other curing agents into the ground. In this process, real-time monitoring of grouting effects and soil changes is the key to ensuring project safety and quality. Electrical Resistivity Tomography (ERT), as an efficient geophysical detection method, can provide high-resolution underground structure images by measuring the resistivity distribution of underground media, thereby monitoring changes in the grouting process in real time.

[0003] The core of ERT technology is to use the potential difference generated when the current passes through the underground medium to infer the distribution of underground resistivity. Due to the large difference in resistivity between different media, ERT can sensitively capture changes in underground structure and fluid flow. However, in practical applications, due to the complexity and heterogeneity of the underground medium, the resolution and accuracy of ERT images may be limited.

[0004] In order to improve the resolution and accuracy of ERT technology, conductive tracer materials are introduced into the grouting process. These materials include carbon black, graphite, conductive polymers, etc., which have good conductivity. When these conductive tracer materials are injected into the ground, they can significantly change the resistivity distribution of the target area, thereby enhancing the imaging effect of ERT. For example, carbon black and graphite have high conductivity and stability, and can significantly change the resistivity of the injected area at low concentrations. Conductive polymers, such as polyaniline (PANI) and polypyrrole (PPy), have good conductivity and chemical stability, and can maintain stable performance under different environmental conditions.

[0005] However, the optimal ratio of conductive tracer materials is affected by many factors, including geological background, soil characteristics, physical and chemical properties of grouting materials, and construction conditions. Under different geological backgrounds, soil parameters such as porosity, water content, and resistivity will significantly affect the diffusion and effect of conductive tracer materials. Therefore, in order to achieve the best ERT imaging effect, the ratio of conductive tracer materials must be accurately adjusted according to the specific geological background and construction conditions.

[0006] Traditional ratio design mainly relies on experience and repeated trials, which is not only time-consuming and labor-intensive, but also difficult to ensure accuracy and stability. In order to improve construction efficiency and monitoring accuracy, it is necessary to design a system that can automatically generate the optimal ratio, and quickly and accurately determine the ratio of conductive tracer materials through intelligent algorithms and data-driven methods. Realize the pretreatment of molding materials; automatic control and proportional mixing of conductive tracer materials. Summary of the invention

[0007] In view of the above technical problems, the present disclosure provides an automated proportioning method and device for conductive grouting tracer materials, which solves the technical problem in the prior art that there is a lack of automated means for the design of conductive tracer material proportioning during foundation pit grouting reinforcement, resulting in poor electrical resistivity tomography (ERT) effect.

[0008] The present invention proposes an automated proportion design method for on-site conductive grouting tracer materials based on convolutional neural network (CNN). The system can automatically calculate the optimal conductive material proportion based on the geological background information input by the user, using relevant geological data and advanced deep learning algorithms to ensure the best resistivity tomography effect. In addition, during the construction process, real-time feedback of the imaging effect can be achieved. When the imaging effect is not good, the automatic dynamic correction of the proportion can be achieved, thereby realizing the automated proportion design of on-site conductive grouting tracer materials with real-time dynamic correction. This system can not only significantly improve the efficiency and accuracy of the proportion design, but also adapt to the complex changes under different geological backgrounds and construction conditions, providing strong technical support for foundation pit grouting reinforcement projects.

[0009] According to one aspect of the present disclosure, there is provided an automated proportioning method for a conductive grouting tracer material, comprising the following steps: (1) Data acquisition and collection: Before starting the conductive tracer grouting construction, data collection is first carried out, and engineering tests are carried out to complete the parameter calibration of numerical analysis. Through numerical analysis, a three-dimensional finite element calculation model of the strata and buildings within the construction influence range is established to simulate the foundation pit grouting construction process; (2) Parameter setting and construction plan formulation: Before construction, parameter setting and construction plan formulation are carried out based on data collection and simulation analysis results; (3) Control the grouting ratio and determine the construction measures: Based on the numerical simulation results and construction monitoring data, determine the conductive tracer material ratio and implement construction measures, including grouting pressure, grouting speed, and grouting volume; (4) Automated ratio generation and output: After entering the operation stage, the user inputs geological background information and calculates the conductive tracer material ratio through a convolutional neural network; (5) Automated real-time feedback and correction: During the construction process, the proportion design of the on-site conductive grouting tracer material realizes real-time imaging dynamic feedback. When the imaging effect is not good, the conductive material proportion is automatically corrected through the optimization algorithm, thereby realizing dynamic control and correction of the conductive material proportion throughout the construction process, thereby achieving the imaging effect.

[0010] In some embodiments of the present disclosure, the simulation of the foundation pit grouting construction process in step (1) includes soil deformation, stress distribution, and pore water pressure changes; the horizontal displacement, settlement, and stress distribution of the stratum are calculated, and combined with the deformation and stress conditions of the building structure, the influence mechanism of the foundation pit grouting construction on the stratum and the structure is revealed; at the same time, the stratum parameters under different geological backgrounds, including porosity, water content, soil type, resistivity, and construction conditions, including grouting pressure and grouting volume, are collected, and the input parameters and output results of each simulation analysis are recorded, so as to establish a geological background database.

[0011] In some embodiments of the present disclosure, the parameter setting in step (2) further includes the following steps: establishing and analyzing a three-dimensional finite element model through numerical simulation software to obtain the deformation and stress distribution of the formation; and determining construction parameters based on numerical simulation and historical data, including: grouting pressure, grouting volume, settlement, and a safe range of horizontal displacement.

[0012] In some embodiments of the present disclosure, the construction plan formulation in step (2) includes: monitoring point arrangement, monitoring frequency, data collection and processing methods.

[0013] In some embodiments of the present disclosure, determining the proportion of the conductive tracer material in step (3) includes the following steps: by collecting construction and experimental data and combining with complete training of convolutional neural networks, a database and prediction model of the proportion of conductive tracer materials under different geological backgrounds are established; during the construction process, the proportion of the conductive tracer material is adjusted according to the real-time monitoring data and the prediction results of the convolutional neural network model to ensure the resistivity tomography effect; a mapping relationship between the input geological background information and the output proportion is constructed by using a convolutional neural network; the input end parameters include soil porosity (φ), water content (θ), soil type (T), and soil resistivity (ρ); the output end parameters are in an integrated slurry storage tank, and for conductive materials, include: the addition amount of carbon black, graphite, conductive polymer and cement mortar, binder, and dispersant; the proportion of the conductive material is obtained and gradually optimized by combining the above-mentioned input end parameters with the convolutional neural network; The input layer of the convolutional neural network receives the geological background information obtained from the data collection stage and passes it to multiple convolutional layers for feature extraction; each convolutional layer uses a convolution kernel to perform a convolution operation on the input data to extract a feature map; the pooling layer reduces the size of the feature map through a downsampling operation, and after multiple convolution and pooling operations, a high-dimensional feature representation is obtained; the high-dimensional feature is input into the fully connected layer, and the ratio of the conductive tracer material is calculated through linear transformation and activation function; In the convolutional neural network model, the convolution operation formula is as follows: (1) Convolution operation is the core of feature extraction. f is the input feature map, g is the convolution kernel, t It is a characteristic response of time or spatial location; is the integration variable; Indicates that the convolution kernel g Pan to t The position of , so as to realize feature extraction; Through convolution operation, local features are extracted from the input data; The pooling process of the convolutional neural network uses the maximum pooling operation to highlight and distinguish the local features of the tomographic image. The formula is: (2) In the above formula y i,j The output feature map after pooling is at position ( i,j ) value; Max m,n : means taking the maximum value within a given window; x i+m,j+n : Input feature map at position ( i + m,j + n ) is the pixel value or feature value within the pooling window; The pooling operation is used for downsampling, reducing the size of the feature map and retaining important features; the maximum pooling selects the maximum value in the local area and retains significant features, thereby improving the generalization ability of the model. The pooled feature map is passed to the fully connected layer for linear transformation and activation to calculate the ratio of the conductive tracer material; The final output ratio is calculated by the following formula: (3) in, is the resistivity after mixing, For the i The ratio of materials, For the i The resistivity of a material; Describe the relationship between the resistivity of the conductive tracer material and the resistivity and ratio of its components, and ensure the resistivity tomography effect by optimizing the ratio; Assume the error function formula is as follows: (4) The mean square error is used to measure the difference between the predicted value and the actual value. is the actual output, is the expected output, N is the number of samples; by minimizing the error function, the prediction ability of the model is optimized; The network weights and biases are adjusted through the back-propagation algorithm, and the formula is as follows: (5) in, is the learning rate, E is the error function, is the network weight; back propagation updates the weight by calculating the gradient of the error relative to the weight to reduce the prediction error and improve the prediction accuracy; During the training phase of the convolutional neural network, historical data is used for training to improve the accuracy and robustness of the model; during the model training process, the network structure and parameters are adjusted to optimize the model performance; and the cross-validation method is used to verify the generalization ability of the model.

[0014] In some embodiments of the present disclosure, the conductive tracer material ratio in step (4) includes the ratio of cement mortar, binder, dispersant, and conductive material to ensure the resistivity tomography effect; the step of generating the conductive tracer material ratio includes a. user input: the user inputs the geological background information of the construction site, including soil parameters, porosity, water content, soil type, and resistivity; b. database retrieval: retrieving similar geological background data as an initial reference; c. convolutional neural network calculation: extracting the features of the input data through multi-layer convolution, pooling, and fully connected layers to perform ratio prediction; d. output: outputting the conductive tracer material ratio and generating a corresponding construction plan, controlling the application amount of the conductive material or grouting additive storage tank through an integrated control circuit, and transporting it into the integrated slurry storage tank through an integrated pipeline for mixing and stirring to form a conductive tracer composite material slurry for construction.

[0015] An automated proportioning device for on-site conductive grouting tracer materials is used to implement the automated proportioning method for the conductive grouting tracer materials described above, comprising a server through which a user can input data and parameters, wherein an automated proportioning design and processing system is arranged in the server, wherein the server is electrically connected to an integrated control circuit, wherein a signal input end of the integrated control circuit is connected to a tomography monitoring device to provide feedback on imaging effects, wherein a signal output end of the integrated control circuit is connected to an integrated slurry storage tank and a plurality of material sub-packaging storage tanks to implement proportioning control to transport materials in the plurality of material sub-packaging storage tanks to the integrated slurry storage tank, wherein a stirring blade for stirring the conductive tracer composite material slurry is installed in the integrated slurry storage tank.

[0016] In some embodiments of the present disclosure, the integrated slurry storage tank is funnel-shaped on the top and an opening and closing device is installed on the bottom.

[0017] In some embodiments of the present disclosure, a plurality of fan-shaped funnels are arranged in a circumferential array above the integrated slurry storage tank, and at least one weighing assembly is arranged above each fan-shaped funnel.

[0018] In some embodiments of the present disclosure, the weighing assembly includes a first weighing device, and a second weighing device is mounted below the first weighing device.

[0019] In some embodiments of the present disclosure, the material packaging storage tank is arranged in a circular array, a screw conveyor is installed at the bottom, and a weighing component is received below the outlet of the screw conveyor.

[0020] The beneficial effects of the present invention are: 1. During the construction process, the ratio of the conductive tracer material is dynamically adjusted through real-time monitoring and feedback mechanisms. The system will automatically generate the optimal ratio based on real-time monitoring data and model prediction results to ensure the accuracy and effectiveness of the ERT effect. At the same time, the real-time feedback of the imaging effect can be used to automatically correct the ratio to optimize the imaging effect. 2. In addition, through the continuously updated database, the system can continuously optimize and improve its prediction ability and scope of application to meet the needs of different geological conditions and construction environments.

[0021] 2. It can realize the selection and generation of tracer materials and proportions suitable for the construction background, and can realize real-time monitoring of tomographic imaging effects. The material proportions can be further adjusted through the optimization algorithm, and then the iterative optimization of the tomographic imaging inversion effect can be gradually realized, so as to provide the material proportion with the best imaging effect for the actual project.

[0022] 3. Through the automated mix design method, the construction efficiency and monitoring accuracy of the foundation pit grouting reinforcement project have been significantly improved, reducing the necessity of manual intervention and potential construction risks. Combined with high-precision sensors and advanced data analysis tools, dynamic monitoring and intelligent feedback adjustment of tomography effects throughout the construction process have been achieved, serving the project quality and construction safety.

[0023] 4. Users input data and parameters through the server, and the customized proportioning process is more flexible and can adapt to different application requirements. The tomographic imaging monitoring equipment provides real-time imaging effect feedback, optimizes the proportioning scheme, and monitors the state of the slurry to meet the expected conductive properties. The top of the integrated slurry storage tank is funnel-shaped, and the design of multiple fan-shaped funnels in a circular array facilitates the inflow of materials; the opening and closing devices installed below can control the outflow of the slurry, which is convenient for adjusting the flow rate or stopping the output. The combined use of the first weigher and the second weigher can achieve more accurate weight measurement; a weighing assembly is set above each fan-shaped funnel to accurately control the amount of each material added, and the material distribution storage tank in a circular array enables various materials to be stored in an orderly manner for easy access. At the same time, the screw conveyor installed at the bottom can smoothly deliver the material into the integrated slurry storage tank. The stirring blades installed in the integrated slurry storage tank can effectively mix the various components. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flowchart of the automated proportioning method of the conductive grouting tracer material; Figure 2 It is a schematic diagram of the structure of the automatic proportioning device of the on-site conductive grouting tracer material; Figure 3 This is a schematic diagram of the structure of an integrated slurry storage tank and a material packaging storage tank; Figure 4 This is a schematic diagram of the material packaging storage tank structure; Figure 5 This is a schematic diagram of the integrated slurry storage tank structure; Figure 6 for Figure 5 Sectional view of the middle AA plane; The names of the components in the figure are: 1. Automatic proportioning design and processing system; 2. Tomography monitoring equipment; 3. Integrated control circuit; 4. Integrated slurry storage tank; 5. Conductive tracer composite material slurry; 6. Stirring blades; 7. Material packaging storage tank; 8. Opening and closing device; 9. Fan-shaped funnel; 10. Weighing assembly; 11. First weighing device; 12. Second weighing device; 13. Screw conveyor. DETAILED DESCRIPTION

[0025] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Example 1

[0026] This example discloses an automated proportioning method for conductive grouting tracer materials, see Figures 1 to 6 ; comprising the following steps: (1) Data acquisition and collection: Before starting the conductive tracer grouting construction, data collection is first carried out, and engineering tests are carried out to complete the parameter calibration of numerical analysis. Through numerical analysis, a three-dimensional finite element calculation model of the strata and buildings within the construction influence range is established to simulate the foundation pit grouting construction process; (2) Parameter setting and construction plan formulation: Before construction, parameter setting and construction plan formulation are carried out based on data collection and simulation analysis results; (3) Control the grouting ratio and determine the construction measures: Based on the numerical simulation results and construction monitoring data, determine the conductive tracer material ratio and implement construction measures, including grouting pressure, grouting speed, and grouting volume; (4) Automated ratio generation and output: After entering the operation stage, the user inputs geological background information and calculates the conductive tracer material ratio through a convolutional neural network; (5) Automated real-time feedback and correction: During the construction process, the proportion design of the on-site conductive grouting tracer material realizes real-time imaging dynamic feedback. When the imaging effect is not good, the conductive material proportion is automatically corrected through the optimization algorithm, thereby realizing dynamic control and correction of the conductive material proportion throughout the construction process, thereby achieving the imaging effect.

[0027] In step (1), the simulation of the foundation pit grouting construction process includes soil deformation, stress distribution, and pore water pressure changes; the horizontal displacement, settlement, and stress distribution of the stratum are calculated, and combined with the deformation and stress conditions of the building structure, the influence mechanism of the foundation pit grouting construction on the stratum and structure is revealed; at the same time, the stratum parameters under different geological backgrounds are collected, including porosity, water content, soil type, resistivity, and construction conditions, including grouting pressure and grouting volume, and the input parameters and output results of each simulation analysis are recorded, so as to establish a geological background database.

[0028] The parameter setting in step (2) also includes the following steps: establishing and analyzing a three-dimensional finite element model through numerical simulation software to obtain the deformation and stress distribution of the stratum; and determining the construction parameters based on numerical simulation and historical data, including: grouting pressure, grouting volume, settlement, and a safe range of horizontal displacement.

[0029] The construction plan in step (2) includes: monitoring point layout, monitoring frequency, data collection and processing methods.

[0030] Determining the proportion of the conductive tracer material in step (3) includes the following steps: by collecting construction and experimental data and combining with complete training of convolutional neural networks, a database and prediction model of the proportion of conductive tracer materials under different geological backgrounds are established; during the construction process, according to real-time monitoring data and prediction results of the convolutional neural network model, the proportion of the conductive tracer material is adjusted to ensure the resistivity tomography effect; a mapping relationship between input geological background information and output proportion is constructed by using a convolutional neural network; the input end parameters include soil porosity (φ), water content (θ), soil type (T), soil resistivity (ρ), and the output end parameters are in an integrated slurry storage tank, and for conductive materials, include: the addition amount of carbon black, graphite, conductive polymer and cement mortar, binder, and dispersant; the proportion of the conductive material is obtained and gradually optimized by combining the above-mentioned input end parameters with the convolutional neural network; The input layer of the convolutional neural network receives the geological background information obtained from the data collection stage and passes it to multiple convolutional layers for feature extraction; each convolutional layer uses a convolution kernel to perform a convolution operation on the input data to extract a feature map; the pooling layer reduces the size of the feature map through a downsampling operation, and after multiple convolution and pooling operations, a high-dimensional feature representation is obtained; the high-dimensional feature is input into the fully connected layer, and the ratio of the conductive tracer material is calculated through linear transformation and activation function; In the convolutional neural network model, the convolution operation formula is as follows: (1) Convolution operation is the core of feature extraction. f is the input feature map, g is the convolution kernel, t It is a characteristic response of time or spatial location; is the integration variable; Indicates that the convolution kernel g Pan to t The position of , so as to realize feature extraction; Through convolution operation, local features are extracted from the input data; The pooling process of the convolutional neural network uses the maximum pooling operation to highlight and distinguish the local features of the tomographic image. The formula is: (2) In the above formula y i,j The output feature map after pooling is at position ( i,j ) value; Max m,n : means taking the maximum value within a given window; x i+m,j+n : Input feature map at position ( i + m,j + n ) is the pixel value or feature value within the pooling window; The pooling operation is used for downsampling, reducing the size of the feature map and retaining important features; the maximum pooling selects the maximum value in the local area and retains significant features, thereby improving the generalization ability of the model. The pooled feature map is passed to the fully connected layer for linear transformation and activation to calculate the ratio of the conductive tracer material; The final output ratio is calculated by the following formula: (3) in, is the resistivity after mixing, For the i The ratio of materials, For the i The resistivity of a material; Describe the relationship between the resistivity of the conductive tracer material and the resistivity and ratio of its components, and ensure the resistivity tomography effect by optimizing the ratio; Assume the error function formula is as follows: (4) The mean square error is used to measure the difference between the predicted value and the actual value. is the actual output, is the expected output, N is the number of samples; by minimizing the error function, the prediction ability of the model is optimized; The network weights and biases are adjusted through the back-propagation algorithm, and the formula is as follows: (5) in, is the learning rate, E is the error function, is the network weight; back propagation updates the weight by calculating the gradient of the error relative to the weight to reduce the prediction error and improve the prediction accuracy; During the training phase of the convolutional neural network, historical data is used for training to improve the accuracy and robustness of the model; during the model training process, the network structure and parameters are adjusted to optimize the model performance; and the cross-validation method is used to verify the generalization ability of the model.

[0031] In step (4), the conductive tracer material ratio includes the ratio of cement mortar, binder, dispersant and conductive material to ensure the resistivity tomography effect; the generation steps of the conductive tracer material ratio include a. user input: the user inputs the geological background information of the construction site, including soil parameters, porosity, water content, soil type and resistivity; b. database retrieval: retrieving similar geological background data as an initial reference; c. convolutional neural network calculation: extracting the features of the input data through multi-layer convolution, pooling and fully connected layers to predict the ratio; d. output: outputting the conductive tracer material ratio and generating a corresponding construction plan, controlling the application amount of the conductive material or grouting additive storage tank through the integrated control circuit, and transporting it into the integrated slurry storage tank through the integrated pipeline for mixing and stirring to form a conductive tracer composite material slurry for construction.

[0032] An automated proportioning device for on-site conductive grouting tracer materials is used to realize an automated proportioning method for conductive grouting tracer materials, including a server through which users can input data and parameters, an automated proportioning design and processing system 1 is arranged in the server, the server is electrically connected to an integrated control circuit 3, a signal input end of the integrated control circuit 3 is connected to a tomography monitoring device 2 to feedback imaging effects, a signal output end of the integrated control circuit 3 is connected to an integrated slurry tank 4 and a plurality of material sub-packaging storage tanks 7 to realize proportioning control to transport materials in the plurality of material sub-packaging storage tanks 7 to the integrated slurry tank 4, a stirring blade 6 for stirring the conductive tracer composite material slurry is installed in the integrated slurry tank 4.

[0033] The integrated slurry storage tank 4 is funnel-shaped on the top and an opening and closing device 8 is installed on the bottom.

[0034] A plurality of fan-shaped funnels 9 are arranged in a circumferential array above the integrated slurry storage tank 4 , and at least one weighing component 10 is arranged above each fan-shaped funnel 9 .

[0035] The weighing assembly 10 comprises a first weighing device 11 , and a second weighing device 12 is mounted below the first weighing device 11 .

[0036] The material packaging storage tank 7 is arranged in a circular array, and a screw conveyor 13 is installed at the bottom. The weighing assembly 10 is received below the outlet of the screw conveyor 13 .

[0037] During the work process, Figures 2 to 6As shown, the user inputs data and parameters through the automated proportioning design and processing system 1, the tomography monitoring device 2 feeds back the imaging effect, the proportioning control is realized through the integrated control circuit 3, the material is transported to the integrated slurry storage tank 4 through a plurality of different material sub-packaging storage tanks 7, and the conductive tracer composite material slurry 5 is formed by stirring of the stirring blade 6. The material in the material sub-packaging storage tank 7 is transported to the first weighing device 11 through the screw conveyor 13, and then falls to the second weighing device 12, flows to the integrated slurry storage tank 4 through the fan-shaped funnel 9, and the slurry outflow is controlled by the opening and closing device 8.

[0038] During the construction process, the ratio of conductive tracer materials is dynamically adjusted through real-time monitoring and feedback mechanisms. The system will automatically generate the optimal ratio based on real-time monitoring data and model prediction results to ensure the accuracy and effectiveness of the ERT effect. At the same time, the system can automatically correct the ratio based on the real-time feedback of the imaging effect to optimize the imaging effect. In addition, through the continuously updated database, the system can continuously optimize and improve its prediction ability and scope of application to meet the needs of different geological conditions and construction environments.

[0039] The present invention can realize that the user inputs relevant parameters such as relevant geological background in the automatic grouting tracer material ratio design and processing system, and then after the aforementioned sufficient convolutional neural network training, it can realize the selection and generation of tracer materials and ratios suitable for the construction background under the conditions of combining tomographic imaging effects and considering the improvement of grouting reinforcement mechanical properties. The integrated circuit and the material quantitative intelligent transportation control system are used to carry out quantitative transportation according to the corresponding ratio, and finally the ideal ratio slurry material obtained according to the automatic grouting tracer material ratio design is obtained in the composite material slurry storage tank, and the tomographic imaging effect can be monitored in real time. When the imaging effect is not good, the material ratio can be further adjusted through the optimization algorithm, and then the iterative optimization of the tomographic imaging inversion effect can be gradually realized, so as to provide the material ratio with the best imaging effect for the actual project.

[0040] The present invention significantly improves the construction efficiency and monitoring accuracy of foundation pit grouting reinforcement projects through an automated mix design method, reducing the necessity of manual intervention and potential construction risks. Combined with high-precision sensors and advanced data analysis tools, dynamic monitoring and intelligent feedback adjustment of tomography effects during the entire construction process are achieved, serving the project quality and construction safety.

[0041] Although some preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0042] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of this application and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An automated proportioning method for conductive grouting tracer materials, characterized in that: The steps include: (1) Data acquisition and collection: Before starting the conductive tracer grouting construction, data collection is first carried out, and engineering tests are carried out to complete the parameter calibration of numerical analysis. Through numerical analysis, a three-dimensional finite element calculation model of the strata and buildings within the construction influence range is established to simulate the foundation pit grouting construction process; (2) Parameter setting and construction plan formulation: Before construction, parameter setting and construction plan formulation are carried out based on data collection and simulation analysis results; (3) Control the grouting ratio and determine the construction measures: Based on the numerical simulation results and construction monitoring data, determine the conductive tracer material ratio and implement construction measures, including grouting pressure, grouting speed, and grouting volume; (4) Automated ratio generation and output: After entering the operation stage, the user inputs geological background information and calculates the conductive tracer material ratio through a convolutional neural network; (5) Automated real-time feedback and correction: During the construction process, the proportion design of the on-site conductive grouting tracer material realizes real-time imaging dynamic feedback. When the imaging effect is not good, the conductive material proportion is automatically corrected through the optimization algorithm, thereby realizing dynamic control and correction of the conductive material proportion throughout the construction process, thereby achieving the imaging effect.

2. The automated proportioning method of the conductive grouting tracer material according to claim 1, characterized in that: The simulation of the foundation pit grouting construction process in step (1) includes soil deformation, stress distribution, and pore water pressure changes; the horizontal displacement, settlement, and stress distribution of the stratum are calculated, and the influence mechanism of the foundation pit grouting construction on the stratum and the structure is revealed in combination with the deformation and stress conditions of the building structure; at the same time, the stratum parameters under different geological backgrounds are collected, including porosity, water content, soil type, resistivity, and construction conditions, including grouting pressure and grouting volume, and the input parameters and output results of each simulation analysis are recorded, so as to establish a geological background database.

3. The automated proportioning method of the conductive grouting tracer material according to claim 1, characterized in that: The parameter setting in step (2) also includes the following steps: establishing and analyzing a three-dimensional finite element model through numerical simulation software to obtain the deformation and stress distribution of the stratum; and determining construction parameters based on numerical simulation and historical data, including: grouting pressure, grouting volume, settlement, and a safe range of horizontal displacement.

4. The automated proportioning method of the conductive grouting tracer material according to claim 1, characterized in that: The construction plan formulation in step (2) includes: monitoring point arrangement, monitoring frequency, data collection and processing methods.

5. The automated proportioning method of the conductive grouting tracer material according to claim 1, characterized in that: Determining the proportion of the conductive tracer material in step (3) includes the following steps: by collecting construction and experimental data and combining with complete training of convolutional neural networks, a database and prediction model of the proportion of conductive tracer materials under different geological backgrounds are established; during the construction process, according to real-time monitoring data and prediction results of the convolutional neural network model, the proportion of the conductive tracer material is adjusted to ensure the resistivity tomography effect; a mapping relationship between input geological background information and output proportion is constructed by using a convolutional neural network; the input end parameters include soil porosity (φ), water content (θ), soil type (T), soil resistivity (ρ), and the output end parameters are in an integrated slurry storage tank, and for conductive materials, include: the addition amount of carbon black, graphite, conductive polymer and cement mortar, binder, and dispersant; the proportion of the conductive material is obtained and gradually optimized by combining the above-mentioned input end parameters with the convolutional neural network; The input layer of the convolutional neural network receives the geological background information obtained from the data collection stage and passes it to multiple convolutional layers for feature extraction; each convolutional layer uses a convolution kernel to perform a convolution operation on the input data to extract a feature map; the pooling layer reduces the size of the feature map through a downsampling operation, and after multiple convolution and pooling operations, a high-dimensional feature representation is obtained; the high-dimensional feature is input into the fully connected layer, and the ratio of the conductive tracer material is calculated through linear transformation and activation function; In the convolutional neural network model, the convolution operation formula is as follows: （1） Convolution operation is the core of feature extraction. f is the input feature map, g is the convolution kernel, t It is a characteristic response of time or spatial location; is the integration variable; Indicates that the convolution kernel g Pan to t The position of , so as to realize feature extraction; Through convolution operation, local features are extracted from the input data; The pooling process of the convolutional neural network uses the maximum pooling operation to highlight and distinguish the local features of the tomographic image. The formula is: （2） In the above formula y i,j The output feature map after pooling is at position ( i,j ) value; Max m,n : means taking the maximum value within a given window; x i+m,j+n : Input feature map at position ( i + m,j + n ) is the pixel value or feature value within the pooling window; The pooling operation is used for downsampling, reducing the size of the feature map and retaining important features; the maximum pooling selects the maximum value in the local area and retains significant features, thereby improving the generalization ability of the model. The pooled feature map is passed to the fully connected layer for linear transformation and activation to calculate the ratio of the conductive tracer material; The final output ratio is calculated by the following formula: （3） in, is the resistivity after mixing, For the i The ratio of materials, For the i The resistivity of a material; describe the relationship between the resistivity of the conductive tracer material and the resistivity and ratio of its components, and ensure the resistivity tomography effect by optimizing the ratio; Assume the error function formula is as follows: （4） The mean square error is used to measure the difference between the predicted value and the actual value. is the actual output, is the expected output, N is the number of samples; by minimizing the error function, the prediction ability of the model is optimized; The network weights and biases are adjusted through the back-propagation algorithm, and the formula is as follows: （5） in, is the learning rate, E is the error function, is the network weight; back propagation updates the weight by calculating the gradient of the error relative to the weight to reduce the prediction error and improve the prediction accuracy; During the training phase of the convolutional neural network, historical data is used for training to improve the accuracy and robustness of the model; during the model training process, the network structure and parameters are adjusted to optimize the model performance; and the cross-validation method is used to verify the generalization ability of the model.

6. The automated proportioning method of the conductive grouting tracer material according to claim 1, characterized in that: The conductive tracer material ratio in step (4) includes the ratio of cement mortar, binder, dispersant and conductive material to ensure the resistivity tomography effect; the conductive tracer material ratio generation steps include a. user input: the user inputs the geological background information of the construction site, including soil parameters, porosity, water content, soil type and resistivity; b. database search: search for similar geological background data as an initial reference; c. convolutional neural network calculation: extract the features of the input data through multi-layer convolution, pooling and fully connected layers to perform ratio prediction; d. Output: Output the proportion of conductive tracer materials and generate a corresponding construction plan. The amount of conductive material or grouting additive storage tank applied is controlled by an integrated control circuit, and transported into the integrated slurry storage tank through an integrated pipeline for mixing and stirring to form a conductive tracer composite material slurry for construction.

7. An automated proportioning device for on-site conductive grouting tracer material, used to implement the automated proportioning method for conductive grouting tracer material according to any one of claims 1 to 6, characterized in that: The invention comprises a server through which a user can input data and parameters, wherein an automatic proportioning design and processing system is arranged in the server, wherein the server is electrically connected to an integrated control circuit, wherein a signal input end of the integrated control circuit is connected to a tomographic imaging monitoring device to provide feedback on imaging effects, and a signal output end of the integrated control circuit is connected to an integrated slurry storage tank and a plurality of material sub-packaging storage tanks to implement proportioning control to transport materials in the plurality of material sub-packaging storage tanks to the integrated slurry storage tank, wherein a stirring blade for stirring the conductive tracer composite material slurry is installed in the integrated slurry storage tank.

8. The automatic proportioning device for on-site conductive grouting tracer material according to claim 7, characterized in that: The upper part of the integrated slurry storage tank is in a funnel shape, and an opening and closing device is installed at the lower part.

9. The automatic proportioning device for on-site conductive grouting tracer material according to claim 7, characterized in that: A plurality of fan-shaped funnels are arranged in a circular array above the integrated slurry storage tank, and at least one weighing assembly is arranged above each fan-shaped funnel; the weighing assembly comprises a first weighing device, and a second weighing device is installed under the first weighing device.

10. The automatic proportioning device for on-site conductive grouting tracer material according to claim 7, characterized in that: The material packaging storage tank is in a circular array, with a screw conveyor installed at the bottom, and a weighing component is placed below the outlet of the screw conveyor.

Images