Real-time dynamic prediction system and method for three-dimensional form of high-pressure jet grouting pile

Abstract

The invention relates to the technical field of real-time prediction of three-dimensional forms, in particular to a real-time dynamic prediction system and method for the three-dimensional form of a high-pressure jet grouting pile. The method comprises the following steps: carrying out stratum characteristic identification on a construction site to obtain stratum distribution data; key monitoring point positions are determined according to the stratum distribution data, and a multi-source sensor monitoring network is established; carrying out anti-interference optimization processing on the multi-source sensor monitoring network, and carrying out real-time acquisition on high-pressure jet grouting parameters to obtain real-time high-pressure jet grouting data; and spatio-temporal evolution processing based on injection pressure, lifting speed and rotating speed is conducted on the real-time high-pressure jet grouting data, and parameter change characteristic data are generated. According to the method, the construction stratum distribution and the high-pressure jet grouting parameters are obtained and analyzed in real time, the parameter-stratum response relation model is established, grout diffusion and pile body forming dynamic prediction are conducted on different stratum characteristics in a targeted mode, and the precise construction capacity under the complex stratum condition is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time prediction of three-dimensional morphology, and in particular to a real-time dynamic prediction system and method for the three-dimensional morphology of high-pressure jet grouting piles. Background Art

[0002] High-pressure jet grouting is a new type of foundation engineering construction technology, which is widely used in civil engineering, foundation treatment and underground engineering. High-pressure jet grouting technology is a foundation treatment method based on high-pressure jet slurry to cut, mix and solidify the soil, mainly used in soft soil foundation reinforcement and groundwater control projects. Its core is to inject slurry into the soil at high pressure by rotating and lifting the nozzle, thereby forming a continuous columnar pile or wall, improving the strength and stability of the soil.

[0003] The traditional real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles often has the following problems: high-pressure jet grouting construction needs to adapt to a variety of strata types, and different strata (such as sand layers, clay layers, etc.) have different responses to the diffusion and penetration of high-pressure jet slurry. Previous pile-making methods usually process different strata based on empirical data, but lack a dynamic prediction and real-time adjustment mechanism, making it difficult to accurately control the pile shape. Previous high-pressure jet grouting construction mainly estimates slurry diffusion and pile forming through experience or limited monitoring methods, but due to the lack of effective prediction and control models, the pile shape is difficult to accurately control, and the pile quality and reinforcement effect are uncertain, increasing the risk of project construction. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a real-time dynamic prediction system and method for the three-dimensional shape of high-pressure rotary jet grouting piles to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above object, a real-time dynamic prediction method for the three-dimensional shape of high-pressure jet grouting piles comprises the following steps:

[0006] Step S1: Identify the stratum characteristics of the construction site and obtain stratum distribution data; determine key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network;

[0007] Step S2: performing anti-interference optimization processing on the multi-source sensor monitoring network, and performing real-time acquisition of high-pressure rotary spraying parameters to obtain real-time high-pressure rotary spraying data; performing spatiotemporal evolution processing on the real-time high-pressure rotary spraying data based on the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data;

[0008] Step S3: establishing a parameter-formation response relationship model based on the parameter change characteristic data and the formation distribution data; performing slurry diffusion characteristic analysis based on the parameter-formation response relationship model, and performing depth-related diffusion radius evaluation to generate slurry diffusion dynamic data;

[0009] Step S4: performing soil improvement correlation processing according to the slurry diffusion dynamic data and the real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; identifying the pile formation influence area based on the soil improvement dynamic characteristic data to generate pile influence range data; coupling the formation stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data to generate soil action coupling data;

[0010] Step S5: construct a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile body to generate morphology prediction data; optimize the construction parameters in real time based on the morphology prediction data to obtain high-pressure rotary jet construction control feedback data.

[0011] The present invention can obtain detailed stratum distribution information of the construction area before construction, including soil layer type, thickness, mechanical properties, etc., by identifying stratum characteristics. By using this information as basic data and determining key monitoring points, the monitoring accuracy and coverage of the sensor network can be effectively improved, and the integrity and accuracy of the data can be ensured. At the same time, the establishment of a multi-source sensor network can monitor the changes in stratum conditions in real time, providing high-precision data support for subsequent predictions. Anti-interference optimization processing effectively reduces the interference of the complex environment of the construction site on the monitoring data through electromagnetic shielding and vibration isolation technology, and improves the reliability of the monitoring data. Real-time collection of parameters (such as injection pressure, lifting speed, rotation speed) of high-pressure rotary spraying construction, and processing of these data with spatiotemporal evolution can capture dynamic changes in the construction process, generate parameter change characteristic data, and provide accurate input features for the prediction model. The parameter-stratum response model constructed by parameter change characteristic data and stratum distribution data can accurately simulate the response relationship of construction parameters to strata. This model not only provides a basis for the analysis of slurry diffusion characteristics, but also helps to identify the diffusion radius of different depths and strata, and realizes accurate prediction of the slurry diffusion range. This helps prevent the waste of resources caused by excessive diffusion while ensuring the depth and uniformity of soil improvement. By combining the dynamic data of slurry diffusion with real-time high-pressure rotary jet data, the effect of different parameter combinations on soil improvement can be analyzed. The dynamic characteristic data of soil improvement generated by this process provides an accurate basis for identifying the area affected by pile formation. Based on the data of the pile influence range, the formation stress mode coupling can clarify the stress changes and transmission of the formation during the construction process, thereby optimizing the forming quality of the pile. The deep learning model trained with the soil action coupling data can predict the three-dimensional morphological changes of the pile in real time, including cross-section, longitudinal changes and spatial distribution. The morphological prediction data generated by the model not only provides intuitive morphological feedback for construction personnel, but also provides a reliable basis for real-time optimization of construction parameters. Through continuous feedback and optimization, the high-pressure rotary jet parameters during the construction process are always in the best state, which improves the construction quality and consistency of pile formation. This method realizes comprehensive monitoring and dynamic prediction of the high-pressure rotary jet pile formation process, forming a closed-loop control system from formation characteristic identification to real-time construction optimization. Through real-time data collection, precise diffusion analysis, soil improvement assessment and deep learning prediction, not only the construction efficiency is improved, but also the quality and stability of piles are significantly improved.

[0012] The present invention also provides a real-time dynamic prediction system for the three-dimensional form of high-pressure rotary grouting piles, which is used to execute the above-mentioned real-time dynamic prediction method for the three-dimensional form of high-pressure rotary grouting piles. The real-time dynamic prediction system for the three-dimensional form of high-pressure rotary grouting piles comprises:

[0013] The site monitoring module is used to identify the stratum characteristics of the construction site and obtain stratum distribution data; determine the key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network;

[0014] The parameter acquisition module is used to perform anti-interference optimization processing on the multi-source sensor monitoring network and to collect high-pressure rotary spray parameters in real time to obtain real-time high-pressure rotary spray data; the real-time high-pressure rotary spray data is processed based on the spatiotemporal evolution of the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data;

[0015] The diffusion prediction module is used to establish a parameter-formation response relationship model based on parameter change characteristic data and formation distribution data; analyze the slurry diffusion characteristics based on the parameter-formation response relationship model, and evaluate the depth-related diffusion radius to generate slurry diffusion dynamic data;

[0016] The impact assessment module is used to perform soil improvement correlation processing based on the slurry diffusion dynamic data and real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; to identify the pile impact area based on the soil improvement dynamic characteristic data to generate pile impact range data; to couple the formation stress mode based on the injection pressure, soil stress and improvement effect based on the pile impact range data to generate soil action coupling data;

[0017] The morphology control module is used to build a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile to generate morphology prediction data; based on the morphology prediction data, the construction parameters are optimized in real time to obtain high-pressure rotary jet construction control feedback data.

[0018] The present invention can obtain accurate stratum distribution information through detailed identification of stratum characteristics, and provide basic data for subsequent construction; determine key monitoring points, and centrally monitor the most important areas to improve monitoring efficiency; establish a multi-source sensor network to ensure the comprehensiveness of monitoring data, improve data reliability and accuracy, and provide support for construction decisions. Through optimization processing, the influence of external interference on sensor data can be effectively reduced, and the reliability of data acquisition can be improved. Real-time acquisition of high-pressure rotary spraying parameters enables each key data in the construction process to be captured in time, providing a basis for dynamically adjusting the construction plan. Based on the spatiotemporal evolution processing of injection pressure, lifting speed and rotation speed, the changing law of construction parameters can be fully understood, laying the foundation for subsequent analysis. By establishing a parameter-stratum response relationship model, the influence of different construction parameters on stratum response can be quantified, and the scientific nature of construction can be improved; by analyzing the diffusion characteristics of slurry and the diffusion radius related to depth, the behavior of slurry in soil can be better understood, and the slurry formula and injection strategy can be optimized; and the slurry diffusion dynamic data is generated, which can provide basic information for subsequent impact assessment. By analyzing the dynamic characteristic data of soil improvement, the influence of slurry on soil can be timely evaluated, providing a basis for construction adjustment. Identifying the impact range data of the pile body can accurately evaluate the impact of the pile construction on the surrounding soil and ensure construction safety. Through the coupling of the stratum stress mode, the injection pressure, soil stress and improvement effect can be comprehensively considered to ensure the scientific and reasonable construction plan. Through the deep learning prediction model, the dynamic prediction of the three-dimensional morphology of the pile body can be realized to improve the prediction accuracy; the construction parameters can be optimized in real time according to the morphological prediction data to ensure flexibility and adaptability during the construction process. Generating high-pressure rotary spraying construction control feedback data can adjust the construction strategy in time, improve construction efficiency and reduce risks. Through the collaborative work of the above modules, the entire system not only improves the automation and intelligence level of high-pressure rotary spraying construction, but also provides comprehensive guarantees for real-time monitoring, data analysis and decision support of the construction process. This integrated solution can effectively improve the safety, efficiency and quality of construction and meet the high requirements of modern engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0020] Figure 1 It is a schematic flow chart of the steps of the real-time dynamic prediction method of the three-dimensional form of high-pressure jet grouting piles of the present invention;

[0021] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0022] Figure 3 for Figure 1Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0023] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0025] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0026] To achieve this, please refer to Figures 1 to 3 The present invention provides a real-time dynamic prediction method for the three-dimensional shape of high-pressure jet grouting piles, the method comprising the following steps:

[0027] Step S1: Identify the stratum characteristics of the construction site and obtain stratum distribution data; determine key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network;

[0028] Step S2: performing anti-interference optimization processing on the multi-source sensor monitoring network, and performing real-time acquisition of high-pressure rotary spraying parameters to obtain real-time high-pressure rotary spraying data; performing spatiotemporal evolution processing on the real-time high-pressure rotary spraying data based on the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data;

[0029] Step S3: establishing a parameter-formation response relationship model based on the parameter change characteristic data and the formation distribution data; performing slurry diffusion characteristic analysis based on the parameter-formation response relationship model, and performing depth-related diffusion radius evaluation to generate slurry diffusion dynamic data;

[0030] Step S4: performing soil improvement correlation processing according to the slurry diffusion dynamic data and the real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; identifying the pile formation influence area based on the soil improvement dynamic characteristic data to generate pile influence range data; coupling the formation stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data to generate soil action coupling data;

[0031] Step S5: construct a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile body to generate morphology prediction data; optimize the construction parameters in real time based on the morphology prediction data to obtain high-pressure rotary jet construction control feedback data.

[0032] In the embodiment of the present invention, reference Figure 1 As shown, it is a schematic diagram of the steps of a real-time dynamic prediction method of the three-dimensional shape of a high-pressure rotary grouting pile according to the present invention. In this example, the real-time dynamic prediction method of the three-dimensional shape of a high-pressure rotary grouting pile comprises the following steps:

[0033] Step S1: Identify the stratum characteristics of the construction site and obtain stratum distribution data; determine key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network;

[0034] In the embodiment of the present invention, in the construction site, the characteristics of the stratum are identified using technologies such as borehole sampling or geological radar to obtain specific stratum distribution data, covering information such as soil layer thickness, soil density, and permeability coefficient. Based on these data, geological information analysis software is used to determine suitable locations for deploying monitoring equipment and set key monitoring points. Then, based on these points, a multi-source monitoring network containing sensors, such as piezoelectric sensors, accelerometers, etc., is arranged to continuously collect key parameters and realize real-time monitoring of the stratum. In the actual scenario of urban subway foundation pit construction, key points can be set at 5 meters, 10 meters and 15 meters from the center of the rotary jet pile to track the stratum response of different radii.

[0035] Step S2: performing anti-interference optimization processing on the multi-source sensor monitoring network, and performing real-time acquisition of high-pressure rotary spraying parameters to obtain real-time high-pressure rotary spraying data; performing spatiotemporal evolution processing on the real-time high-pressure rotary spraying data based on the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data;

[0036] The embodiment of the present invention implements anti-interference optimization for the multi-source sensor monitoring network, and uses filtering technology to reduce the impact of electromagnetic interference of construction equipment on the signal. Subsequently, the embedded acquisition system collects high-pressure rotary spraying parameters, such as injection pressure, lifting speed, rotation speed, etc. in real time to generate real-time rotary spraying data. During the data acquisition process, combined with the rotary spraying equipment used in the actual site, the injection pressure is set at 15MPa, the lifting speed is controlled at 0.2m / min, and the rotation speed is adjusted according to the formation conditions. The spatiotemporal evolution analysis is performed to obtain parameter change characteristic data to ensure that the monitoring data reflects the dynamic characteristics of the formation reaction.

[0037] Step S3: establishing a parameter-formation response relationship model based on the parameter change characteristic data and the formation distribution data; performing slurry diffusion characteristic analysis based on the parameter-formation response relationship model, and performing depth-related diffusion radius evaluation to generate slurry diffusion dynamic data;

[0038] The embodiment of the present invention uses the parameter change characteristic data and formation distribution data of step S2 to construct a parameter-formation response relationship model to analyze the diffusion effect of parameters such as injection pressure and lifting speed in different formations. Based on this model, the diffusion characteristics of the slurry are analyzed using fluid diffusion simulation software to evaluate the diffusion radius of the slurry at different depths. For example, in a sandy formation, the slurry diffusion radius is 0.6 meters, while in a clay layer it diffuses to 0.3 meters, generating corresponding slurry diffusion dynamic data to ensure that the diffusion effect meets the design requirements.

[0039] Step S4: performing soil improvement correlation processing according to the slurry diffusion dynamic data and the real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; identifying the pile formation influence area based on the soil improvement dynamic characteristic data to generate pile influence range data; coupling the formation stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data to generate soil action coupling data;

[0040] The embodiment of the present invention uses slurry diffusion dynamic data and real-time high-pressure rotary jet data to establish a soil improvement correlation model, associates parameters such as diffusion radius, rotary jet speed and jet pressure with soil improvement effects, and generates dynamic characteristic data. Based on this data, the influence area of ​​the pile is further identified, and the influence range of the pile on the surrounding soil is analyzed. Then, based on the pile influence range data, the jet pressure, soil stress and improvement effect are coupled and analyzed to generate soil action coupling data. In the soft soil layer, it can be known through coupling analysis that under the condition of a jet pressure of 12MPa, the soil improvement effect is significant, and regional data with an influence radius of 0.5 meters is generated.

[0041] Step S5: construct a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile body to generate morphology prediction data; optimize the construction parameters in real time based on the morphology prediction data to obtain high-pressure rotary jet construction control feedback data.

[0042] The embodiment of the present invention applies a convolutional neural network to construct a deep learning model based on the coupling data of soil action, and trains the model to predict the three-dimensional morphology of the pile body. The input includes characteristics such as jet grouting parameters and stratum stress. The jet grouting pile data at different time nodes are used as training samples to optimize the model weights and generate morphology prediction data. During on-site construction, the construction parameters are optimized in real time based on the predicted morphology, such as appropriately reducing the injection pressure to 10MPa in clay formations to ensure uniform forming of the pile body. The generated control feedback data is transmitted back to the construction system in real time for dynamic adjustment of the jet grouting construction parameters.

[0043] The present invention can obtain detailed stratum distribution information of the construction area before construction, including soil layer type, thickness, mechanical properties, etc., by identifying stratum characteristics. By using this information as basic data and determining key monitoring points, the monitoring accuracy and coverage of the sensor network can be effectively improved, and the integrity and accuracy of the data can be ensured. At the same time, the establishment of a multi-source sensor network can monitor the changes in stratum conditions in real time, providing high-precision data support for subsequent predictions. Anti-interference optimization processing effectively reduces the interference of the complex environment of the construction site on the monitoring data through electromagnetic shielding and vibration isolation technology, and improves the reliability of the monitoring data. Real-time collection of parameters (such as injection pressure, lifting speed, rotation speed) of high-pressure rotary spraying construction, and processing of these data with spatiotemporal evolution can capture dynamic changes in the construction process, generate parameter change characteristic data, and provide accurate input features for the prediction model. The parameter-stratum response model constructed by parameter change characteristic data and stratum distribution data can accurately simulate the response relationship of construction parameters to strata. This model not only provides a basis for the analysis of slurry diffusion characteristics, but also helps to identify the diffusion radius of different depths and strata, and realizes accurate prediction of the slurry diffusion range. This helps prevent the waste of resources caused by excessive diffusion while ensuring the depth and uniformity of soil improvement. By combining the dynamic data of slurry diffusion with real-time high-pressure rotary jet data, the effect of different parameter combinations on soil improvement can be analyzed. The dynamic characteristic data of soil improvement generated by this process provides an accurate basis for identifying the area affected by pile formation. Based on the data of the pile influence range, the formation stress mode coupling can clarify the stress changes and transmission of the formation during the construction process, thereby optimizing the forming quality of the pile. The deep learning model trained with the soil action coupling data can predict the three-dimensional morphological changes of the pile in real time, including cross-section, longitudinal changes and spatial distribution. The morphological prediction data generated by the model not only provides intuitive morphological feedback for construction personnel, but also provides a reliable basis for real-time optimization of construction parameters. Through continuous feedback and optimization, the high-pressure rotary jet parameters during the construction process are always in the best state, which improves the construction quality and consistency of pile formation. This method realizes comprehensive monitoring and dynamic prediction of the high-pressure rotary jet pile formation process, forming a closed-loop control system from formation characteristic identification to real-time construction optimization. Through real-time data collection, precise diffusion analysis, soil improvement assessment and deep learning prediction, not only the construction efficiency is improved, but also the quality and stability of piles are significantly improved.

[0044] Preferably, step S1 comprises the following steps:

[0045] Step S11: Perform multi-point drilling sampling on the construction site to obtain formation physical sample data;

[0046] Step S12: Identify the type, distribution depth and thickness of the soil layer of the stratum physical sample data, so as to obtain stratum structure data;

[0047] Step S13: performing mechanical property analysis based on density, cohesion, internal friction angle and water content on the physical sample data of the stratum according to the stratum structure data, thereby obtaining mechanical property data of the soil layer;

[0048] Step S14: Performing hydrogeological characteristic analysis based on the soil mechanical property data to obtain hydrogeological characteristic data, wherein the hydrogeological characteristic analysis includes water level observation around the construction site, permeability testing of stratum physical sample data, and groundwater distribution analysis based on water content and permeability;

[0049] Step S15: merging the hydrogeological characteristic data, the soil layer mechanical property data and the stratum structure data into stratum distribution data;

[0050] Step S16: Determine key monitoring points based on the formation distribution data and establish a multi-source sensor monitoring network.

[0051] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0052] Step S11: Perform multi-point drilling sampling on the construction site to obtain formation physical sample data;

[0053] The embodiment of the present invention uses drilling equipment to perform distributed drilling sampling at multiple locations on the construction site to ensure that the samples can represent the characteristics of different depths and stratum distributions, and samples are usually taken at a depth of every 10-15 meters at key locations. When drilling, select a suitable sampler, such as a rotary sampler or a double-tube sampler, to keep the soil in its original state to the greatest extent. After taking out the sample, it is sealed and stored for subsequent laboratory analysis. Each sample is marked with specific depth information and drilling location information when sampling, so that position correspondence can be performed in subsequent analysis.

[0054] Step S12: Identify the type, distribution depth and thickness of the soil layer of the stratum physical sample data, so as to obtain stratum structure data;

[0055] The embodiment of the present invention conducts a detailed analysis of the collected stratum physical sample data in the laboratory, uses a laser particle size analyzer or a sieving method to test the particle size distribution of the soil layer, and uses magnetic or optical recognition technology to identify the mineral components in the soil. For the identification of depth and thickness, by comparing the sample data of different boreholes and combining the similarity of the strata to speculate, the depth change and thickness distribution of different soil layers are confirmed layer by layer, thereby forming complete stratum structure data, which provides a basis for subsequent geological analysis.

[0056] Step S13: performing mechanical property analysis based on density, cohesion, internal friction angle and water content on the physical sample data of the stratum according to the stratum structure data, thereby obtaining mechanical property data of the soil layer;

[0057] The embodiment of the present invention analyzes the mechanical properties of the physical sample based on the stratum structure data. The density is measured using a soil densimeter, and the cohesion and internal friction angle are measured using a uniaxial compression test and a direct shear test. A triaxial shear test is performed according to different moisture content conditions to accurately obtain the mechanical property data of the soil layer under different hydrological conditions. Through multi-point testing and statistical processing, comprehensive mechanical property data is obtained to provide a reference for further stratum stability analysis.

[0058] Step S14: Performing hydrogeological characteristic analysis based on the soil mechanical property data to obtain hydrogeological characteristic data, wherein the hydrogeological characteristic analysis includes water level observation around the construction site, permeability testing of stratum physical sample data, and groundwater distribution analysis based on water content and permeability;

[0059] The embodiment of the present invention conducts a detailed analysis of the hydrogeological characteristics based on the mechanical property data of the soil layer. Water level monitoring is carried out on site, and water level changes are recorded regularly by installing water level observation wells. At the same time, the permeability of the physical samples of the formation is tested in the laboratory, and the permeability coefficient is measured using a permeameter. The groundwater distribution characteristics are analyzed in combination with the soil moisture content and permeability data, and the finite element method is used to simulate the flow of groundwater in different soil layers to obtain accurate hydrogeological characteristic data to provide guidance for construction design.

[0060] Step S15: merging the hydrogeological characteristic data, the soil layer mechanical property data and the stratum structure data into stratum distribution data;

[0061] The embodiment of the present invention integrates hydrogeological characteristic data, soil mechanical property data and stratum structure data. A geographic information system (GIS) platform is used to perform spatial processing on various data to generate a three-dimensional stratum distribution model. The structure, mechanical properties and hydrological characteristics of different strata are displayed in layers on the GIS platform to ensure that the comprehensive situation of each stratum attribute can be intuitively viewed during the analysis. The generated stratum distribution data has a layered visualization function, which is convenient for real-time reference during the construction process.

[0062] Step S16: Determine key monitoring points based on the formation distribution data and establish a multi-source sensor monitoring network.

[0063] The embodiment of the present invention selects the stratum area with greater impact on construction stability as the key monitoring point based on the stratum distribution data, such as the junction of the high permeability soil layer and the clay soil layer. Multi-source sensors, including soil pressure gauges, osmometers and water level monitors, are installed at the key points to form a full coverage sensor monitoring network. To ensure the real-time and reliability of the data, the sensor network transmits the monitoring data to the central control system in real time through wireless communication technology for subsequent construction monitoring and analysis.

[0064] The present invention can directly obtain physical samples of strata at different depths of the construction site through multi-point drilling sampling, ensuring accurate actual measurement data for geological conditions. Drilling sampling forms multiple data points in the area, laying the foundation for subsequent analysis and reducing the uncertainty caused by stratum heterogeneity. The identification of the type, depth and thickness of the soil layer can clarify the distribution of each layer of soil. The stratum structure data obtained in this process is helpful for the formulation and optimization of the construction plan, especially in the key soil layer areas that need to be paid attention to in pile formation, thereby reducing the construction risks caused by misjudgment and unexpected stratum mutations. The mechanical properties of the soil layer directly affect the pressure and slurry diffusion behavior during the pile formation process. Through the analysis of mechanical parameters such as density, cohesion, and internal friction angle, the compressive capacity and stability of each layer of soil can be deeply understood, providing guidance for the optimization of parameters such as pressure, speed, and slurry volume required for pile formation. These data can also help predict the resistance and deformation characteristics that may be encountered during the construction process, reducing construction risks. Hydrogeological characteristics have a direct impact on slurry diffusion and soil layer improvement effects. Water level observation, permeability test and groundwater distribution analysis help to understand the mobility and diffusion of water in the soil layer. By clarifying the hydrological conditions, the construction team can reasonably adjust the slurry pressure and diffusion rate, prevent excessive diffusion or leakage of slurry, improve construction accuracy, and avoid unnecessary construction costs. Integrating hydrogeological characteristics, mechanical properties and stratigraphic structure data into a comprehensive stratigraphic distribution data can provide an overall geological profile and provide a more intuitive reference for construction decisions. This integrated data facilitates the construction team to fully grasp the geological characteristics of the site and improve the efficiency and accuracy of data use. Based on the stratigraphic distribution data, key monitoring points are reasonably arranged and a multi-source sensor network is established, so that the stratigraphic and construction parameters can be monitored in real time during the pile construction process. Through the sensor monitoring network, the construction team can continuously track key indicators such as soil stress, slurry diffusion, and stratigraphic displacement to ensure that various operations during the construction process meet the predetermined safety and quality standards, timely discover and handle abnormal situations, and ensure construction quality and safety. These steps provide detailed geological data and real-time monitoring methods for the construction process from stratigraphic sampling to data integration and then to the establishment of a monitoring network. The integrated data and monitoring network provide a scientific basis for the subsequent adjustment and optimization of high-pressure rotary grouting construction parameters, helping the construction team to achieve efficient and precise control during the construction process, ultimately improving the quality and safety of pile formation and reducing potential risks and costs of engineering construction.

[0065] Preferably, step S16 comprises the following steps:

[0066] Step S161: performing mutation analysis on soil layer interface characteristics according to stratum distribution data to obtain soil layer mutation point data, wherein the mutation analysis specifically identifies locations where soil layer types and mechanical properties change significantly;

[0067] In the mutation analysis of the embodiment of the present invention, the stratum distribution data is first input into the geological modeling software, and the interface where the soil type and mechanical properties of the stratum change significantly is identified by analyzing the soil profile and combining the collected soil physical sample data. Specifically, the location of sudden changes in the soil properties of the stratum is observed by using data such as density, cohesion and internal friction angle. For example, if the density in a certain area changes from 1800kg / m 3 Sudden change to 2100kg / m 3 , then the point can be considered as a soil mutation point. All mutation points are marked in three-dimensional space to form complete soil mutation point data in order to identify the possible impact on engineering stability.

[0068] Step S162: performing sensitivity analysis on the soil layer stress state according to the stratum distribution data, identifying areas where soil stress changes greatly, and obtaining stress sensitive area data, wherein the stress sensitivity analysis includes shear strength evaluation based on cohesion and internal friction angle, compressibility evaluation based on density and water content, and consolidation characteristic evaluation based on permeability;

[0069] The embodiment of the present invention performs a sensitivity analysis of the soil layer stress state on the stratum distribution data to identify areas where the soil stress changes greatly. During the analysis process, the shear strength is first evaluated, and the effects of cohesion and internal friction angle on the shear resistance of different soil layers are calculated. Then, based on the density and moisture content of the soil, its compressibility characteristics are evaluated, and the compression modulus is obtained through a one-dimensional compression test to further analyze the stability of the soil. Finally, based on the permeability test, the consolidation characteristics of different soil layers are evaluated using consolidation theory. Areas with significant changes in shear strength, compressibility or consolidation are marked as stress sensitive areas, and their spatial coordinates and depth information are recorded to form stress sensitive area data, providing basic data for the subsequent layout of monitoring points.

[0070] Step S163: Determine the spatial distribution of key monitoring points according to preset site construction conditions and engineering requirements data, soil layer mutation point data, and stress sensitive area data to obtain monitoring point layout data;

[0071] The embodiment of the present invention arranges key monitoring points according to the construction conditions of the site (such as excavation depth, support type, etc.) and engineering requirements, combined with soil mutation point data and stress sensitive area data. By analyzing the distribution characteristics of mutation points and stress sensitive areas, monitoring points are preferentially arranged in these areas to monitor the stratum status of key areas in real time. For example, stress and displacement sensors are arranged at mutation points found in the area with a foundation excavation depth of 10 meters, or water level and pressure sensors are arranged in highly compressible stress sensitive areas to monitor the stress changes of the soil. Finally, the monitoring point layout data is generated to provide guidance for the next step of sensor type configuration.

[0072] Step S164: Determine the sensor type according to the monitoring point layout data, and establish a multi-source sensor monitoring network including a pressure sensor, a displacement sensor, a strain sensor, and a water level sensor.

[0073] The embodiment of the present invention selects appropriate sensor types based on the monitoring point layout data and establishes a multi-source sensor monitoring network. For points that require displacement monitoring, displacement sensors are selected; pressure sensors are installed at points that need to monitor soil stress changes; water level sensors are configured in areas where groundwater levels change significantly; and strain sensors are configured in stress-sensitive soil areas. The sensors are connected to the central control system via wired or wireless networks to ensure the transmission and recording of real-time data, forming a complete multi-source sensor monitoring network to provide real-time feedback and early warning for subsequent engineering construction.

[0074] The present invention identifies the location where the soil layer type and mechanical properties change significantly through mutation analysis, and can locate the area where the geological characteristics suddenly change (i.e., soil layer mutation point), such as the interface between the hard soil layer and the loose soil layer or the area where the groundwater level changes significantly. These mutation points have an important impact on the construction process. Mutation analysis helps to consider the requirements of different soil layers for pile parameters in advance in the construction design, avoid excessive disturbance of sensitive areas during the construction process, thereby reducing the probability of unexpected situations and ensuring smooth construction. Stress sensitivity analysis can identify areas where soil stress changes greatly (stress sensitive areas) and evaluate the shear strength, compressibility and consolidation characteristics of the soil layer in detail. During the pile forming process, the soil layer in the stress sensitive area is easily affected by force and causes deformation or instability. By identifying these areas, the injection pressure and construction speed can be controlled more specifically. Knowing the stress characteristics of these areas in advance can avoid the instability of the stratum caused by excessive diffusion of slurry in sensitive areas or uneven pressure distribution, and further improve the quality of pile forming and construction safety. Key monitoring points are arranged according to site construction conditions, engineering requirements, and data of soil layer mutation points and stress sensitive areas to ensure that the monitoring network reasonably covers the key positions of the construction area. In this way, the monitoring points can accurately cover important geological structures and stress-sensitive areas, avoiding monitoring blind spots. This layout can monitor the stress, deformation and water level changes of important soil layers in real time during the construction process, so that any abnormalities in the construction process can be discovered and adjusted in time, thereby effectively improving the construction accuracy and safety. Multi-source sensors including pressure sensors, displacement sensors, strain sensors and water level sensors are installed at the determined monitoring points to fully capture the key parameters of the soil layer. This multi-source sensor network can not only monitor the pressure and deformation of the soil layer in real time, but also detect changes in the groundwater level, helping the construction team to obtain soil status information in real time. The establishment of the monitoring network enables the construction team to grasp the mechanical response and slurry diffusion of the soil layer at the first time, which is conducive to dynamically adjusting the construction parameters according to the monitoring data, and further enhancing the intelligence and safety of the construction process. Through the analysis of soil mutation and stress sensitivity, the construction team can deeply understand the stratum characteristics and stress distribution characteristics of the site, and ensure that the monitoring network covers important locations. Reasonably arranged monitoring points and multi-source sensor networks can provide real-time feedback on key parameters during the construction process, helping the construction team to accurately control and dynamically optimize, and significantly improve the quality and safety of pile construction. At the same time, these measures can effectively reduce the construction risks caused by stratum heterogeneity, provide a more scientific monitoring basis for the entire project, ensure smooth construction and reduce costs.

[0075] Preferably, step S2 comprises the following steps:

[0076] Step S21: performing anti-interference optimization processing based on electromagnetic shielding and vibration isolation on the multi-source sensor monitoring network, and collecting real-time parameters during the high-pressure rotary grouting construction process, thereby obtaining real-time high-pressure rotary grouting data;

[0077] Step S22: segmenting the injection pressure in the real-time high-pressure rotary spraying data based on the time series, thereby establishing a pressure-time relationship curve;

[0078] Step S23: Calculate the frequency characteristics and amplitude distribution of pressure fluctuations according to the pressure-time relationship curve, and perform correlation analysis between pressure fluctuations and depth, so as to obtain pressure-depth evolution characteristic data;

[0079] Step S24: establishing a relationship curve between lifting speed and depth based on real-time high-pressure jetting data, and calculating the speed change rate at different depths, thereby obtaining speed spatiotemporal evolution characteristic data;

[0080] Step S25: recording the real-time changes of the rotation speed in the real-time high-pressure rotary spraying data, and performing correlation characteristic analysis between the rotation speed and the depth, thereby obtaining rotation spatiotemporal evolution characteristic data;

[0081] Step S26: performing a comprehensive analysis of the influence of parameter combination on pile formation based on the pressure-depth evolution characteristic data, the velocity spatiotemporal evolution characteristic data, and the rotation spatiotemporal evolution characteristic data, and identifying the critical value and optimal interval of the key parameter combination, thereby generating spatiotemporal evolution characteristic data of the parameter combination;

[0082] Step S27: establishing a dynamic evolution model of the three-dimensional parameter space according to the spatiotemporal evolution characteristic data, and extracting parameter change characteristics, thereby obtaining parameter change characteristic data.

[0083] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0084] Step S21: performing anti-interference optimization processing based on electromagnetic shielding and vibration isolation on the multi-source sensor monitoring network, and collecting real-time parameters during the high-pressure rotary grouting construction process, thereby obtaining real-time high-pressure rotary grouting data;

[0085] When performing anti-interference optimization processing on a multi-source sensor monitoring network, the embodiment of the present invention first installs an electromagnetic shielding device for the sensor and the line to reduce electromagnetic radiation interference generated by surrounding equipment. In addition, a rubber pad or shock absorber is added to the bracket on which the sensor is installed to effectively isolate the vibration interference from the construction equipment. Next, key parameters are collected in real time during the high-pressure rotary grouting construction process, including injection pressure, lifting speed, and rotation speed, to ensure the continuity and accuracy of data collection, so as to provide high-quality real-time high-pressure rotary grouting data for subsequent analysis.

[0086] Step S22: segmenting the injection pressure in the real-time high-pressure rotary spraying data based on the time series, thereby establishing a pressure-time relationship curve;

[0087] The embodiment of the present invention processes the injection pressure in the collected real-time high-pressure rotary spray data in segments according to the time series. Specifically, the pressure data is segmented at fixed intervals of 10 seconds each using the time series analysis method, so as to draw a pressure-time relationship curve. Through these segmented data, the variation law of the injection pressure in different time periods can be accurately displayed, laying a data foundation for the subsequent pressure fluctuation analysis.

[0088] Step S23: Calculate the frequency characteristics and amplitude distribution of pressure fluctuations according to the pressure-time relationship curve, and perform correlation analysis between pressure fluctuations and depth, so as to obtain pressure-depth evolution characteristic data;

[0089] The embodiment of the present invention calculates the frequency characteristics and amplitude distribution of pressure fluctuations based on the pressure-time relationship curve. First, the frequency characteristics of pressure fluctuations are extracted by Fourier transform to obtain the main fluctuation frequency and the secondary frequency, and then the amplitude distribution at each frequency is calculated. In addition, the correlation analysis between pressure fluctuations and depth is performed in combination with data at different depths, and the significance of pressure fluctuations at different depths is calculated by regression analysis method, thereby obtaining pressure-depth evolution characteristic data, which provides basic data for judging the influence of pressure on pile forming at different depths.

[0090] Step S24: establishing a relationship curve between lifting speed and depth based on real-time high-pressure jetting data, and calculating the speed change rate at different depths, thereby obtaining speed spatiotemporal evolution characteristic data;

[0091] When analyzing the lifting speed in real-time high-pressure jet grouting data, the embodiment of the present invention first divides the lifting speed into segments according to the depth, and establishes a relationship curve between the depth and the lifting speed. Specifically, the segmentation interval is set to every 0.5 meters, and the speed change rate of each segment is calculated to form the speed spatiotemporal evolution characteristic data. By observing the curve, the speed change trend of different depth segments can be found. For example, the lower speed at a deeper position may be affected by the density of the soil layer, providing the evolution characteristics of depth and speed, and providing a basis for optimizing the lifting speed control parameters.

[0092] Step S25: recording the real-time changes of the rotation speed in the real-time high-pressure rotary spraying data, and performing correlation characteristic analysis between the rotation speed and the depth, thereby obtaining rotation spatiotemporal evolution characteristic data;

[0093] The embodiment of the present invention records the rotation speed in the real-time high-pressure rotary grouting data, especially the rotation speed changes at different depths. Whenever the depth changes by 0.5 meters, the rotation speed is recorded and its correlation characteristics with the depth are analyzed. For example, at a depth of 10 meters, the rotation speed is stable at 20 rpm, while at a deeper depth, the rotation speed increases to 25 rpm. Through this depth correlation feature analysis, the rotation spatiotemporal evolution feature data is obtained, and control suggestions for the rotation speed are provided to help optimize the construction effects at different depths.

[0094] Step S26: performing a comprehensive analysis of the influence of parameter combination on pile formation based on the pressure-depth evolution characteristic data, the velocity spatiotemporal evolution characteristic data, and the rotation spatiotemporal evolution characteristic data, and identifying the critical value and optimal interval of the key parameter combination, thereby generating spatiotemporal evolution characteristic data of the parameter combination;

[0095] The embodiment of the present invention analyzes the comprehensive influence of different parameter combinations on pile body forming based on pressure-depth evolution characteristic data, velocity spatiotemporal evolution characteristic data, and rotation spatiotemporal evolution characteristic data. The critical value of the key parameter combination is identified by statistical analysis method. For example, the combination of injection pressure of 15MPa, lifting speed of 2m / min, and rotation speed of 20rpm has the best effect on pile body forming. Further, the optimal interval of these parameters is determined, and the spatiotemporal evolution characteristic data of the parameter combination is generated to provide clear parameter control suggestions for subsequent construction.

[0096] Step S27: establishing a dynamic evolution model of the three-dimensional parameter space according to the spatiotemporal evolution characteristic data, and extracting parameter change characteristics, thereby obtaining parameter change characteristic data.

[0097] According to the spatiotemporal evolution characteristic data, the embodiment of the present invention uses three-dimensional modeling software to establish a dynamic evolution model of parameter space, showing the changes of pressure, speed and rotation speed at different depths and times. Next, the key parameter change characteristics are identified by feature extraction technology, such as the peak value or mutation point of parameter change at a specific depth. Through these data, parameter change characteristic data are obtained, which provides an accurate basis for real-time monitoring and parameter adjustment during the construction process to ensure that the pile body molding meets the expected quality.

[0098] The present invention can effectively reduce the interference of the external environment on the sensor data and improve the accuracy and stability of data acquisition through the anti-interference processing of electromagnetic shielding and vibration isolation. The real-time parameters such as the injection pressure, lifting speed and rotation speed in the high-pressure rotary spraying construction process are accurately collected to ensure the reliability of the monitoring data and provide high-quality basic data for subsequent analysis. By segmenting the injection pressure in time series and establishing a pressure-time relationship curve, the pressure change trend can be intuitively observed. This process can identify the stage characteristics of pressure during the construction process, help the construction team to grasp the pressure changes at different time points, and provide basic data support for the further optimization of construction parameters. The frequency characteristics and amplitude distribution of pressure fluctuations are calculated, and the correlation characteristics of pressure fluctuations and depth can be obtained by combining the depth for analysis. This data can help the construction team better understand the pressure response characteristics at different depths, especially when the pressure fluctuates abnormally, it can be adjusted in time to ensure stability under depth changes and effectively prevent the risk of soil layer mutation or instability. By establishing a relationship curve between lifting speed and depth and calculating the speed change rate of different depth sections, it can help the construction team identify the difference in speed in depth, which helps to optimize the construction parameters of different depth sections. The speed change rate analysis can also avoid construction quality problems caused by uneven lifting speed, and help maintain the uniformity and stability of the pile body. By recording the real-time changes in the rotation speed and analyzing it in combination with the depth, we can deeply understand the performance characteristics of the rotation speed at different depths. The analysis of the correlation characteristics of the rotation speed and depth can reveal the impact of depth changes on the rotation effect, which helps to adjust the rotation speed in the key depth section, so as to achieve the best rotary spraying effect and ensure the integrity and uniformity of the pile structure. By comprehensively analyzing the evolution characteristics of pressure, lifting speed and rotation speed in time and space, the critical value and optimal interval of the key parameter combination in the pile formation process can be identified. This process can help the construction team determine the most effective parameter range for pile formation, so as to make real-time parameter adjustments during construction to ensure the quality and structural stability of the pile body. By establishing a dynamic evolution model of the three-dimensional parameter space, the spatiotemporal characteristics of parameter changes can be comprehensively analyzed. The dynamic characteristic extraction of the model helps the construction team understand the changing trend and influencing mechanism of the parameters throughout the construction process, so as to achieve the optimal control of the parameters. The model can provide a scientific basis for real-time construction optimization, realize intelligent construction control, and improve construction efficiency and pile quality. The systematic analysis method collects and analyzes multi-source sensor data to establish a deep correlation and spatiotemporal evolution model of various parameters, so that the construction process of high-pressure jet grouting piles can be dynamically adjusted according to real-time data. This data-based parameter control method not only improves the accuracy and stability of construction, but also greatly reduces the risks caused by uncertain factors during the construction process.At the same time, through comprehensive analysis and optimization of parameters, the construction team can grasp the key influencing factors in the pile-building process, optimize the pile-building quality, and ultimately improve the safety and reliability of the entire project.

[0099] Preferably, step S21 includes the following steps:

[0100] The multi-source sensor monitoring network is optimized for anti-interference based on electromagnetic shielding and vibration isolation, and real-time parameters during the high-pressure rotary spraying construction process are collected to obtain real-time high-pressure rotary spraying data; the electromagnetic shielding includes metal shell packaging of the sensor, shielding layer grounding of the signal cable, and planning of the sensor grounding point; the vibration isolation includes setting shock-absorbing brackets, using flexible connectors, and anti-vibration optimization of the sensor installation position.

[0101] When implementing the anti-interference optimization processing of the multi-source sensor monitoring network, the embodiment of the present invention first installs a metal shell for each sensor to form an electromagnetic shielding layer, and selects an aluminum alloy shell with a thickness of 1mm to resist external electromagnetic interference; at the same time, the signal cable is equipped with a shielding layer, and one end of it is grounded to form a good shielding effect. The specific grounding point is selected at the shortest distance from the sensor to reduce signal attenuation. In order to further improve the anti-interference effect, the grounding point of the sensor is also planned, and its grounding point is separated from other power grounding systems to avoid interference caused by ground return. In addition, in terms of vibration isolation, a dedicated shock-absorbing bracket is installed for each sensor bracket, and a shock-absorbing cushion layer made of rubber material is used to absorb high-frequency vibrations. At the same time, a flexible connector is used to connect the sensor to the fixed bracket to effectively reduce the vibration transmission generated during construction. Finally, based on the vibration characteristics of the construction site, the installation position of the sensor is optimized, and it is kept as far away from high-frequency vibration sources or areas with greater vibration interference as possible, such as away from the main engine of the rotary spraying machine and set in a stable area to ensure the stability and accuracy of data acquisition. After the implementation of the above-mentioned anti-interference optimization measures, real-time parameters of the high-pressure rotary grouting construction process were collected, including injection pressure, lifting speed, rotation speed, etc. These data were recorded in real time for subsequent analysis, thereby obtaining high-precision, low-interference real-time high-pressure rotary grouting data, which provided reliable data support for dynamic monitoring during subsequent construction.

[0102] The present invention performs metal shell packaging on the sensor, which can effectively block external electromagnetic interference (EMI). In the high-pressure rotary spraying construction environment, there are a large number of mechanical equipment and power equipment, which generate strong electromagnetic fields and may have adverse effects on the data acquisition of the sensor. The metal shell can form a Faraday cage effect, isolate the electronic components inside the sensor from the external electromagnetic field, and significantly improve the signal stability of the sensor. The shielding layer of the signal cable is grounded to effectively suppress the external electromagnetic interference conducted along the cable. The grounded shielding layer can provide a reliable grounding path for the cable, transfer the interference signal to the ground, prevent the interference signal from entering the data stream of the sensor, and ensure the accuracy of the sensor output data. Reasonable planning of the grounding point of the sensor to ensure that the grounding path of each sensor is short and stable can minimize the electromagnetic interference caused by the grounding current. At the same time, the reasonable arrangement of the grounding point can also avoid the formation of potential difference between different grounding points, avoid the interference signal caused by this, and ensure the stable operation of the sensor network. The shock-absorbing bracket can absorb and buffer the mechanical vibration from the high-pressure rotary spraying equipment and the construction site. The operation of the equipment during the construction process will produce large vibrations, which will directly act on the sensor to cause data fluctuations and reduce the measurement accuracy. By installing the shock-absorbing bracket, this vibration can be effectively isolated and attenuated, ensuring that the sensor collects data in a stable environment and improving the reliability of the monitoring data. The flexible connector has a certain elasticity and flexibility, which can provide a buffer between the sensor and the mounting bracket, further absorb external vibration, and prevent the vibration from being directly transmitted to the sensor. In this way, the sensor can reduce the vibration impact under the protection of the flexible connector, thereby obtaining more accurate real-time data. By analyzing the distribution of vibration sources at the construction site, the sensor is installed in an area far away from the vibration source and with less vibration. Optimizing the installation position can significantly reduce the vibration impact on the sensor, thereby improving the accuracy of its collected data. Through the above-mentioned electromagnetic shielding and vibration isolation measures, the multi-source sensor monitoring network can significantly improve the anti-interference ability of data collection in the complex environment of high-pressure rotary grouting construction. Electromagnetic shielding and vibration isolation not only ensure the stability of the sensor network, but also reduce the impact of the external environment on the monitoring data, making the collected high-pressure rotary grouting real-time data more accurate and reliable. These optimization processes provide solid data support for subsequent data analysis, parameter control and construction optimization, while reducing construction errors caused by interference, thereby improving construction efficiency and the reliability of engineering quality.

[0103] Preferably, step S27 includes the following steps:

[0104] Step S271: constructing a three-dimensional parameter space with injection pressure, lifting speed and rotation speed as coordinate axes according to the spatiotemporal evolution characteristic data, thereby obtaining parameter space coordinate data;

[0105] The embodiment of the present invention uses the injection pressure, lifting speed and rotation speed data collected in real time as the X, Y and Z axes of the three-dimensional space to construct a parameter space coordinate system. Then, each data point is mapped to the coordinate system in sequence according to the construction time to generate parameter space coordinate data, which is used to show the position relationship of each parameter in the three-dimensional space at different construction times, laying the foundation for analyzing the parameter change trend. This operation needs to consider the specific depth of each data sampling point, and the association and visualization of multi-layer data are realized through depth marking.

[0106] Step S272: establishing a parameter change trajectory based on the parameter space coordinate data, and constructing a parameter motion trajectory curve through spatial mapping of the time series sampling points, thereby obtaining parameter trajectory data;

[0107] The embodiment of the present invention uses the aforementioned parameter space coordinate data to plot the parameter change trajectory of the injection pressure, lifting speed and rotation speed in sequence according to the construction sequence, and generates a parameter motion trajectory curve by connecting adjacent sequence points. The curve can reflect the change trend of each parameter at different depths and times, thereby obtaining parameter trajectory data. The trajectory helps to identify how the parameters change with time and depth during the construction process, and provides a basis for analyzing the stability and mutation of parameter changes during the construction process.

[0108] Step S273: performing depth-based layered processing on the parameter trajectory data to identify parameter change characteristics of different depth segments, including change trends, change rates, and mutual correlations, thereby obtaining layered feature data;

[0109] The embodiment of the present invention performs depth stratification processing based on parameter trajectory data. According to the trajectory data of different depth segments, the parameter change trend, rate and correlation are decomposed layer by layer, the parameter change characteristics of each depth layer are identified, and stratified characteristic data are generated. For example, the parameter change rate and trend in the depth segment of 10-20 meters can reveal the injection effect of this depth segment, thereby providing a basis for adjusting the process parameters at different depths.

[0110] Step S274: constructing a dynamic model of the three-dimensional parameter space according to the hierarchical feature data, establishing a continuous expression of parameter changes through spatial interpolation and numerical fitting, thereby obtaining dynamic evolution model data;

[0111] The embodiment of the present invention combines hierarchical feature data to construct a dynamic model of a three-dimensional parameter space. By performing spatial interpolation and numerical fitting on the data of each depth segment in the model, a continuous expression of parameter changes is generated to obtain dynamic evolution model data. This operation smoothes the discrete sampling data into a continuous surface curve, thereby more intuitively showing the parameter changes in different depth levels, making it easier to observe the parameter evolution process more comprehensively.

[0112] Step S275: extracting the gradient features, curvature features and speed features of the dynamic evolution model data based on parameter changes, thereby obtaining model feature data;

[0113] The embodiment of the present invention performs detailed feature extraction on the dynamic evolution model data. Specifically, the gradient calculation is used to extract the rate of change of the parameter, the curvature calculation reflects the shape change of the parameter trajectory, and the velocity feature analysis is used to identify the amplitude and direction of parameter fluctuations to generate model feature data. For example, the gradient change of the pressure parameter can reveal the stability of the injection pressure and provide suggestions for process adjustments.

[0114] Step S276: performing parameter correlation analysis based on the coupling relationship and mutual feedback mechanism between parameters based on the model characteristic data, thereby obtaining parameter correlation data;

[0115] The embodiment of the present invention carries out coupling relationship analysis between parameters based on model characteristic data, further identifies the correlation between injection pressure, lifting speed and rotation speed through the mutual feedback mechanism between parameters, and generates parameter correlation data. This step analyzes the correlation and mutual influence of the three parameters through numerical simulation and regression analysis methods, revealing the coupling behavior between parameters during the construction process.

[0116] Step S277: extracting indicators characterizing the dynamic change rules of parameters based on the model characteristic data and the parameter association data, thereby generating parameter change characteristic data.

[0117] The embodiment of the present invention combines the model characteristic data and the parameter association data to extract key indicators that can characterize the dynamic change law of the parameters, such as the average change rate, the maximum change amplitude, the correlation coefficient between parameters, etc., and generates parameter change characteristic data. These indicators are used for further construction parameter optimization. For example, when the injection pressure is low and changes smoothly, the lifting speed can be appropriately increased to improve efficiency, thereby completing the feedback of parameter optimization and construction control.

[0118] The present invention constructs a three-dimensional parameter space with injection pressure, lifting speed and rotation speed as coordinate axes, so that the dynamic changes of each parameter in the same space are clear at a glance. In this way, the mutual relationship between the three key parameters can be fully observed, which is convenient for more accurate analysis of their comprehensive influence on pile formation, and lays the foundation for subsequent trajectory analysis. The parameter change trajectory can intuitively show the evolution path of the parameters in different construction stages. By constructing the parameter motion trajectory curve, the dynamic change trend of each parameter in the time series can be observed, which is helpful to identify the key time points and the rate of parameter change, effectively reveal the mutual influence and coupling relationship between the parameters, and enhance the understanding of the dynamic behavior of the parameters. By deeply layering the parameter trajectory data, the parameter change characteristics of different depth segments can be more clearly identified, including the change trend, rate and mutual correlation. The layered feature data not only helps to deeply understand the law of parameter change with depth, but also provides a reference for parameter adjustment and optimization under different formation conditions. After the dynamic model is established, the parameter change trend can be presented in the form of continuous expression, which is convenient for more intuitive observation of its evolution process. Through spatial interpolation and numerical fitting, the dynamic model data has high-precision temporal and spatial continuity, which is helpful for real-time optimization of construction parameters and improves the accuracy of prediction and control. By extracting the gradient, curvature and velocity characteristics of the parameters, we can have a deeper understanding of the rate, direction and amplitude of parameter changes. This feature extraction provides more representative quantitative indicators for analyzing the dynamic change pattern of parameters, which helps to determine the sensitive areas and key intervals of parameter changes during the construction process, thereby guiding the fine adjustment of construction parameters. Based on the coupling relationship and mutual feedback mechanism between parameters, the correlation of parameters can be analyzed to reveal the intrinsic connection between different parameters and identify potential interactions. Parameter correlation data helps to judge the degree of influence of parameter changes on other parameters, enhance the grasp of the comprehensive effects of multiple parameters during the construction process, and thus achieve overall optimization of the construction effect. By extracting the characterization indicators of the dynamic change law, a scientific basis is provided for parameter control in high-pressure rotary grouting construction. This step not only helps to grasp the law of parameter changes, but can also be further used for key indicator monitoring of construction control, realize real-time feedback and adjustment, and ensure the accuracy and stability of the construction process. This series of steps not only established a three-dimensional parameter space, dynamic model and hierarchical characteristic data, providing a multi-dimensional and refined characterization method for parameter analysis of high-pressure rotary grouting construction, but also revealed the dynamic characteristics of the construction process by extracting the coupling relationship and change law between parameters. The parameter change characteristic data finally generated can guide the real-time optimization of construction parameters and provide key data support for improving pile molding quality and construction efficiency.

[0119] Preferably, step S3 comprises the following steps:

[0120] Step S31: establishing a parameter-stratum characteristic correspondence table including a mapping relationship between parameter changes and soil responses in different strata according to the parameter change characteristic data and the stratum distribution data, thereby obtaining parameter response data;

[0121] The embodiment of the present invention constructs a parameter-stratum characteristic correspondence table based on parameter change characteristic data and stratum distribution data. The specific operation is first to associate the stratum information of different depth sections with the corresponding parameter change data, such as injection pressure, lifting speed and rotation speed, etc., which come from the real-time monitoring results of the previous steps. Then, these parameter change data are combined with the corresponding stratum type, thickness and mechanical properties (such as cohesion and internal friction angle) to form a mapping relationship table, and finally generate parameter response data. This table can reflect the influence of construction parameters on soil response under different stratum conditions, providing a basis for subsequent analysis.

[0122] Step S32: performing data standardization and feature dimension reduction processing on the parameter response data, and establishing a mathematical model of parameter-formation response, thereby obtaining response model data;

[0123] The embodiment of the present invention performs standardization processing on the parameter response data so that parameters of different magnitudes and units can be compared within the same range. Common methods include Z-score standardization or Min-Max scaling. Subsequently, feature dimensionality reduction techniques such as principal component analysis (PCA) or t-SNE are applied to convert high-dimensional data into low-dimensional data, extract features that have a significant impact on the response, and then establish a mathematical model of parameter-formation response to obtain response model data. This model will make the response trend of different parameters when the formation characteristics change clearer, laying the foundation for the effectiveness of the model.

[0124] Step S33: Perform machine learning training based on the response model data to construct a nonlinear mapping relationship between the parameter and the formation response, thereby obtaining a parameter-formation response relationship model;

[0125] The embodiment of the present invention is based on the response model data and uses machine learning algorithms such as support vector machines (SVM), random forests or neural networks for training to construct a nonlinear mapping relationship between parameters and formation responses. The training process requires that the training data be divided into a training set and a validation set, and the model parameters are adjusted through an iterative optimization algorithm until the model can accurately predict the response of the formation to different construction parameters. This will make the obtained parameter-formation response relationship model have good generalization ability and can be applied to unseen data scenarios.

[0126] Step S34: performing slurry diffusion characteristic analysis including pressure propagation, soil penetration and slurry solidification according to the parameter-stratum response relationship model, thereby obtaining diffusion characteristic data;

[0127] The embodiment of the present invention conducts slurry diffusion characteristic analysis based on the parameter-stratum response relationship model, specifically including comprehensive analysis of pressure propagation, soil permeability and slurry solidification process. First, the model is used to predict the slurry pressure changes under different stratum conditions and its diffusion process in the soil; secondly, the permeability characteristics of the soil are analyzed to affect the solidification effect of the slurry at different depths; finally, through numerical simulation or experimental data, the diffusion characteristic data of the slurry under different conditions are obtained to ensure the comprehensiveness and accuracy of the analysis.

[0128] Step S35: Calculating the diffusion range related to the depth based on the diffusion characteristic data to obtain diffusion radius data, wherein the diffusion range calculation includes the influence analysis of the slurry diffusion pressure attenuation law and the soil shear strength;

[0129] The embodiment of the present invention uses diffusion characteristic data to calculate the depth-dependent diffusion range. The specific operation includes studying the pressure attenuation law during the slurry diffusion process, and using nonlinear regression or finite element analysis methods to simulate the diffusion behavior of the slurry in the soil. At the same time, it is also necessary to consider the influence of the shear strength of the soil on the diffusion range, establish a calculation model for the diffusion radius, and identify the effective influence range of the slurry at different depths, and finally obtain the diffusion radius data to optimize the construction parameters.

[0130] Step S36: identifying key features and mutation points in the slurry diffusion process on the diffusion dynamic data, thereby obtaining the slurry diffusion dynamic data.

[0131] The embodiment of the present invention identifies key features and mutation points of diffusion dynamic data. This process first requires setting thresholds for monitoring parameters such as pressure and permeability, and identifying mutation points in the data through time series analysis methods (such as sliding average or autoregressive moving average model). These mutation points often indicate major changes or abnormal phenomena in the slurry diffusion process. The identified key features and mutation points will provide important information for the slurry diffusion dynamic data, help understand the performance and potential risks of the slurry during the construction process, and then optimize subsequent construction decisions.

[0132] The present invention combines parameter change characteristic data with stratum distribution data to create a parameter-stratum characteristic correspondence table, which helps to identify the impact of parameter changes in different strata on soil response. This mapping relationship provides comprehensive data support for subsequent analysis, can accurately capture the response characteristics of parameters under different stratum conditions, and improve the adaptability to stratum conditions. Through standardization, the dimensional differences in parameter response data are eliminated to ensure data consistency; through feature dimension reduction, the data structure is simplified and the model complexity is reduced, thereby improving the accuracy and efficiency of subsequent modeling. This step provides a concise and reliable data basis for constructing a parameter-stratum response mathematical model, which helps to improve the stability and computational efficiency of the response model. A nonlinear mapping model is constructed through machine learning training to capture the complex nonlinear relationship between parameters and stratum response, and to achieve accurate prediction of parameter changes and soil response. The model can adjust parameters such as injection pressure and speed according to different stratum conditions, effectively improve the adaptability and accuracy of slurry diffusion control, and ensure the uniformity and stability of slurry diffusion during pile formation. Based on the parameter-stratum response relationship model, the diffusion characteristics of pressure propagation, soil penetration and slurry solidification are analyzed to fully understand the diffusion law of slurry in different strata. This step ensures the uniform diffusion and effective solidification of the slurry in the soil, prevents insufficient or excessive diffusion, and avoids unnecessary impact on the surrounding soil, thereby improving the forming quality and structural stability of the pile. By analyzing the attenuation law of the slurry diffusion pressure and the shear strength of the soil, the diffusion range and diffusion radius data are accurately calculated, which is convenient for controlling the effective range of slurry diffusion. This calculation provides a reference for pile design, ensuring that the slurry can effectively fill the soil voids during pile construction, improve the bonding strength between the pile and the surrounding soil layer, and thus enhance the bearing capacity and durability of the pile. Through dynamic data analysis, the key features and mutation points in the slurry diffusion process can be identified, and abnormal conditions in the diffusion process can be discovered in time and necessary adjustments can be made. This identification process effectively prevents the instability of the slurry during the diffusion process, improves the uniformity and control accuracy of the slurry diffusion, and ensures the overall quality of the pile formation. These steps jointly construct a systematic method from the parameter-stratum response relationship model to the slurry diffusion characteristic analysis, which not only improves the prediction accuracy of the model, but also ensures the efficient diffusion of the slurry during pile construction. This process provides scientific guidance for pile construction. Through real-time monitoring and adjustment of parameters, it ensures the quality of the pile body and the controllability of construction, optimizes construction efficiency and pile-forming effects, and provides a solid technical foundation for subsequent construction management and improvement of pile body quality.

[0133] Preferably, step S4 comprises the following steps:

[0134] Step S41: analyzing the influence of different parameter combinations on soil improvement effects based on slurry diffusion dynamic data and real-time high-pressure rotary jetting data, thereby obtaining improvement impact data;

[0135] The embodiment of the present invention analyzes the influence of different parameter combinations on the soil improvement effect based on the slurry diffusion dynamic data and real-time high-pressure rotary jet data. The specific operation includes first collecting key parameters related to slurry diffusion and high-pressure rotary jet, such as injection pressure, slurry solidification time and physical properties of soil, and then using multivariate regression analysis or machine learning methods to establish a relationship model between parameter combinations and soil improvement effects. Through experimental design of different parameter combinations (such as orthogonal experiments or response surface methodology), the influence of each parameter on the soil improvement effect is systematically evaluated, thereby obtaining improvement impact data to help guide subsequent construction optimization.

[0136] Step S42: evaluating the performance of the improved body based on the strength development law, curing time characteristics and deformation characteristics of the improved impact data, thereby obtaining performance evaluation data;

[0137] The embodiment of the present invention evaluates the performance of the improved body based on the strength development law, curing time characteristics and deformation characteristics of the improvement impact data. The specific operation includes using standard test methods (such as unconfined compressive strength test, permeability test, etc.) to evaluate the performance of the improved soil body, combining the improvement impact data, and using statistical analysis (such as variance analysis) to study the strength development trend of the improved soil body, the influence of curing time on the performance, and the deformation characteristics of the improved body, so as to obtain performance evaluation data. These data will provide engineers with information on the effectiveness and applicability of soil improvement.

[0138] Step S43: analyzing the evolution characteristics of the improvement effect over time on the performance evaluation data, thereby obtaining soil improvement dynamic characteristic data;

[0139] The embodiment of the present invention analyzes the evolution characteristics of the improvement effect over time on the performance evaluation data. This process requires collecting performance evaluation data at different time points, applying time series analysis methods (such as autoregressive integrated moving average model, ARIMA) to analyze the performance change trend over time, and identify the key time nodes that affect the soil improvement effect. By analyzing these dynamic characteristic data, the changes in the improvement effect at different stages after construction can be revealed, and finally the dynamic characteristic data of soil improvement can be obtained.

[0140] Step S44: numerically simulate the pile influence area according to the soil improvement dynamic characteristic data, and establish a multi-field coupling analysis model including stress field, displacement field and seepage field, so as to obtain the influence area data;

[0141] The embodiment of the present invention performs numerical simulation of the area affected by pile formation based on the dynamic characteristic data of soil improvement. The specific method includes using finite element analysis (FEA) software to establish a three-dimensional numerical model to simulate the influence of the load applied by the pile on the stress field, displacement field and seepage field of the surrounding soil. In the model, it is necessary to set the physical properties of the soil (such as elastic modulus, Poisson's ratio, etc.) and boundary conditions, run the simulation to obtain the impact area data, analyze the impact of pile construction on the surrounding environment, and provide a basis for subsequent analysis.

[0142] Step S45: performing boundary recognition and spatial partitioning on the impact area data, thereby obtaining pile body impact range data;

[0143] The embodiment of the present invention performs boundary recognition and spatial partitioning on the impact area data. The specific operation includes using the numerical simulation results and applying image processing technology (such as edge detection algorithm) to identify the boundaries of stress and displacement fields. At the same time, the impact area is divided into different spatial partitions (such as according to stress size, displacement change, etc.) for easy analysis and visualization. This process will obtain the pile body impact range data and clarify the specific impact area of ​​the pile body on the surrounding soil.

[0144] Step S46: coupling the stratum stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data, identifying the stress field evolution law and critical state characteristics, and generating soil action coupling data.

[0145] The embodiment of the present invention couples the formation stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data. The specific operation includes establishing a multi-physics field coupling model that comprehensively considers various influencing factors, and simulating the stress field evolution law of the soil during the improvement process by introducing the changes in injection pressure and soil stress. Combining experimental data and numerical simulation results, key critical state characteristics are identified, and finally soil action coupling data is generated to provide a scientific basis for further soil improvement and construction optimization.

[0146] The present invention can identify which parameter combinations have the most significant soil improvement effect by analyzing different parameter combinations of slurry diffusion dynamic data and real-time high-pressure rotary spraying data. This analysis provides a scientific basis for optimizing construction parameters, which helps to dynamically adjust parameters during construction, thereby improving the efficiency and effect of soil improvement. The improvement impact data is evaluated based on the strength development law, curing time characteristics and deformation characteristics, so as to fully understand the performance of the improved body. This evaluation can help engineers predict the mechanical properties of the improved soil during construction, and provide data support for subsequent construction plan adjustments, thereby ensuring the effectiveness and durability of soil improvement. By analyzing the evolution characteristics of performance evaluation data over time, it is possible to identify the change law of soil improvement effect over time and provide dynamic monitoring of the improvement effect. This dynamic characteristic data is crucial for maintenance and evaluation in the later stage of construction, and helps to ensure the stability and safety of the soil in long-term use. By establishing a multi-field coupling analysis model of stress field, displacement field and seepage field, the influence area of ​​the pile body after soil improvement can be accurately simulated. This simulation provides important theoretical support for construction, ensuring that the interaction of various environmental factors in pile construction is fully considered, thereby optimizing the design and construction strategy of the pile body. Boundary identification and spatial partitioning of the impact area data can clarify the impact range of the pile. This step helps to accurately delineate the construction area, ensure that the construction activities will not have a negative impact on the surrounding environment, and provide basic data for subsequent monitoring and evaluation. Coupling analysis of the stratum stress pattern based on the pile impact range data can identify the evolution law and critical state characteristics of the stress field. This analysis helps to reveal the interaction between the pile and the surrounding soil, ensuring the safety and stability of the pile during construction and long-term use. By understanding the impact of different construction parameters on stress distribution, the pile design can be optimized to ensure that the pile can adapt to actual construction conditions and environmental changes. These steps together build a systematic analysis framework to help engineers fully understand the impact of slurry diffusion on soil improvement, and to perform scientific and effective construction parameter optimization and evaluation under dynamically changing construction conditions. Through these analyses, the effectiveness and stability of soil improvement can be ensured, the bearing capacity and service life of the pile can be improved, thereby effectively reducing construction risks and ensuring the successful implementation of the project.

[0147] Preferably, step S5 comprises the following steps:

[0148] Step S51: extracting features from soil action coupling data and constructing a training sample set for a deep learning model to obtain training sample data, wherein the training sample data includes input features and morphological labels;

[0149] The embodiment of the present invention performs feature extraction on the soil action coupling data and constructs a training sample set for the deep learning model. The specific operation includes first cleaning and preprocessing the soil action coupling data, extracting key features such as stress distribution, displacement change, slurry diffusion radius, etc., and then organizing these features into an input feature set. At the same time, the corresponding soil morphology is marked and a morphological label is generated. For example, image segmentation technology is used to mark different soil morphologies (such as diffusion morphology, solidification morphology, etc.). In this way, training sample data containing input features and morphological labels are constructed to provide a basis for the subsequent training of deep learning models.

[0150] Step S52: constructing a deep learning network structure based on a hybrid framework of a convolutional neural network and a recurrent neural network according to the training sample data, and performing model training to obtain a morphology prediction model;

[0151] The embodiment of the present invention constructs a deep learning network structure based on a hybrid architecture of a convolutional neural network (CNN) and a recurrent neural network (RNN) according to training sample data. The specific method is to design a multi-layer convolutional neural network to extract spatial features, and combine a long short-term memory network (LSTM) to process temporal information to form an effective hybrid architecture. Use a back propagation algorithm and an optimization algorithm (such as Adam or SGD) to train the model, iteratively adjust the network parameters, so that the model can effectively capture the relationship between the input features and the morphological labels, and finally obtain a morphological prediction model with predictive ability.

[0152] Step S53: optimizing the model prediction accuracy of the morphology prediction model based on a cross-validation method, thereby obtaining prediction model data;

[0153] The embodiment of the present invention optimizes the model prediction accuracy of the morphological prediction model based on the cross-validation method. The specific operation is to divide the training sample data into multiple subsets, use the K-fold cross-validation method to train and verify the model, calculate the prediction accuracy (such as accuracy, F1 value, etc.) of the model on different subsets through multiple training and evaluation, and then adjust the model's hyperparameters (such as learning rate, batch size, etc.) to improve the generalization ability of the model, and finally obtain the optimized prediction model data to ensure the stability and accuracy of the model on new data.

[0154] Step S54: dynamically predicting the three-dimensional shape of the pile body using the prediction model data, thereby obtaining shape prediction data, wherein the shape prediction data includes cross-sectional shape, vertical change, and spatial distribution;

[0155] The embodiment of the present invention uses the prediction model data to dynamically predict the three-dimensional shape of the pile body. The specific operation includes inputting the high-pressure rotary grouting data collected in real time into the optimized shape prediction model to generate the three-dimensional shape prediction results of the pile body during the construction process. This process can use software visualization tools to present the prediction results as a three-dimensional model, showing the cross-sectional shape, vertical changes and spatial distribution of the pile body, helping construction personnel to monitor and adjust the construction strategy in real time to ensure construction quality.

[0156] Step S55: The key stages and morphological mutation characteristics of the pile body formation process are analyzed for the morphological prediction data, thereby obtaining morphological evolution data, wherein the key stages include the initial injection stage, the diffusion development stage, the initial solidification stage, the strength development stage, the stable forming stage, and the adjacent pile body influence stage;

[0157] The embodiment of the present invention analyzes the key stages and morphological mutation characteristics of the pile body formation process on the morphological prediction data. This process first identifies the key stages in the morphological prediction data, including the initial injection stage, the diffusion development stage, the initial solidification stage, the strength development stage, the stable forming stage and the adjacent pile body influence stage. By analyzing the pile body morphological changes at different stages, data mining techniques (such as cluster analysis and feature extraction) are used to identify morphological mutation points and obtain morphological evolution data. These data will provide an important basis for understanding the dynamic changes in the pile body formation process.

[0158] Step S56: Optimize the construction parameters of each stage in real time according to the morphological evolution data, so as to obtain high-pressure jet grouting construction control feedback data.

[0159] The embodiment of the present invention optimizes the construction parameters of each stage in real time according to the morphological evolution data. The specific operation includes establishing a relationship model between the construction parameters and the pile body morphological evolution, analyzing the influence of the construction parameters (such as injection pressure, slurry curing time, lifting speed, etc.) at different stages on the pile body morphology, monitoring the key parameters in the construction process in real time, and adjusting the construction parameters using an optimization algorithm (such as a genetic algorithm or a particle swarm optimization), achieving real-time feedback and optimization, and finally obtaining the high-pressure rotary spraying construction control feedback data, providing a scientific basis and effective guidance for the construction process.

[0160] The present invention can identify important factors affecting the pile morphology by extracting features from soil action coupling data. The training sample set constructed in this step includes input features (such as soil parameters, construction conditions, etc.) and morphological labels (pile morphological features), which provide rich data support for subsequent deep learning models and ensure the effectiveness and accuracy of model training. Using a hybrid architecture of convolutional neural networks (CNN) and recurrent neural networks (RNN), the complex relationship between spatial features and time series features can be effectively captured. This deep learning network structure can improve the model's ability to predict changes in pile morphology, enhance the model's adaptability, and thus improve the accuracy of prediction. By optimizing the morphological prediction model through a cross-validation method, overfitting can be effectively avoided, ensuring the generalization ability of the model on unseen data. This optimization process helps to improve the reliability of the model and provide more accurate morphological prediction for actual construction. Using a trained prediction model to dynamically predict the three-dimensional morphology of the pile body, comprehensive information on cross-sectional shape, vertical changes, and spatial distribution can be obtained. These prediction data provide a scientific basis for monitoring and adjustment during the construction process, which helps to grasp the formation of the pile body in real time and ensure construction quality. Identifying the key stages and morphological mutation characteristics in the pile formation process can provide a deep understanding of the growth process and influencing factors of the pile. This analysis helps engineers grasp the key links in the construction process and take timely measures to deal with potential problems, thereby improving the safety and effectiveness of construction. The construction parameters of each stage are optimized in real time according to the morphological evolution data, and the construction strategy of high-pressure rotary grouting can be dynamically adjusted. This real-time feedback mechanism ensures flexible response during the construction process, maximizes construction efficiency, reduces construction risks, and ensures the final quality and performance of the pile. Through this series of steps, the three-dimensional morphology of the pile can be accurately predicted and dynamically optimized, effectively improving the intelligence level of high-pressure rotary grouting construction and the overall efficiency of the project. Using deep learning technology to analyze the complex dynamics in the soil improvement process can better deal with the uncertainty in construction and ensure the safety and reliability of the pile structure. This method not only improves the accuracy of construction, but also provides strong data support for subsequent monitoring and evaluation, and provides a scientific decision-making basis for modern engineering construction management.

[0161] The present invention also provides a real-time dynamic prediction system for the three-dimensional form of high-pressure rotary grouting piles, which is used to execute the above-mentioned real-time dynamic prediction method for the three-dimensional form of high-pressure rotary grouting piles. The real-time dynamic prediction system for the three-dimensional form of high-pressure rotary grouting piles comprises:

[0162] The site monitoring module is used to identify the stratum characteristics of the construction site and obtain stratum distribution data; determine the key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network;

[0163] The parameter acquisition module is used to perform anti-interference optimization processing on the multi-source sensor monitoring network and to collect high-pressure rotary spray parameters in real time to obtain real-time high-pressure rotary spray data; the real-time high-pressure rotary spray data is processed based on the spatiotemporal evolution of the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data;

[0164] The diffusion prediction module is used to establish a parameter-formation response relationship model based on parameter change characteristic data and formation distribution data; analyze the slurry diffusion characteristics based on the parameter-formation response relationship model, and evaluate the depth-related diffusion radius to generate slurry diffusion dynamic data;

[0165] The impact assessment module is used to perform soil improvement correlation processing based on the slurry diffusion dynamic data and real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; to identify the pile impact area based on the soil improvement dynamic characteristic data to generate pile impact range data; to couple the formation stress mode based on the injection pressure, soil stress and improvement effect based on the pile impact range data to generate soil action coupling data;

[0166] The morphology control module is used to build a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile to generate morphology prediction data; based on the morphology prediction data, the construction parameters are optimized in real time to obtain high-pressure rotary jet construction control feedback data.

[0167] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0168] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A real-time dynamic prediction method for the three-dimensional shape of high-pressure jet grouting piles, characterized in that: The following steps are involved: Step S1: Identify the stratum characteristics of the construction site and obtain stratum distribution data; determine key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network; Step S2: performing anti-interference optimization processing on the multi-source sensor monitoring network, and performing real-time collection of high-pressure rotary spraying parameters to obtain real-time high-pressure rotary spraying data; The real-time high-pressure jet spraying data is processed based on the spatiotemporal evolution of the jet pressure, lifting speed and rotation speed to generate parameter change characteristic data; Step S3: establishing a parameter-formation response relationship model based on the parameter change characteristic data and the formation distribution data; performing slurry diffusion characteristic analysis based on the parameter-formation response relationship model, and performing depth-related diffusion radius evaluation to generate slurry diffusion dynamic data; Step S4: performing soil improvement correlation processing according to the slurry diffusion dynamic data and the real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; identifying the pile formation influence area based on the soil improvement dynamic characteristic data to generate pile influence range data; coupling the formation stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data to generate soil action coupling data; Step S5: construct a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile body to generate morphology prediction data; optimize the construction parameters in real time based on the morphology prediction data to obtain high-pressure rotary jet construction control feedback data.

2. The real-time dynamic prediction method of the three-dimensional morphology of high-pressure jet grouting piles according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Perform multi-point drilling sampling on the construction site to obtain formation physical sample data; Step S12: Identify the type, distribution depth and thickness of the soil layer of the stratum physical sample data, so as to obtain stratum structure data; Step S13: performing mechanical property analysis based on density, cohesion, internal friction angle and water content on the physical sample data of the stratum according to the stratum structure data, thereby obtaining mechanical property data of the soil layer; Step S14: Performing hydrogeological characteristic analysis based on the soil mechanical property data to obtain hydrogeological characteristic data, wherein the hydrogeological characteristic analysis includes water level observation around the construction site, permeability testing of stratum physical sample data, and groundwater distribution analysis based on water content and permeability; Step S15: merging the hydrogeological characteristic data, the soil layer mechanical property data and the stratum structure data into stratum distribution data; Step S16: Determine key monitoring points based on the formation distribution data and establish a multi-source sensor monitoring network.

3. The real-time dynamic prediction method of the three-dimensional morphology of high-pressure jet grouting piles according to claim 2 is characterized in that: Step S16 includes the following steps: Step S161: performing mutation analysis on soil layer interface characteristics according to stratum distribution data to obtain soil layer mutation point data, wherein the mutation analysis specifically identifies locations where soil layer types and mechanical properties change significantly; Step S162: performing sensitivity analysis on the soil layer stress state according to the stratum distribution data, identifying areas where soil stress changes greatly, and obtaining stress sensitive area data, wherein the stress sensitivity analysis includes shear strength evaluation based on cohesion and internal friction angle, compressibility evaluation based on density and water content, and consolidation characteristic evaluation based on permeability; Step S163: Determine the spatial distribution of key monitoring points according to preset site construction conditions and engineering requirements data, soil layer mutation point data, and stress sensitive area data to obtain monitoring point layout data; Step S164: Determine the sensor type according to the monitoring point layout data, and establish a multi-source sensor monitoring network including a pressure sensor, a displacement sensor, a strain sensor, and a water level sensor.

4. The real-time dynamic prediction method of the three-dimensional morphology of high-pressure jet grouting piles according to claim 3 is characterized in that: Step S2 includes the following steps: Step S21: performing anti-interference optimization processing based on electromagnetic shielding and vibration isolation on the multi-source sensor monitoring network, and collecting real-time parameters during the high-pressure rotary grouting construction process, thereby obtaining real-time high-pressure rotary grouting data; Step S22: segmenting the injection pressure in the real-time high-pressure rotary spraying data based on the time series, thereby establishing a pressure-time relationship curve; Step S23: Calculate the frequency characteristics and amplitude distribution of pressure fluctuations according to the pressure-time relationship curve, and perform correlation analysis between pressure fluctuations and depth, so as to obtain pressure-depth evolution characteristic data; Step S24: establishing a relationship curve between lifting speed and depth based on real-time high-pressure jetting data, and calculating the speed change rate at different depths, thereby obtaining speed spatiotemporal evolution characteristic data; Step S25: recording the real-time changes of the rotation speed in the real-time high-pressure rotary spraying data, and performing correlation characteristic analysis between the rotation speed and the depth, thereby obtaining rotation spatiotemporal evolution characteristic data; Step S26: performing a comprehensive analysis of the influence of parameter combination on pile formation based on the pressure-depth evolution characteristic data, the velocity spatiotemporal evolution characteristic data, and the rotation spatiotemporal evolution characteristic data, and identifying the critical value and optimal interval of the key parameter combination, thereby generating spatiotemporal evolution characteristic data of the parameter combination; Step S27: establishing a dynamic evolution model of the three-dimensional parameter space according to the spatiotemporal evolution characteristic data, and extracting parameter change characteristics, thereby obtaining parameter change characteristic data.

5. The real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles according to claim 4 is characterized in that: Step S21 includes the following steps: The multi-source sensor monitoring network is optimized for anti-interference based on electromagnetic shielding and vibration isolation, and real-time parameters during the high-pressure rotary spraying construction process are collected to obtain real-time high-pressure rotary spraying data; the electromagnetic shielding includes metal shell packaging of the sensor, shielding layer grounding of the signal cable, and planning of the sensor grounding point; the vibration isolation includes setting shock-absorbing brackets, using flexible connectors, and anti-vibration optimization of the sensor installation position.

6. The real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles according to claim 5 is characterized in that: Step S27 includes the following steps: Step S271: constructing a three-dimensional parameter space with injection pressure, lifting speed and rotation speed as coordinate axes according to the spatiotemporal evolution characteristic data, thereby obtaining parameter space coordinate data; Step S272: establishing a parameter change trajectory based on the parameter space coordinate data, and constructing a parameter motion trajectory curve through spatial mapping of the time series sampling points, thereby obtaining parameter trajectory data; Step S273: performing depth-based layered processing on the parameter trajectory data to identify parameter change characteristics of different depth segments, including change trends, change rates, and mutual correlations, thereby obtaining layered feature data; Step S274: constructing a dynamic model of the three-dimensional parameter space according to the hierarchical feature data, establishing a continuous expression of parameter changes through spatial interpolation and numerical fitting, thereby obtaining dynamic evolution model data; Step S275: extracting the gradient features, curvature features and speed features of the dynamic evolution model data based on parameter changes, thereby obtaining model feature data; Step S276: performing parameter correlation analysis based on the coupling relationship and mutual feedback mechanism between parameters based on the model characteristic data, thereby obtaining parameter correlation data; Step S277: extracting indicators characterizing the dynamic change rules of parameters based on the model characteristic data and the parameter association data, thereby generating parameter change characteristic data.

7. The real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles according to claim 6 is characterized in that: Step S3 includes the following steps: Step S31: establishing a parameter-stratum characteristic correspondence table including a mapping relationship between parameter changes and soil responses in different strata according to the parameter change characteristic data and the stratum distribution data, thereby obtaining parameter response data; Step S32: performing data standardization and feature dimension reduction processing on the parameter response data, and establishing a mathematical model of parameter-formation response, thereby obtaining response model data; Step S33: Perform machine learning training based on the response model data to construct a nonlinear mapping relationship between the parameter and the formation response, thereby obtaining a parameter-formation response relationship model; Step S34: performing slurry diffusion characteristic analysis including pressure propagation, soil penetration and slurry solidification according to the parameter-stratum response relationship model, thereby obtaining diffusion characteristic data; Step S35: Calculating the diffusion range related to the depth based on the diffusion characteristic data to obtain diffusion radius data, wherein the diffusion range calculation includes the influence analysis of the slurry diffusion pressure attenuation law and the soil shear strength; Step S36: identifying key features and mutation points in the slurry diffusion process on the diffusion dynamic data, thereby obtaining the slurry diffusion dynamic data.

8. The real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles according to claim 7 is characterized in that: Step S4 includes the following steps: Step S41: analyzing the influence of different parameter combinations on soil improvement effects based on slurry diffusion dynamic data and real-time high-pressure rotary jetting data, thereby obtaining improvement impact data; Step S42: evaluating the performance of the improved body based on the strength development law, curing time characteristics and deformation characteristics of the improved impact data, thereby obtaining performance evaluation data; Step S43: analyzing the evolution characteristics of the improvement effect over time on the performance evaluation data, thereby obtaining soil improvement dynamic characteristic data; Step S44: numerically simulate the pile influence area according to the soil improvement dynamic characteristic data, and establish a multi-field coupling analysis model including stress field, displacement field and seepage field, so as to obtain the influence area data; Step S45: performing boundary recognition and spatial partitioning on the impact area data, thereby obtaining pile body impact range data; Step S46: coupling the stratum stress mode based on the injection pressure, soil stress and improvement effect according to the pile influence range data, identifying the stress field evolution law and critical state characteristics, and generating soil action coupling data.

9. The real-time dynamic prediction method of the three-dimensional shape of high-pressure jet grouting piles according to claim 8 is characterized in that: Step S5 includes the following steps: Step S51: extracting features from soil action coupling data and constructing a training sample set for a deep learning model to obtain training sample data, wherein the training sample data includes input features and morphological labels; Step S52: constructing a deep learning network structure based on a hybrid framework of a convolutional neural network and a recurrent neural network according to the training sample data, and performing model training to obtain a morphology prediction model; Step S53: optimizing the model prediction accuracy of the morphology prediction model based on a cross-validation method, thereby obtaining prediction model data; Step S54: dynamically predicting the three-dimensional shape of the pile body using the prediction model data, thereby obtaining shape prediction data, wherein the shape prediction data includes cross-sectional shape, vertical change, and spatial distribution; Step S55: The key stages and morphological mutation characteristics of the pile body formation process are analyzed for the morphological prediction data, thereby obtaining morphological evolution data, wherein the key stages include the initial injection stage, the diffusion development stage, the initial solidification stage, the strength development stage, the stable forming stage, and the adjacent pile body influence stage; Step S56: Optimize the construction parameters of each stage in real time according to the morphological evolution data, so as to obtain high-pressure jet grouting construction control feedback data.

10. A real-time dynamic prediction system for the three-dimensional shape of high-pressure jet grouting piles, characterized in that: The method for real-time dynamic prediction of the three-dimensional shape of high-pressure jet grouting piles according to claim 1, wherein the real-time dynamic prediction system for the three-dimensional shape of high-pressure jet grouting piles comprises: The site monitoring module is used to identify the stratum characteristics of the construction site and obtain stratum distribution data; determine the key monitoring points based on the stratum distribution data and establish a multi-source sensor monitoring network; The parameter acquisition module is used to perform anti-interference optimization processing on the multi-source sensor monitoring network and to collect high-pressure rotary spray parameters in real time to obtain real-time high-pressure rotary spray data; the real-time high-pressure rotary spray data is processed based on the spatiotemporal evolution of the injection pressure, lifting speed and rotation speed to generate parameter change characteristic data; The diffusion prediction module is used to establish a parameter-formation response relationship model based on parameter change characteristic data and formation distribution data; analyze the slurry diffusion characteristics based on the parameter-formation response relationship model, and evaluate the depth-related diffusion radius to generate slurry diffusion dynamic data; The impact assessment module is used to perform soil improvement correlation processing based on the slurry diffusion dynamic data and real-time high-pressure rotary jetting data to generate soil improvement dynamic characteristic data; to identify the pile impact area based on the soil improvement dynamic characteristic data to generate pile impact range data; to couple the formation stress mode based on the injection pressure, soil stress and improvement effect based on the pile impact range data to generate soil action coupling data; The morphology control module is used to build a deep learning prediction model based on the soil action coupling data, and use the deep learning prediction model to dynamically predict the three-dimensional morphology of the pile to generate morphology prediction data; based on the morphology prediction data, the construction parameters are optimized in real time to obtain high-pressure rotary jet construction control feedback data.

Images