Microbial circulating grouting foundation reinforcement system and method based on multi-source information fusion and artificial intelligence feedback

Abstract

The invention provides a microbial circulating grouting foundation reinforcement system and method based on multi-source information fusion and artificial intelligence feedback. The reinforcement system comprises a sensing module, a decision control module and an execution module. The sensing module collects multi-dimensional physical and chemical field data representing microbial reaction states and soil body properties in real time through a multi-source sensor network, and the decision control module analyzes the data collected by the multi-source sensor network, issues a control instruction, controls working parameters of an electromagnetic valve and a metering pump of the execution module in real time, and achieves control over grouting reinforcement. According to the method, on-demand generation of calcium carbonate precipitates in a three-dimensional space is achieved through real-time sensing and feedback control, the problem of uneven reinforcement is fundamentally solved, and the engineering quality is remarkably improved; through real-time intervention on key indexes such as pH and ion concentration, the biochemical reaction is always maintained in an optimal active window, the inhibition effect of by-products is avoided, and the reaction efficiency and the material utilization rate are improved.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of geotechnical engineering, environmental microbiology technology and intelligent control systems, and in particular relates to a system and method for microbial cyclic grouting reinforcement of foundations based on multi-source information fusion and artificial intelligence feedback. Background Technology

[0002] Microbially induced calcium carbonate precipitation (MICP) technology, as a green soil improvement method, can enhance soil strength and stiffness by using microbial metabolites to bind soil particles. However, traditional MICP technology has significant bottlenecks: Uncontrollable process: The grouting process relies on pre-set experience and fixed parameters, making adjustments impossible based on real-time biochemical reactions within the soil, leading to uneven reinforcement effects; Uneven migration of reaction fluids: Due to the anisotropy and heterogeneity of the soil, the infiltration paths of the bacterial solution and cementing fluid are complex, easily forming dominant channels, resulting in local over-reinforcement and insufficient reinforcement in other areas; Inhibition of reaction byproducts: Rapid hydrolysis of urea leads to a sharp increase in local pH and ammonia accumulation, inhibiting microbial activity and even terminating the reaction, reducing reinforcement efficiency; Delayed quality assessment: The assessment of reinforcement effects relies on point-based, destructive testing such as core drilling or static cone penetration testing after reinforcement, making real-time, comprehensive monitoring and feedback of the reinforcement process impossible. It is evident that existing technologies lack a solution that can deeply integrate process monitoring and real-time control, making it difficult for MICP technology to be widely applied in critical engineering projects with high uniformity requirements. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a system and method for microbial cyclic grouting reinforcement of foundations based on multi-source information fusion and artificial intelligence feedback, achieving deep integration of process monitoring and real-time control.

[0004] The present invention achieves the above-mentioned technical objectives through the following technical means.

[0005] A system for reinforcing foundations using microbial cyclic grouting based on multi-source information fusion and artificial intelligence feedback includes a sensing module, a decision control module, and an execution module. The sensing module includes a multi-source sensor network deployed within the foundation reinforcement area to collect multi-dimensional physical and chemical field data characterizing the microbial reaction state and soil properties in real time. The decision control module includes a computing unit and an intelligent regulation algorithm model stored therein to receive data streams from the sensing module, diagnose the current reaction state and predict development trends in real time through data fusion and analysis, and generate targeted grouting regulation commands. The execution module includes a multi-channel grouting unit controlled by the decision control module, which delivers bacterial solutions, cementing solutions, or flushing solutions with different ratios and flow rates to different locations within the reinforcement area according to the regulation commands.

[0006] Furthermore, the sensing module includes one or more of the following sensors: a distributed fiber optic sensor for continuously monitoring the temperature and strain field of the soil; a resistivity tomography system for inverting the three-dimensional distribution of soil resistivity and visualizing the calcium carbonate precipitation process; a pH sensor array for monitoring the spatial distribution and dynamic changes of pH value in the soil; an ion-selective electrode for monitoring the consumption of calcium ion concentration; and a pore water pressure sensor for monitoring the effect of grouting pressure on soil structure.

[0007] Furthermore, the intelligent regulation algorithm model in the decision control module is a machine learning-based model. It is trained with historical data to establish a mapping relationship between sensor data and microbial activity, sedimentation rate and location, and outputs optimization instructions including grouting location, grouting rate, grouting pressure, grouting liquid type and ratio.

[0008] Furthermore, the multi-channel grouting unit includes storage tanks for storing bacterial solution, urea solution, and calcium solution respectively, and also includes multiple grouting pipelines connected to the storage tanks and independently controlled by metering pumps and solenoid valves. The solenoid valves and metering pumps are signal-connected to the decision control module.

[0009] A method for microbial cyclic grouting reinforcement of foundations utilizing the aforementioned microbial cyclic grouting reinforcement system based on multi-source information fusion and artificial intelligence feedback includes the following steps: S1: Based on the geological survey results, grouting holes and a multi-source sensor network are planned and deployed in the area to be reinforced; S2: Start the system. Under the control of the decision control module, perform preliminary grouting with conservative parameters, and collect data from the multi-source sensor network to establish a site response baseline. S3: Closed-loop intelligent control cycle: The sensing module collects multi-dimensional data in real time and transmits it to the decision control module; the decision control module uses the intelligent control algorithm model to process the data, diagnose the current reaction state, predict future trends, and generate real-time control instructions; the execution module receives and executes the control instructions to dynamically adjust the parameters of specific grouting channels; S4: When the decision control module analyzes and determines that the foundation reinforcement has reached the preset target based on the data transmitted by the multi-source sensor network, or when it analyzes and determines that the pH value change reaction in each area tends to be stable, the grouting cycle is terminated; then, a reinforcement effect evaluation report is generated based on the final data from the sensing module.

[0010] The present invention has the following beneficial effects: Precise and uniform: Through real-time sensing and feedback control, calcium carbonate precipitation is generated on demand in three-dimensional space, which fundamentally solves the problem of uneven reinforcement and significantly improves the quality of the project.

[0011] Process controllability: By intervening in key indicators such as pH and ion concentration in real time, the biochemical reaction is always maintained at the optimal activity window, avoiding the inhibitory effect of by-products and improving reaction efficiency and material utilization.

[0012] Intelligent and adaptive: The system has self-learning and self-decision-making capabilities, can adapt to changes in geological conditions at different sites, reduces reliance on operator experience, and promotes the standardization and industrialization of the technology.

[0013] Non-destructive real-time monitoring: The entire reinforcement process is also a quality inspection process, realizing full-area, real-time, and non-destructive monitoring of the reinforcement effect, replacing the traditional post-incident destructive testing, and saving costs and time. Attached Figure Description

[0014] Figure 1 This is a flowchart of the microbial cyclic grouting method for foundation reinforcement described in this invention. Detailed Implementation

[0015] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0016] Example 1: A microbial cyclic grouting foundation reinforcement system based on multi-source information fusion and artificial intelligence feedback includes a sensing module, a decision control module, and an execution module.

[0017] Sensing module: includes a multi-source sensor network deployed in the foundation reinforcement area, used to collect multi-dimensional physical and chemical field data that characterize the microbial reaction state and soil properties in real time.

[0018] Decision control module: This module includes a computing unit and an intelligent control algorithm model stored within it. It is configured to receive data streams from the sensing module, and through data fusion and analysis, diagnose the reaction status in real time, predict development trends, and generate targeted grouting control commands.

[0019] Execution module: includes a multi-channel grouting unit; the multi-channel grouting unit is controlled by the decision control module and can independently and accurately deliver bacterial solution, cementing solution or flushing solution with different ratios and flow rates to different locations in the reinforced area according to the control instructions.

[0020] The sensing module includes one or more of the following sensors: a distributed fiber optic sensor for continuous monitoring of soil temperature and strain field; a resistivity tomography system for inverting the three-dimensional distribution of soil resistivity and visualizing the calcium carbonate precipitation process; a pH sensor array for monitoring the spatial distribution and dynamic changes of pH value in the soil; an ion-selective electrode for monitoring the consumption of calcium ion concentration; and a pore water pressure sensor for monitoring the effect of grouting pressure on soil structure.

[0021] The intelligent regulation algorithm model in the decision control module is a machine learning-based model. Through training with historical data, it can establish a mapping relationship between sensor data and microbial activity, sedimentation rate and location, and output optimization instructions including grouting location, grouting rate, grouting pressure, grouting liquid type and ratio.

[0022] The multi-channel grouting unit in the execution module includes storage tanks for storing bacterial solution, urea solution, and calcium solution, as well as a metering pump and multiple grouting pipelines controlled by solenoid valves. The solenoid valves and metering pumps are connected to the decision control module via signals.

[0023] The microbial cyclic grouting foundation reinforcement method based on multi-source information fusion and artificial intelligence feedback described in this invention is as follows: Figure 1 As shown, the process includes the following: S1: Site survey and system deployment: Based on the geological survey results, a multi-source sensor network of grouting holes and sensing modules was planned and deployed in the reinforcement area. S2: System Initialization and Baseline Establishment The system was started to perform preliminary grouting with conservative parameters, while data collected from a multi-source sensor network was collected to establish a site response baseline. S3: Closed-loop intelligent control cycle: S31: The sensing module collects multi-dimensional data in real time and transmits it to the decision control module; S32: The decision control module uses an intelligent regulation algorithm model to process data, diagnose the current response state, predict future trends, and generate real-time regulation commands. S33: The execution module receives and executes control commands to dynamically adjust the parameters of a specific grouting channel; S4: Cycle Termination and Effect Evaluation: When the decision control module determines that the reinforcement indicators have reached the preset target, or the pH value change reaction in each area tends to stabilize, the grouting cycle is terminated; then, a reinforcement effect evaluation report is generated based on the final data from the sensing module.

[0024] Example 2: This embodiment uses a coastal soft soil foundation reinforcement project as an example to illustrate the foundation reinforcement method: S1: System Deployment: Grouting holes are arranged in a 1.2m × 1.2m quincunx pattern in the area to be reinforced, with a hole depth of 8m. At the same time, distributed fiber optic sensors, ERT electrodes, and pH sensor arrays are arranged in parallel in the area to be reinforced. All sensors are connected to the central decision control cabinet on the ground (which contains a decision control module). The corresponding pipelines of the three storage tanks containing bacterial solution, urea solution, and calcium chloride solution in the multi-channel grouting unit of the execution module are connected to each grouting hole.

[0025] S2: Initialization: The system was started, and a low-concentration bacterial solution and cementing solution were injected into all grouting holes at a pressure of 0.15 MPa and a rate of 1.5 L / min. During this period, the system recorded the initial background fields of temperature, resistivity, and pH.

[0026] S3: Intelligent Control Cycle S31: Real-time perception: The multi-source sensor network begins to continuously send data streams to the decision control module; S32: Intelligent Decision-Making: The AI ​​model (i.e., intelligent regulation algorithm model) of the decision control module analyzes the data and finds that the resistivity and pH value increase rate of region A are much faster than those of region B. The model diagnoses that region A is reacting too quickly and has the risk of pore blockage; region B is reacting slowly. Based on this, the model generates the instruction: "Reduce the cementing fluid injection rate of all grouting holes in region A to 0.8 L / min and pause the injection for one round; at the same time, increase the bacterial solution injection pressure in region B to 0.2 MPa to enhance its permeability."

[0027] S33: Precise Execution: The decision control module immediately drives the solenoid valve and metering pump on the corresponding grouting pipeline to complete the above adjustments.

[0028] S4: Loop Continuation and Termination: The aforementioned intelligent control cycle continues. In subsequent cycles, the system detects that the calcium ion concentration in area C is decreasing slowly, indicating insufficient microbial activity. Therefore, it sends a command to inject a new round of highly active bacterial solution. Throughout the process, the underground resistivity cloud map displayed by the ERT system clearly reflects that the reinforced soil is developing towards homogenization. When the soil stiffness increment measured by the distributed fiber optic sensor reaches the design standard and the pH value change in each area tends to be gradual, the AI ​​model determines that the reinforcement target has been achieved, automatically stops all grouting operations, and generates a reinforcement effect evaluation report.

[0029] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A microbial cyclic grouting foundation reinforcement system based on multi-source information fusion and artificial intelligence feedback, characterized in that, It includes a sensing module, a decision control module, and an execution module. The sensing module includes a multi-source sensor network deployed in the area to be reinforced, used to collect multi-dimensional physical and chemical field data characterizing the microbial reaction state and soil properties in real time. The decision control module includes a computing unit and an intelligent regulation algorithm model stored therein, used to receive the data stream from the sensing module, and through data fusion and analysis, diagnose the current reaction state in real time, predict the development trend, and generate targeted grouting regulation commands. The execution module includes a multi-channel grouting unit, which is controlled by the decision control module and delivers bacterial solutions, cementing solutions, or flushing solutions with different ratios and flow rates to different locations in the area to be reinforced according to the regulation commands.

2. The microbial cyclic grouting foundation reinforcement system based on multi-source information fusion and artificial intelligence feedback as described in claim 1, characterized in that, The sensing module includes one or more of the following sensors: a distributed fiber optic sensor for continuous monitoring of soil temperature and strain field; and a resistivity tomography system for inverting the three-dimensional distribution of soil resistivity and visualizing the calcium carbonate precipitation process. pH sensor array is used to monitor the spatial distribution and dynamic changes of pH value in soil. An ion-selective electrode is used to monitor the consumption of calcium ion concentration; a pore water pressure sensor is used to monitor the impact of grouting pressure on soil structure.

3. The microbial cyclic grouting foundation reinforcement system based on multi-source information fusion and artificial intelligence feedback as described in claim 1, characterized in that, The intelligent regulation algorithm model in the decision control module is a machine learning-based model. It is trained with historical data to establish a mapping relationship between sensor data and microbial activity, sedimentation rate and location, and outputs optimization instructions including grouting location, grouting rate, grouting pressure, grouting liquid type and ratio.

4. The microbial cyclic grouting foundation reinforcement system based on multi-source information fusion and artificial intelligence feedback as described in claim 1, characterized in that, The multi-channel grouting unit includes storage tanks for storing bacterial solution, urea solution, and calcium solution respectively, and also includes multiple grouting pipelines connected to the storage tanks and independently controlled by metering pumps and solenoid valves. The solenoid valves and metering pumps are signal-connected to the decision control module.

5. A method for microbial cyclic grouting reinforcement of a foundation using the microbial cyclic grouting reinforcement system based on multi-source information fusion and artificial intelligence feedback as described in claim 1, characterized in that, The process includes the following: S1: Based on the geological survey results, grouting holes and a multi-source sensor network are planned and deployed in the area to be reinforced; S2: Start the system. Under the control of the decision control module, perform preliminary grouting with conservative parameters, and collect data from the multi-source sensor network to establish a site response baseline. S3: Closed-loop intelligent control cycle: The sensing module collects multi-dimensional data in real time and transmits it to the decision control module; the decision control module uses the intelligent control algorithm model to process the data, diagnose the current reaction state, predict future trends, and generate real-time control instructions; the execution module receives and executes the control instructions to dynamically adjust the parameters of specific grouting channels; S4: When the decision control module analyzes and determines that the foundation reinforcement has reached the preset target based on the data transmitted by the multi-source sensor network, or analyzes and determines that the pH value change reaction in each area tends to be stable, the grouting cycle is terminated. Subsequently, an evaluation report on the reinforcement effect is generated based on the final data from the sensing module.

Images