Stiff composite pile construction quality control method based on global intelligent sensing

Abstract

The invention discloses a universe intelligent sensing stiff composite pile construction quality control method which comprises the following steps: before construction, arranging a quantum gravity gradiometer array and an electromagnetic signal receiving and transmitting device on a site, and obtaining universe gravity field baseline data and underground dielectric constant distribution before construction; according to material design, iron powder or titanium powder and other electromagnetic sensitive materials are doped into cement soil, the mass ratio is 0.01%-5%, and excitable-responsive intelligent cement soil is formed; sensing in construction; and verification after construction: based on quantum-electromagnetic cooperative detection and an AI prediction model, directly outputting a pile body integrity, bearing capacity and settlement characteristic evaluation report to replace traditional static load test and core drilling detection. The quality is controllable, the construction period is shortened, and sedimentation is reduced.

Description

Technical Field

[0001] This invention relates to a method for quality control during the construction of rigid composite piles. Background Technology

[0002] Traditional quality inspection methods for reinforced composite piles have the following problems: Delayed testing: Static load tests and core drilling tests require 28 days of curing, which cannot guide construction; Spot sampling: A 1% sampling rate cannot represent the overall quality, and the risks are hidden. Parameter isolation: It is impossible to obtain the continuous evolution of moisture content-density-stress-cementation degree; Reliance on experience: The water-cement ratio, spraying volume, and pile driving force control depend on the experience of the workers.

[0003] Existing technologies can only be used for post-piling testing and cannot be used for control during construction; quantum gravity exploration is mainly used for mineral exploration and has not been integrated with geotechnical engineering construction. Summary of the Invention

[0004] The purpose of this invention is to provide a method for controlling the construction quality of stiffened composite piles with full-area intelligent sensing, which enables quality control, reduces construction period, and reduces settlement.

[0005] The technical solution of this invention is: A method for controlling the construction quality of stiffened composite piles with full-domain intelligent sensing, characterized by the following steps: Before construction: A quantum gravity gradiometer array and an electromagnetic signal transceiver were deployed on site to obtain baseline data of the global gravity field and the distribution of underground dielectric constant before construction. Material design: Electromagnetically sensitive materials such as iron powder or titanium powder are added to cement-soil, accounting for 0.01%-5% by mass, to form intelligent cement-soil that can be activated and responded to. Sensed during construction: Data on current, back pressure, and depth are collected by sensing the self-tapping mechanism of the drill bit's spiral. Using a quantum gravity gradiometer to monitor changes in underground density caused by construction; The electromagnetic response of the magnetically doped cement soil was excited and received by the wide-area electromagnetic method, and the water content, porosity, and cementation degree were inverted. Data fusion and reverse regulation: Built with trained Physical Information Neural Network (PINN) and Graph Neural Network (GNN) models, this AI system is capable of: inverse identification: deducing key mechanical parameters (such as elastic modulus and shear strength) of soil and cement-soil based on real-time physical field changes; forward prediction: predicting diagenetic strength, final bearing capacity, and post-construction settlement under different construction parameters; real-time control: dynamically outputting optimal construction instructions; and inputting multi-source data from gravity, electromagnetics, mechanics, and acoustic waves into the Physical Information Neural Network (PINN). Dynamic inversion of stress field and diagenetic process in pile-soil-concrete integrated system; The water-cement ratio, spraying volume, re-mixing frequency, and pile driving force are adjusted in real time based on the inversion results. Post-construction verification: Based on quantum-electromagnetic collaborative detection and AI prediction model, it directly outputs evaluation reports on pile integrity, bearing capacity and settlement characteristics, replacing traditional static load tests and core drilling.

[0006] The system adopts a full-domain intelligent perception and control system, including: Quantum-electromagnetic cooperative detection module: A quantum gravity gradiometer array with a sensitivity of 1 μGal is used to monitor the evolution of underground density fields. A wide-area electromagnetic transmitting-receiving system, with a frequency range of 0.1-100 Hz, is used to excite magnetically doped cement-soil and collect the response. Smart Materials Module: Electromagnetically sensitive cement-soil is composed of cement, in-situ soil, iron / titanium powder (0.01%-5%), and geopolymers. Intelligent pipe piles with built-in optical fibers and piezoelectric ceramics are used for stress-strain-fluctuation signal acquisition; Data fusion and AI analysis platform: Multi-source database, integrating gravity, electromagnetic, acoustic, mechanical, and thermal data; PINN-Graph Neural Network (GNN) hybrid model for material property prediction and process reverse engineering; A 3D digital twin engine enables real-time visualization of the integrated pile-soil-concrete process.

[0007] The process of exciting and receiving the electromagnetic response of the magnetically doped cement-soil using a wide-area electromagnetic method, and then inverting the water content, porosity, and cementation degree, specifically includes: The underground density distribution ρ(x,y,z) is inferred using the gravity gradient variation Δgz; By combining electromagnetic apparent resistivity ρe and phase φ, water content w, porosity n, and cementation degree Ic are jointly inverted. By coupling the inversion results with the pile driving force-depth curve through a data assimilation algorithm, a stress-strain-wave velocity joint constitutive model is constructed.

[0008] The AI ​​analysis platform has reverse engineering capabilities and can: Based on the target bearing capacity and settlement requirements, the optimal water-cement ratio, cement content, and pile driving force are derived in reverse. Based on the GNoME-style material exploration algorithm, the optimal ratio of geopolymer-cement-sensitive material is recommended; Through reinforcement learning, the agent autonomously adjusts the stirring speed, back pressure, and pressure stabilization time.

[0009] The post-construction verification includes: Quantum gravity residual analysis: Comparison of the changes in the gravity field before and after construction with the design expectation; a deviation of <5% is considered acceptable. Electromagnetic response spectrum identification: The amplitude-phase characteristics of the response in a specific frequency band match the benchmark cases in the database with a degree of >90%; AI-based comprehensive rating: Outputs a digital certificate of conformity for each pile, including predicted bearing capacity, settlement curve, and durability index.

[0010] The core innovation of this invention: 1. Integration of detection technologies: The quantum gravity gradiometer captures subtle changes in underground density fields, with a sensitivity 100 times greater than that of traditional gravimeters; The degree of cementation and water content of magnetically doped cement soil were inverted by using the dispersion characteristics after the wide-area electromagnetic method was used to excite the magnetically doped cement soil. The sensing drill bit and intelligent pipe pile collect mechanical, thermal, and acoustic data to form a global sensing network.

[0011] 2. Material-Sensing Integration: The incorporation of electromagnetically sensitive materials transforms cement-soil into a smart medium with its own tags. Quantum-electromagnetic synergy enables non-contact, non-destructive, and continuous tracking of medium properties.

[0012] 3. AI-driven reverse regulation: The PINN+GNN hybrid model integrates physical equations and big data to achieve multi-field coupled inversion. The reinforcement learning agent autonomously adjusts construction parameters based on the inversion results, forming an intelligent closed loop.

[0013] Technical effects: Quality controllable: Real-time evaluation and control of each pile and each process; A revolutionary testing method: replacing traditional testing methods such as static load, core drilling, and low strain, saving over 80% of the project time; Performance Enhancement: Through AI reverse engineering, load-bearing capacity is increased by 20%, and settlement is reduced by 50%; Standardization: Output digital certificates of conformity to achieve quality traceability and standardized delivery.

[0014] The present invention will be further described below with reference to the embodiments. Detailed Implementation

[0015] A method for controlling the construction quality of stiffened composite piles with full-domain intelligent sensing includes the following steps: Before construction: A quantum gravity gradiometer array and an electromagnetic signal transceiver were deployed on site to obtain baseline data of the global gravity field and the distribution of underground dielectric constant before construction. Material design: Electromagnetically sensitive materials such as iron powder or titanium powder are added to cement-soil, accounting for 0.01%-5% by mass, to form intelligent cement-soil that can be activated and responded to. Sensed during construction: Data on current, back pressure, and depth are collected by sensing the self-tapping mechanism of the drill bit's spiral. Using a quantum gravity gradiometer to monitor changes in underground density caused by construction; The electromagnetic response of the magnetically doped cement soil was excited and received by the wide-area electromagnetic method, and the water content, porosity, and cementation degree were inverted. Data fusion and reverse regulation: Built with trained Physical Information Neural Network (PINN) and Graph Neural Network (GNN) models, this AI system is capable of: inverse identification: deducing key mechanical parameters (such as elastic modulus and shear strength) of soil and cement-soil based on real-time physical field changes; forward prediction: predicting diagenetic strength, final bearing capacity, and post-construction settlement under different construction parameters; real-time control: dynamically outputting optimal construction instructions; and inputting multi-source data from gravity, electromagnetics, mechanics, and acoustic waves into the Physical Information Neural Network (PINN). Dynamic inversion of stress field and diagenetic process in pile-soil-concrete integrated system; The water-cement ratio, spraying volume, re-mixing frequency, and pile driving force are adjusted in real time based on the inversion results. Post-construction verification: Based on quantum-electromagnetic collaborative detection and AI prediction model, it directly outputs evaluation reports on pile integrity, bearing capacity and settlement characteristics, replacing traditional static load tests and core drilling.

[0016] The system adopts a full-domain intelligent perception and control system, including: Quantum-electromagnetic cooperative detection module: A quantum gravity gradiometer array with a sensitivity of 1 μGal is used to monitor the evolution of underground density fields. A wide-area electromagnetic transmitting-receiving system, with a frequency range of 0.1-100 Hz, is used to excite magnetically doped cement-soil and collect the response. Smart Materials Module: Electromagnetically sensitive cement-soil is composed of cement, in-situ soil, iron / titanium powder (0.01%-5%), and geopolymers. Intelligent pipe piles with built-in optical fibers and piezoelectric ceramics are used for stress-strain-fluctuation signal acquisition; Data fusion and AI analysis platform: Multi-source database, integrating gravity, electromagnetic, acoustic, mechanical, and thermal data; PINN-Graph Neural Network (GNN) hybrid model for material property prediction and process reverse engineering; A 3D digital twin engine enables real-time visualization of the integrated pile-soil-concrete process.

[0017] The process of exciting and receiving the electromagnetic response of the magnetically doped cement-soil using a wide-area electromagnetic method, and then inverting the water content, porosity, and cementation degree, specifically includes: The underground density distribution ρ(x,y,z) is inferred using the gravity gradient variation Δgz; By combining electromagnetic apparent resistivity ρe and phase φ, water content w, porosity n, and cementation degree Ic are jointly inverted. By coupling the inversion results with the pile driving force-depth curve through a data assimilation algorithm, a stress-strain-wave velocity joint constitutive model is constructed.

[0018] The AI ​​analysis platform has reverse engineering capabilities and can: Based on the target bearing capacity and settlement requirements, the optimal water-cement ratio, cement content, and pile driving force are derived in reverse. Based on the GNoME-style material exploration algorithm, the optimal ratio of geopolymer-cement-sensitive material is recommended; Through reinforcement learning, the agent autonomously adjusts the stirring speed, back pressure, and pressure stabilization time.

[0019] The post-construction verification includes: Quantum gravity residual analysis: Comparison of the changes in the gravity field before and after construction with the design expectation; a deviation of <5% is considered acceptable. Electromagnetic response spectrum identification: The amplitude-phase characteristics of the response in a specific frequency band match the benchmark cases in the database with a degree of >90%; AI-based comprehensive rating: Outputs a digital certificate of conformity for each pile, including predicted bearing capacity, settlement curve, and durability index.

[0020] System deployment: Eight quantum gravity gradiometers were arranged around the tank to monitor the density field; The wide-area electromagnetic system is laid out with intersecting measurement lines, and the excitation frequency is 0.1-10 Hz; 0.03% iron powder is added to the cement-soil mixture, and fiber optic gratings are pre-embedded in the intelligent pipe piles.

[0021] Adjustments during construction: Gravity data showed that the density in a certain area was low, and the AI ​​platform suggested increasing the amount of ash sprayed by 10%. Electromagnetic inversion revealed that the water content of the bearing layer was too high, so the process of re-stirring and re-pressurizing was initiated three times. The stress bubble distribution is fed back in real time during pile driving, and the stabilization time is adjusted to 30 minutes.

[0022] Results Verification: AI outputs a full-domain digital certificate of conformity, with a deviation of less than 5% between predicted and actual load-bearing capacity; Compared with traditional testing, the pass rate is 100%, and the testing cost is reduced by 70%.

[0023] Benefits: Construction period shortened by 60%, costs reduced by 40%.

Claims

1. A method for controlling the construction quality of stiffened composite piles with full-domain intelligent sensing, characterized in that, Includes the following steps: Before construction: A quantum gravity gradiometer array and an electromagnetic signal transceiver were deployed on site to obtain baseline data of the global gravity field and the distribution of underground dielectric constant before construction. Material design: Electromagnetically sensitive materials such as iron powder or titanium powder are added to cement-soil, accounting for 0.01%-5% by mass, to form intelligent cement-soil that can be activated and responded to. Sensed during construction: Data on current, back pressure, and depth are collected by sensing the self-tapping mechanism of the drill bit's spiral. Using a quantum gravity gradiometer to monitor changes in underground density caused by construction; The electromagnetic response of the magnetically doped cement soil was excited and received by the wide-area electromagnetic method, and the water content, porosity, and cementation degree were inverted. Data fusion and reverse regulation: Built with a trained physical information neural network and graph neural network model, this AI system is capable of: reverse identification: deducing key mechanical parameters of soil and cement-soil based on real-time physical field changes; forward prediction: predicting diagenetic strength, final bearing capacity and post-construction settlement under different construction parameters; real-time control: dynamically outputting optimal construction instructions; and inputting multi-source data of gravity, electromagnetics, mechanics and sound waves into the physical information neural network. Dynamic inversion of stress field and diagenetic process in pile-soil-concrete integrated system; The water-cement ratio, spraying volume, re-mixing frequency, and pile driving force are adjusted in real time based on the inversion results. Post-construction verification: Based on quantum-electromagnetic collaborative detection and AI prediction model, it directly outputs evaluation reports on pile integrity, bearing capacity and settlement characteristics, replacing traditional static load tests and core drilling.

2. The method for controlling the construction quality of rigid composite piles with full-domain intelligent sensing according to claim 1, characterized in that, The system adopts a full-domain intelligent perception and control system, including: Quantum-electromagnetic cooperative detection module: A quantum gravity gradiometer array with a sensitivity of 1 μGal is used to monitor the evolution of underground density fields. A wide-area electromagnetic transmitting-receiving system, with a frequency range of 0.1-100 Hz, is used to excite magnetically doped cement-soil and collect the response. Smart Materials Module: Electromagnetically sensitive cement-soil is composed of cement, in-situ soil, iron / titanium powder, and geopolymer. Intelligent pipe piles with built-in optical fibers and piezoelectric ceramics are used for stress-strain-fluctuation signal acquisition; Data fusion and AI analysis platform: Multi-source database, integrating gravity, electromagnetic, acoustic, mechanical, and thermal data; PINN-Graph Neural Network Hybrid Model for Material Property Prediction and Process Reverse Engineering; A 3D digital twin engine enables real-time visualization of the integrated pile-soil-concrete process.

3. The method for controlling the construction quality of rigid composite piles with full-domain intelligent sensing according to claim 1, characterized in that, The process of exciting and receiving the electromagnetic response of the magnetically doped cement-soil using a wide-area electromagnetic method, and then inverting the water content, porosity, and cementation degree, specifically includes: The underground density distribution ρ(x,y,z) is inferred using the gravity gradient variation Δgz; By combining electromagnetic apparent resistivity ρe and phase φ, water content w, porosity n, and cementation degree Ic are jointly inverted. By coupling the inversion results with the pile driving force-depth curve through a data assimilation algorithm, a stress-strain-wave velocity joint constitutive model is constructed.

4. The method for controlling the construction quality of rigid composite piles with full-domain intelligent sensing according to claim 2, characterized in that, The AI ​​analysis platform has reverse engineering capabilities and can: Based on the target bearing capacity and settlement requirements, the optimal water-cement ratio, cement content, and pile driving force are derived in reverse. Based on the GNoME-style material exploration algorithm, the optimal ratio of geopolymer-cement-sensitive material is recommended; Through reinforcement learning, the agent autonomously adjusts the stirring speed, back pressure, and pressure stabilization time.

5. The method for controlling the construction quality of rigid composite piles with full-domain intelligent sensing according to claim 1, characterized in that: The post-construction verification includes: Quantum gravity residual analysis: Comparison of the changes in the gravity field before and after construction with the design expectation; a deviation of <5% is considered acceptable. Electromagnetic response spectrum identification: The amplitude-phase characteristics of the response in a specific frequency band match the benchmark cases in the database with a degree of >90%; AI-based comprehensive rating: Outputs a digital certificate of conformity for each pile, including predicted bearing capacity, settlement curve, and durability index.