Expressway pile foundation construction hole forming quality intelligent monitoring and early warning system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGMEI ENGINEERING GROUP LTD
- Filing Date
- 2026-03-11
- Publication Date
- 2026-05-29
Smart Images
Figure CN122106552A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of geotechnical engineering monitoring technology, specifically to an intelligent monitoring and early warning system for the quality of borehole formation in highway pile foundation construction. Background Technology
[0002] In the construction of large-scale infrastructure such as highways and bridges, pile foundations are key structural units that bear the load of the superstructure and transfer it to the deep, stable foundation. The bearing capacity and stability of pile foundations largely depend on the quality of their borehole formation. Pipe hole formation is a hidden project carried out underground. The complexity and uncertainty of geological conditions during construction, such as weak interlayers, karst development, or groundwater activity, often lead to borehole instability, resulting in quality defects such as borehole collapse, diameter reduction, borehole deviation, or excessive sediment. These defects directly affect the smooth progress of subsequent reinforcement cage placement and concrete pouring, and pose a serious threat to the final bearing capacity and long-term service safety of the pile foundation.
[0003] Existing borehole quality monitoring technologies mainly rely on monitoring drilling rig operating parameters and post-drilling inspection. By monitoring parameters such as drilling rig torque, rotation speed, and feed pressure, construction personnel can indirectly infer lithological changes in the strata where the drill bit is located. However, this method cannot provide direct information about the borehole geometry and lacks effective perception of local instability in the borehole wall.
[0004] Post-drilling inspection methods, such as borehole inspection or traditional ultrasonic testing, can obtain some borehole diameter information, but these inspections occur after the drilling process is completed. This is a kind of delayed quality verification and cannot provide real-time feedback and early warning during construction. Once a serious defect is discovered, the best time to deal with it has often been missed. Rework or remedial measures are not only costly, but also difficult to guarantee the effectiveness.
[0005] To achieve process control, some technical solutions attempt to measure during drilling. However, these solutions can usually only obtain borehole diameter information at discrete depth points and cannot form a continuous and complete three-dimensional borehole wall geometric model. Therefore, it is difficult to detect complex geometric anomalies such as local enlargement or asymmetric collapse.
[0006] Furthermore, existing process monitoring technologies lack direct means to assess the physical stability of the borehole wall. The geometric integrity of the borehole is closely related to the physical and mechanical state of its surrounding rock or soil, but current technologies cannot quantify key physical parameters such as borehole wall compaction and the degree of microcrack development, thus failing to predict the probability of geological risks such as borehole collapse from a mechanistic perspective. The limited and incomplete nature of information prevents current technologies from establishing an effective closed loop from multi-dimensional state perception to proactive risk warning, hindering further improvements in the quality control and safety assurance of pile foundation construction. Summary of the Invention
[0007] To address the shortcomings of existing technologies, this invention provides an intelligent monitoring and early warning system for the quality of borehole formation in highway pile foundation construction. This system solves the problems of existing pile foundation borehole quality monitoring technologies, such as limited data dimensions, inability to perform real-time three-dimensional imaging, and lack of early warning capabilities for borehole wall instability risks.
[0008] To achieve the above objectives, the present invention provides the following technical solution: an intelligent monitoring and early warning system for the quality of borehole formation in highway pile foundation construction, comprising: A multimodal sensing module, deployed on the drilling equipment, is used to simultaneously acquire the original mixed signals inside the borehole, mud environment parameters, and drilling rig operating parameters during the hole formation process. The original mixed signals include echo signals from actively emitted acoustic pulses and passive acoustic-vibration coupling signals generated by the drilling rig's operation. The signal processing module is used to decouple the original mixed signal into an active detection acoustic signal and a passive acoustic-vibration coupling signal; The model reconstruction and evaluation module is used to reconstruct the three-dimensional geometric model of the borehole wall based on the active acoustic signal and the mud environment parameters, and to evaluate the physical stability of the borehole wall based on the passive acoustic-vibration coupling signal. The risk warning module is used to integrate the three-dimensional geometric model of the borehole wall, the physical stability of the borehole wall, and the drilling rig operating parameters to generate a three-dimensional risk probability map of the borehole and output warning information.
[0009] Preferably, the model reconstruction and evaluation module is specifically used for: Based on the mud environment parameters, a dynamic sound velocity field is constructed inside the borehole; Using the dynamic sound velocity field as input to the acoustic inverse scattering problem, and combining it with the actively detected acoustic signal, the three-dimensional geometric model of the hole wall is solved and reconstructed.
[0010] Preferably, the model reconstruction and evaluation module is further used for: The vibration generated by the drilling rig is regarded as the excitation source, and the acoustic response signal of the borehole wall to the excitation source is extracted from the passive acoustic-vibration coupling signal. By calculating the acoustic transfer function between the acoustic response signal and the reference vibration source signal, parameters characterizing the physical stability of the hole wall are obtained.
[0011] Preferred options also include: An adaptive detection control module is used to receive the three-dimensional risk probability map and, based on the high-risk areas identified by the three-dimensional risk probability map, control the multimodal sensing module to adjust its detection strategy of actively emitting acoustic pulses.
[0012] Preferably, when adjusting the detection strategy, the adaptive detection control module is specifically used to control multiple array elements of the distributed acoustic topology array sensor in the multimodal sensing module to perform cooperative beamforming, focusing the sound field energy on the high-risk area for refined detection.
[0013] Preferably, the multimodal sensing module includes: A distributed acoustic topology array sensor is used to emit the active acoustic pulses and acquire the raw mixed signal; A fluid dynamics-assisted real-time correction module is used to measure the mud environment parameters in situ, including the local flow velocity, density, and compressibility of the mud. The data interface unit is used to connect to the control system of the drilling rig, thereby obtaining the operating parameters of the drilling rig.
[0014] Preferably, the signal processing module employs an independent component analysis algorithm to separate the original mixed signal into the active detection acoustic signal and the passive acoustic-vibration coupling signal.
[0015] Preferably, the risk warning module uses a Bayesian network model to probabilistically fuse the three-dimensional geometric model of the borehole wall, the physical stability of the borehole wall, and the drilling rig operating parameters, thereby calculating the three-dimensional risk probability map.
[0016] Preferably, the risk warning module is used to spatially align the three-dimensional geometric model of the borehole wall and the parameters characterizing the physical stability of the borehole wall, and to synchronize them with the drilling rig operating parameters in time, thereby realizing the spatiotemporal fusion of multi-source information.
[0017] Preferred options also include: The visualization and decision support module is used to present the three-dimensional geometric model of the hole wall, the physical stability of the hole wall, and the three-dimensional risk probability map in a three-dimensional dynamic visualization, and to provide decision support suggestions based on the early warning information.
[0018] This invention provides an intelligent monitoring and early warning system for the drilling quality of highway pile foundation construction. It has the following beneficial effects: 1. This invention enables real-time, multi-dimensional, and comprehensive perception of borehole quality. By integrating a distributed acoustic topology array sensor and a fluid dynamics-assisted real-time correction module, the system can simultaneously acquire active acoustic signals and passive acoustic-vibration coupling signals, and perform corrections in conjunction with dynamic mud environment parameters. This allows the system to not only reconstruct the three-dimensional geometric model of the borehole wall, but also simultaneously assess its physical stability. Compared to existing technologies that can only perform delayed and discrete geometric parameter measurements, this invention provides a complete and dynamic view of the borehole wall geometry and physical state during construction, significantly improving information dimensionality and real-time performance.
[0019] 2. This invention achieves a leap from condition monitoring to risk warning, improving construction safety. The system innovatively sets up a risk warning module, which adopts a Bayesian network model to fuse multi-source information such as the reconstructed three-dimensional geometric model of the borehole wall, physical stability assessment parameters, and real-time drilling rig operating parameters. This fusion analysis can quantitatively assess the posterior probability of quality defects such as borehole collapse and diameter reduction occurring at specific locations, generating a three-dimensional risk probability map. Compared with the traditional method of relying on the operator's experience for indirect judgment, this invention provides a quantitative and objective risk warning mechanism.
[0020] 3. This invention constructs an intelligent closed-loop control system from risk identification to precise verification, improving the targeting and efficiency of monitoring. By setting an adaptive detection control module, the system can dynamically adjust the detection strategy of the multimodal sensing module according to the high-risk areas identified by the risk warning module. Specifically, by controlling the acoustic array to perform cooperative beamforming, the sound field energy is focused on the high-risk area for refined detection, thereby improving the signal-to-noise ratio and resolution of key areas. This closed-loop feedback mechanism enables the system to conduct in-depth investigation of potential problem points, effectively improving the accuracy of diagnosis and the efficiency of resource utilization. Attached Figure Description
[0021] Figure 1 This is a block diagram of the system functional modules of the present invention; Figure 2 This is a schematic diagram of the deployment of the multimodal sensing module of the present invention; Figure 3 This is a flowchart of the signal processing and model reconstruction process of the present invention; Figure 4 This is a flowchart of the risk assessment and adaptive detection closed-loop control of the present invention. Detailed Implementation
[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] See attached document Figure 1 The present invention provides an intelligent monitoring and early warning system for the quality of hole formation in highway pile foundation construction. In a specific embodiment, it may include: a multimodal perception module, a signal processing module, a model reconstruction and evaluation module, and a risk early warning module.
[0024] The multimodal sensing module, deployed on the drilling equipment, is designed to simultaneously acquire three types of data during the drilling process: the original mixed signal inside the borehole, mud environment parameters, and drilling rig operating parameters. The original mixed signal includes the echo signal of actively emitted acoustic pulses and the passive acoustic-vibration coupling signal generated by the drilling rig.
[0025] The signal processing module is connected to the multimodal sensing module via a data bus. Its function is to receive the raw mixed signal and decouple it into two independent signals: an active acoustic detection signal and a passive acoustic-vibration coupling signal. The output data of this module is transmitted to the model reconstruction and evaluation module.
[0026] The model reconstruction and evaluation module is connected to both the signal processing module and the multimodal sensing module. This module performs two parallel processing tasks: First, it receives active acoustic signals and mud environment parameters, and reconstructs the three-dimensional geometric model of the borehole wall based on these two types of input data; Secondly, it receives passive acoustic-vibration coupling signals and, based on this input data, assesses the physical stability of the borehole wall. The three-dimensional geometric model of the borehole wall generated by this module and the assessment results of the physical stability of the borehole wall are transmitted to the risk warning module.
[0027] The risk warning module is connected to both the model reconstruction and evaluation module and the multimodal perception module. Its function is to receive and fuse three types of data: the three-dimensional geometric model of the borehole wall, the physical stability assessment results of the borehole wall, and the drilling rig's operating parameters. Using a built-in fusion evaluation model, this module calculates and generates a three-dimensional risk probability map of the borehole and outputs warning information based on preset threshold conditions.
[0028] In the system's workflow, the data flow and control flow are as follows: The multimodal sensing module serves as the data source, continuously collecting on-site data. The original mixed signals are decoupled by the signal processing module and then sent to the model reconstruction and evaluation module along with the mud environment parameters for geometric reconstruction and physical state evaluation. The evaluation results are spatiotemporally aligned and data fused with the original drilling rig operating parameters in the risk warning module, and finally, the quantified risk probability is calculated and a warning is triggered.
[0029] In a preferred embodiment, the system further includes an adaptive detection module. This adaptive detection module is connected to the risk warning module and the multimodal perception module. It receives a three-dimensional risk probability map generated by the risk warning module and generates control commands based on high-risk areas identified in the map. These control commands are sent to the multimodal perception module to adjust its detection strategy for actively emitting acoustic pulses, such as adjusting the focusing direction or scanning accuracy of the sound field energy, thereby forming a closed-loop control circuit from risk identification to precise verification.
[0030] In a preferred embodiment, the system further includes a visualization and decision support module, which is connected to the model reconstruction and evaluation module and the risk warning module. This module receives the three-dimensional geometric model of the borehole wall, the physical stability evaluation results of the borehole wall, and the three-dimensional risk probability map. It then presents these multi-dimensional data in a three-dimensional dynamic visualization under a unified coordinate system. At the same time, based on the warning information output by the risk warning module, this module generates and displays corresponding decision support suggestions.
[0031] See attached document Figure 2 In one specific embodiment, the multimodal sensing module serves as the system's field data acquisition unit. It is fixedly integrated into the outer wall of the drill rod or casing of the drilling equipment and enters the pile hole synchronously with the drilling equipment. The multimodal sensing module includes a distributed acoustic topology array sensor, a fluid dynamics-assisted real-time correction module, and a data interface unit.
[0032] A distributed acoustic topology array sensor is a component that performs acoustic signal transmission and reception. In one embodiment, the sensor consists of multiple broadband piezoelectric transducer elements made of piezoelectric ceramic materials (e.g., lead zirconate titanate, PZT).
[0033] These array elements are integrated and arranged on a flexible circuit board substrate, and distributed in a preset spiral matrix form along the axial and circumferential directions of the drilling equipment, thus forming a cylindrical conformal array covering a 360-degree detection range. To adapt to the harsh working conditions inside the borehole, the entire sensor array surface is encapsulated in a layer of polyurethane material with acoustic impedance matching that of the drilling mud and wear resistance. This sensor performs a dual function: Firstly, under control commands, some or all of the array elements actively transmit encoded acoustic pulses (e.g., linear frequency modulated signals). Secondly, during the transmission interval, the entire array acts as a passive vibration pickup, continuously acquiring passive acoustic-vibration coupling signals generated by drilling operations, soil-rock interactions, and other behaviors.
[0034] The fluid dynamics-assisted real-time correction module is a component used for in-situ measurement of mud environment parameters within boreholes. This module is deployed alongside and near a distributed acoustic topology array sensor to ensure that the measured parameters accurately reflect the medium characteristics along the sound wave propagation path. In one embodiment, the fluid dynamics-assisted real-time correction module integrates a set of miniature sensors, including: an ultrasonic Doppler velocimeter for measuring local mud flow velocity, a tuning fork densitometer for measuring mud density, and other sensors for determining parameters required to determine mud compressibility. The fluid dynamics-assisted real-time correction module continuously outputs a high-precision set of mud environment parameters corresponding to specific spatial locations and time points at a preset sampling frequency.
[0035] The data interface unit is a component that enables data communication between this system and the drilling rig's main control system. In one embodiment, the data interface unit is a hardware unit with a physical communication interface and protocol parsing function. Its physical interface establishes a connection with the drilling rig's programmable logic controller (PLC) via a controller area network (CAN) bus or industrial Ethernet. The unit 130 periodically reads drilling rig operating parameters from specific data registers of the drilling rig PLC by executing a preset communication protocol (such as Modbus TCP / IP protocol). The drilling rig operating parameters specifically include drilling depth, drill rod speed, feed torque, and feed pressure. The acquired drilling rig operating parameters are timestamped and transmitted to the subsequent processing module along with other acquired data.
[0036] See attached document Figure 3 The signal processing module receives the raw mixed signal acquired by the multimodal sensing module. The original mixed signal is a multi-channel time-domain signal matrix, which contains the linear superposition of the echo signal of the actively emitted acoustic pulse and the passive acoustic vibration coupling signal generated by the drilling rig. The function of the signal processing module is to decouple the original mixed signal by applying blind source separation technology.
[0037] In one embodiment, an Independent Component Analysis (ICA) algorithm, such as the FastICA algorithm, is employed. The goal of this algorithm is to find an unmixing matrix. This makes the estimated value of the source signal... The following formula is used to calculate: ; in, It is a vector containing the raw signals acquired by multiple array elements. It is an estimated vector containing multiple independent source signals (e.g., active probe acoustic signals and passive acoustic-vibration coupling signals). It is the unmixing matrix.
[0038] The optimal unmixing matrix is obtained by maximizing the non-Gaussianity of the output signal through an iterative optimization algorithm. After separation, the signal in the output vector is identified based on the prior characteristics of the signal (such as high correlation with the active pulse emission time or spectral characteristics), thereby obtaining the active detection acoustic signal. and passive acoustic-vibration coupling signal Subsequently, the two signals were preprocessed with bandpass filtering and noise suppression to eliminate power frequency interference and random noise.
[0039] The model reconstruction and evaluation module receives decoupled signals from the signal processing module and mud environment parameters from the multimodal sensing module, and performs three-dimensional geometric model reconstruction of the borehole wall and evaluation of the physical stability of the borehole wall.
[0040] When performing the reconstruction of the three-dimensional geometric model of the borehole wall, the model reconstruction and evaluation module first uses discrete mud environment parameters (such as the density of the mud fluid phase) measured by the fluid dynamics-assisted real-time correction module. Compression ratio and volume fraction of solid particles By employing three-dimensional spatial interpolation algorithms (such as Kriging interpolation or inverse distance weighted interpolation), a continuous and dynamic three-dimensional sound velocity field covering the entire detection area is constructed. The calculation of this sound velocity field follows Wood's equations: ; in, and The calculation method is as described in the invention content section.
[0041] Subsequently, the model reconstruction and evaluation module constructs the aperture wall reconstruction problem as an acoustic inverse scattering problem and solves it based on the Distorted Wave Born Approximation (DWBA) theory. The calculated dynamic sound velocity field is then... As a non-uniform background medium, the physical property functions characterizing the geometric boundaries of the aperture walls are solved using actively probed acoustic signals. Scattered sound field Described by the following integral equation: ; in, The scattered sound field is obtained by acquiring and decoupling from a distributed acoustic topology array sensor. and These are the position vectors of the transmitting and receiving array elements for actively detecting acoustic pulses, respectively. Angular frequency, This refers to the borehole space area to be reconstructed. It is the physical property function to be solved, and its value reflects the change of the acoustic parameters of the medium relative to the background medium. The non-zero value region constitutes the geometric boundary of the pore wall.
[0042] Is In the described non-uniform background medium, by The sound source at that location The incident sound field generated at that location It is the Green's function in this non-homogeneous medium, representing the state from which the green function originates. arrive The sound wave propagation characteristics were determined by solving the integral equation using iterative optimization algorithms such as the conjugate gradient method, resulting in the material property function. optimal solution The spatial distribution of non-zero values constitutes the three-dimensional geometric model of the borehole wall. .
[0043] When performing a physical stability assessment of the borehole wall, the model reconstruction and assessment module utilizes passive acoustic-vibration coupling signals. First, the mechanical vibrations generated by drilling rig operations (e.g., drill bit rotation, drill rod vibration) and transmitted to the borehole wall via the drill rod and mud are considered as a known broadband excitation source. The Fourier spectrum of the reference vibration source signal is obtained either by placing an additional reference accelerometer near the drill bit or by establishing a model through analysis of drilling rig operating parameters. Simultaneously, time-frequency analysis (e.g., short-time Fourier transform) is performed on the passive acoustic-vibration coupling signal to extract the signals originating from specific locations on the borehole wall. The acoustic signal that responds to and radiates from the excitation source is calculated, and its spectrum is determined. Then, the model reconstruction and evaluation module calculates the acoustic-vibration transfer function. The formula for evaluating the physical and mechanical properties of the hole wall is as follows: ; in, Position of the hole wall At angular frequency The acoustic transfer function under these conditions, It is extracted from the passive acoustic-vibration coupling signal, corresponding to the position of the borehole wall. The Fourier spectrum of the response signal, It is the Fourier spectrum of the reference vibration source signal, through this transfer function A thorough analysis of its frequency domain characteristics, such as identifying the frequency position, peak amplitude, and quality factor (Q factor) of its resonance peaks, can quantify the characterization of the aperture wall at that location. The physical stability parameters, such as density, elastic modulus, degree of internal crack development, and structural damping, are spatially mapped and combined to form a three-dimensional geometric model of the borehole wall. Spatially aligned 3D hole wall physical stability map .
[0044] See attached document Figure 4 This process is mainly executed by the risk warning module and the adaptive detection module. The risk warning module receives the three-dimensional geometric model of the hole wall from the model reconstruction and evaluation module. and parameters characterizing the physical stability of the pore wall Simultaneously, the risk warning module receives drilling rig operating parameters transmitted from the data interface unit of the multimodal perception module. The risk warning module first performs spatiotemporal fusion processing of multi-source data.
[0045] Specifically, the three-dimensional geometric model of the hole wall and parameters characterizing the physical stability of the pore wall The data is mapped to a unified three-dimensional coordinate system and aligned spatially to ensure the correspondence between different data sources in spatial location. For example, grid-based interpolation or point cloud registration techniques can be used. Simultaneously, the fused geometric and physical stability data is integrated with drilling rig operating parameters. Timestamp synchronization is performed to ensure the consistency of all input information in the time dimension. Methods such as linear interpolation or nearest neighbor matching are used to unify data with different sampling frequencies into the same time series.
[0046] Subsequently, the risk warning module uses a Bayesian network model to probabilistically fuse these spatiotemporally fused multi-source information to calculate a three-dimensional risk probability map of the borehole. In one embodiment, the Bayesian network model is constructed as a directed acyclic graph containing the following types of nodes: geometric state nodes (e.g., borehole diameter deviation, borehole wall roughness, skewness, etc.), physical stability nodes (e.g., local density, fracture density, elastic modulus, etc.), drilling rig condition nodes (e.g., abnormal drilling torque, sudden change in feed rate, mud pump pressure fluctuation, etc.), and finally, risk event nodes (e.g., borehole collapse risk, diameter reduction risk, stuck drill risk).
[0047] The conditional dependencies between nodes are established through expert experience and historical construction data. The conditional probability table (CPT) of each node is trained and initialized through statistical analysis of a large number of historical success and failure cases or through domain expert knowledge. This Bayesian network model calculates risk events under given observational evidence through inference. Posterior probability of occurrence Its calculation follows Bayes' theorem: ; In practical calculations, its equivalent proportional form is usually used: ; in, Indicates given observational evidence (From the 3D geometric model of the hole wall) (From the physical stability assessment of the pore wall) and Risk events (based on drilling rig operating parameters) The posterior probability of occurrence; , , Each piece of observational evidence in a risk event The conditional probability of occurrence; Risk event The prior probability. The calculation results are presented in a three-dimensional risk probability map. The form in which each spatial location point is represented, In time Each comes with a quantified risk probability value.
[0048] When the risk probability value of a certain area reaches a preset threshold (for example, 0.75 is a moderate warning and 0.90 is a local warning), the risk warning module is triggered and outputs the corresponding warning information, which includes the risk type, risk location, risk level and recommended measures.
[0049] The adaptive detection module is connected to the risk warning module and the multimodal perception module. The adaptive detection module receives a three-dimensional borehole risk probability map generated by the risk warning module. The adaptive detection module performs real-time analysis of the risk probability map, identifying high-risk areas where the risk probability value exceeds a specific threshold. High-risk areas are determined by their spatial coordinate range (e.g., depth range, radial distance, and circumferential angle).
[0050] Based on the spatial location and risk level of the identified high-risk areas, the adaptive detection module generates control commands to dynamically adjust the detection strategy of the active acoustic pulse emission of the distributed acoustic topology array sensor 110 in the multimodal sensing module. The adjustments include the selection of detection modes, the focusing of sound field energy, and the improvement of local scanning accuracy.
[0051] Specifically, if no high-risk area is detected, the system will maintain the normal omnidirectional scanning mode; if a high-risk area is identified, the system will switch from the normal scanning mode to a localized refined detection mode. In this mode, the adaptive detection module calculates and adjusts the complex weights of each element in the distributed acoustic topology array sensor. This enables coordinated beamforming, precisely focusing acoustic energy onto high-risk areas. (Array beam pattern)
[0052] Determined by the following formula: ; in, For beamforming The complex weights of each array element can be expressed as: ,in For amplitude weighting coefficients, For phase delay; The total number of array elements participating in collaborative work; The imaginary unit; It is the direction of the angle. The direction of the wave vector, its magnitude From angular frequency Local sound velocity in mud Decide; It is the first The precise position vectors of each array element in the array coordinate system are obtained by dynamically loading new complex weights. The sensor can accurately direct and focus the sound field energy towards high-risk areas in three-dimensional space, thereby effectively improving the detection signal-to-noise ratio and spatial resolution of the area, obtaining more detailed acoustic data, providing data support for the subsequent risk warning module to conduct more refined risk assessment, and forming a closed loop of monitoring, early warning, adjustment and re-monitoring.
[0053] The visualization and decision support module provided by this invention is connected to the model reconstruction and evaluation module and the risk warning module, serving as a human-computer interaction interface and information output terminal. The functions of the visualization and decision support module include providing three-dimensional dynamic visualization of multi-dimensional data and offering decision support suggestions. This visualization and decision support module receives a three-dimensional geometric model of the hole wall from the model reconstruction and evaluation module. and parameters characterizing the physical stability of the pore wall And the three-dimensional risk probability map of the borehole from the risk warning module. And early warning information.
[0054] In terms of 3D dynamic visualization, this visualization and decision support module will display the 3D geometric model of the hole wall. As a basic graphical element, such as a 3D mesh model or point cloud rendered on the user interface, parameters characterizing the physical stability of the hole wall. The texture or color gradient mapped to the surface of the geometric model, for example, using a color gradient from blue (high stability) to red (low stability) to represent the local density or degree of fracture development of the pore walls, in a three-dimensional risk probability map. The data is then overlaid on the geometric model using transparent volume rendering or isosurface extraction. Areas with higher risk probabilities are displayed with darker colors or higher opacity to visually indicate the potential risk distribution. Users can rotate, scale, and translate the model using interface controls, as well as slice along the borehole axis. They can also click on specific areas to query detailed data.
[0055] Regarding the tiered early warning mechanism, the visualization and decision support module uses multi-level visual and auditory prompts based on the early warning information output by the risk early warning module.
[0056] For example, when drilling three-dimensional risk probability map When the risk probability value of a certain area reaches the first threshold (e.g., 0.65), the system highlights the area in yellow on the visualization interface and displays a "Attention" text prompt; when the risk probability value reaches the second threshold (e.g., 0.80), the interface highlights it in flashing orange, accompanied by a brief warning sound, and displays a "Warning" text prompt; when the risk probability value reaches the third threshold (e.g., 0.95), the interface highlights it in rapidly flashing red, plays a continuous alarm sound, and displays a "Danger" text prompt.
[0057] These visual and auditory feedbacks are used to quickly attract the operator's attention. Regarding decision support suggestions, the visualization and decision support module generates and displays structured decision support suggestions based on the specific risk type, location, and level identified by the risk warning module. These suggestions are based on preset construction specifications, geological databases, and engineering experience rule bases.
[0058] For example, when the system identifies a high risk of borehole diameter reduction at a depth of X meters, decision support suggestions may include: reducing the drilling feed rate to Y meters per hour at depth X meters; considering adjusting the mud formula to increase the mud density to Z kilograms per cubic meter; or performing localized borehole enlargement. These suggestions are displayed clearly in text on the interface, providing operators with immediate, specific, and actionable directions for adjusting construction parameters or countermeasures, thereby assisting operators in making quick and informed decisions and effectively mitigating risks to the quality of pile foundation borehole formation.
Claims
1. An intelligent monitoring and early warning system for the quality of hole formation in highway pile foundation construction, characterized in that, include: A multimodal sensing module, deployed on the drilling equipment, is used to simultaneously acquire the original mixed signals inside the borehole, mud environment parameters, and drilling rig operating parameters during the hole formation process. The original mixed signals include echo signals from actively emitted acoustic pulses and passive acoustic-vibration coupling signals generated by the drilling rig's operation. The signal processing module is used to decouple the original mixed signal into an active detection acoustic signal and a passive acoustic-vibration coupling signal; The model reconstruction and evaluation module is used to reconstruct the three-dimensional geometric model of the borehole wall based on the active acoustic signal and the mud environment parameters, and to evaluate the physical stability of the borehole wall based on the passive acoustic-vibration coupling signal. The risk warning module is used to integrate the three-dimensional geometric model of the borehole wall, the physical stability of the borehole wall, and the drilling rig operating parameters to generate a three-dimensional risk probability map of the borehole and output warning information.
2. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, The model reconstruction and evaluation module is specifically used for: Based on the mud environment parameters, a dynamic sound velocity field is constructed inside the borehole; Using the dynamic sound velocity field as input to the acoustic inverse scattering problem, and combining it with the actively detected acoustic signal, the three-dimensional geometric model of the hole wall is solved and reconstructed.
3. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, The model reconstruction and evaluation module is also specifically used for: The vibration generated by the drilling rig is regarded as the excitation source, and the acoustic response signal of the borehole wall to the excitation source is extracted from the passive acoustic-vibration coupling signal. By calculating the acoustic transfer function between the acoustic response signal and the reference vibration source signal, parameters characterizing the physical stability of the hole wall are obtained.
4. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, Also includes: An adaptive detection control module is used to receive the three-dimensional risk probability map and, based on the high-risk areas identified by the three-dimensional risk probability map, control the multimodal sensing module to adjust the detection strategy of actively emitting acoustic pulses.
5. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 4, characterized in that, When adjusting the detection strategy, the adaptive detection control module specifically controls multiple array elements of the distributed acoustic topology array sensor in the multimodal sensing module to perform cooperative beamforming, focusing the sound field energy on the high-risk area for refined detection.
6. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, The multimodal sensing module includes: A distributed acoustic topology array sensor is used to emit the active acoustic pulses and acquire the raw mixed signal; A fluid dynamics-assisted real-time correction module is used to measure the mud environment parameters in situ, including the local flow velocity, density, and compressibility of the mud. The data interface unit is used to connect to the control system of the drilling rig, thereby obtaining the operating parameters of the drilling rig.
7. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, The signal processing module uses an independent component analysis algorithm to separate the original mixed signal into the active detection acoustic signal and the passive acoustic-vibration coupling signal.
8. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, The risk warning module uses a Bayesian network model to probabilistically fuse the three-dimensional geometric model of the borehole wall, the physical stability of the borehole wall, and the drilling rig operating parameters, thereby calculating the three-dimensional risk probability map.
9. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 2, characterized in that, The risk warning module is used to spatially align the three-dimensional geometric model of the borehole wall and the parameters characterizing the physical stability of the borehole wall, and to synchronize them with the drilling rig operating parameters in time, thereby realizing the spatiotemporal fusion of multi-source information.
10. The intelligent monitoring and early warning system for the quality of pile foundation drilling in highway construction according to claim 1, characterized in that, Also includes: The visualization and decision support module is used to present the three-dimensional geometric model of the hole wall, the physical stability of the hole wall, and the three-dimensional risk probability map in a three-dimensional dynamic visualization, and to provide decision support suggestions based on the early warning information.