A smart management and control system for pile foundation construction that integrates karst cave detection and mud regulation

Abstract

This invention relates to the field of pile foundation construction technology and discloses an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud control. The system includes a multi-source data synchronous acquisition module, a fluid resistance adaptive calibration module, a parasitic torque decoupling module, a formation energy fingerprint identification module, and a fluid-structure interaction decision module. The acquisition module constructs a full-dimensional state vector; the calibration module calculates the working condition correction coefficient; the decoupling module removes the fluid parasitic torque to output the net cutting torque; the identification module calculates the corrected mechanical specific energy and the normalized change gradient; and the decision module calculates the formation equivalent hydraulic impedance, integrates mechanical and hydraulic characteristics to generate a karst cave critical breakdown early warning index, and outputs pressure regulation and flow compensation commands when the index exceeds the limit or the impedance is abnormal. This invention achieves proactive early warning of karst cave penetration risk and intelligent control of mud pressure by decoupling fluid resistance interference and coupling dual-physics field characteristics.

Description

Technical Field

[0001] This invention relates to the field of pile foundation construction technology, specifically to an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud regulation. Background Technology

[0002] In deep rotary drilling operations, the rotation of the drill rod agitates the drilling mud, generating fluid viscous resistance. The torque values ​​collected by the ground monitoring system actually include the fracturing reaction torque generated by cutting the rock and the parasitic torque of the mud acting on the drill rod surface. As the drilling depth increases, the mud column pressure and mud viscosity increase, and the proportion of parasitic torque in the total torque also increases. If the total torque values ​​collected by the ground monitoring system are directly used to infer the lithology of the formation, the parasitic torque will mask the true characteristics of the fracturing reaction torque changes in the formation, leading to a lag or misjudgment in the identification of the soft and hard interfaces of the formation, and failing to truly reflect the interaction state between the drill bit and the rock and soil mass.

[0003] Existing pile foundation construction monitoring methods typically analyze mechanical drilling parameters or fluid circulation parameters independently. Monitoring mechanical parameters alone is insufficient to distinguish between weak interlayers and early-stage karst development zones, while monitoring fluid parameters alone usually relies on significant anomalies in slurry return flow to determine leakage. Due to the lack of a coupled analysis mechanism between mechanical and fluid physical fields, the monitoring system cannot quantify the formation's resistance to high-pressure mud intrusion, nor can it utilize the correlation between rock strength decay trends and formation permeability abrupt changes for cross-validation. Furthermore, it struggles to identify critical breakdown precursors before the drill bit penetrates the karst cave roof.

[0004] Due to the lack of a feedforward early warning mechanism for the cavern breakthrough process, the control system cannot establish pressure balance within the borehole before the leakage channel forms. When sudden mud leakage occurs, the control system lacks a flow compensation algorithm based on formation impedance characteristics, making it difficult to calculate the mud supply rate required to maintain borehole stability. This leads to lag in mud pump regulation, resulting in the risk of borehole collapse or excessive mud injection. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud regulation. This system solves the problems in existing pile foundation construction where, as drilling depth increases, the interference of fluid viscous resistance on mechanical torque leads to deviations in stratum identification, making it difficult to accurately detect the location of karst caves. At the same time, monitoring of single mechanical parameters cannot reflect the permeability characteristics of stratum fluids, resulting in the failure of mud pressure balance at the moment of karst cave penetration, which can cause drill bit drops, borehole collapse, or mud leakage accidents.

[0006] This invention provides an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud control. The system includes a multi-source data synchronous acquisition module, a fluid resistance adaptive calibration module, a parasitic torque decoupling module, a formation energy fingerprint identification module, and a fluid-structure interaction decision module.

[0007] Furthermore, the multi-source data synchronous acquisition module is used to read data and construct a full-dimensional state vector encapsulating mechanical and fluid components. This full-dimensional state vector encapsulates both mechanical and fluid components. The mechanical components include the total torque measured by the sensors, the drilling pressure applied by the drill bit, the drill pipe rotation speed, the mechanical drilling speed, and the current borehole depth. The fluid components include mud injection flow rate, return flow rate, mud pump outlet pressure, mud plastic viscosity, and mud yield value. The multi-source data synchronous acquisition module uses a unified reference clock source and maps all parameter values ​​to the same time axis coordinate point by downsampling and averaging high-frequency data and interpolating and fitting low-frequency data.

[0008] Furthermore, the fluid resistance adaptive calibration module uses mechanical components to identify the hover calibration state. Combining the mechanical friction constant with the theoretical fluid resistance torque calculated using fluid components and borehole geometry parameters, it calculates the operating condition correction coefficient. The fluid resistance adaptive calibration module has preset drill pressure dead zone thresholds and drill speed dead zone thresholds. When the drill pressure applied by the drill bit is less than the drill pressure dead zone threshold, the absolute value of the mechanical drilling speed is less than the drill speed dead zone threshold, and the drill pipe rotation speed is greater than zero, the fluid resistance adaptive calibration module determines that the drilling rig has entered the hover calibration state through logical judgment.

[0009] Furthermore, the fluid resistance adaptive calibration module triggers the calculation process of the working condition correction coefficient after the duration of the hover calibration state exceeds the calibration time threshold. Utilizing the principle of a concentric cylinder rotational viscometer, the module calculates the effective annular shear rate using the drill pipe outer radius, borehole radius, and drill pipe rotation speed, and combines this with the mud plastic viscosity and mud yield value to calculate the theoretical fluid resistance torque. The module obtains the net fluid resistance torque by subtracting the mechanical friction constant from the total torque measured by the sensor in the hover calibration state, and calculates the ratio of the net fluid resistance torque to the theoretical fluid resistance torque, thus obtaining the working condition correction coefficient characterizing the combined effects of borehole geometric irregularities and mechanical transmission efficiency.

[0010] Furthermore, the parasitic torque decoupling module reconstructs the real-time fluid parasitic torque using a working condition correction coefficient, separating the real-time fluid parasitic torque from the total torque of the full-dimensional state vector to output the net cutting torque. The parasitic torque decoupling module is used to calculate the current theoretical fluid resistance torque in real time when the drilling rig is in mechanical drilling mode. It then multiplies the current theoretical fluid resistance torque by the working condition correction coefficient to obtain the corrected fluid resistance component, and adds the fluid resistance component to the mechanical friction constant to reconstruct the real-time fluid parasitic torque.

[0011] Furthermore, the parasitic torque decoupling module calculates the difference between the total torque measured by the sensor and the real-time fluid parasitic torque, and performs a non-negative logic judgment on the difference: if the difference is less than zero, it is forcibly set to zero; if the difference is greater than or equal to zero, the difference is retained as the net cutting torque.

[0012] Furthermore, the formation energy fingerprinting module utilizes net cutting torque to calculate the corrected mechanical specific energy and its normalized gradient. Replacing traditional surface measurement torque with net cutting torque, the module combines the drill bit applied drilling pressure, drill pipe rotation speed, mechanical drilling speed, and drill bit radius to calculate the corrected mechanical specific energy. Employing the finite difference method, the module obtains the normalized gradient characterizing the rate of change of formation strength per unit depth by dividing the difference in corrected mechanical specific energy between adjacent sampling times by the corresponding difference in borehole depth.

[0013] Furthermore, the fluid-structure interaction decision module is used to calculate the equivalent hydraulic impedance of the formation, and integrates the normalized variation gradient with the equivalent hydraulic impedance of the formation to calculate the critical breakdown warning index of the karst cave. When the critical breakdown warning index of the karst cave exceeds the danger trigger threshold and the equivalent hydraulic impedance of the formation is lower than the critical impedance value, feedforward pressure regulation command and flow compensation command are output respectively to coordinate the control of the mud state.

[0014] Furthermore, the fluid-structure interaction (FSI) decision-making module utilizes Darcy's law and fluid network theory to calculate the formation's equivalent hydraulic impedance by summing the absolute value of the difference between the mud pump outlet pressure and the mud injection flow rate and the return flow rate, along with the anti-singularity regularization constant. The FSI quantifies the current formation's resistance to high-pressure mud intrusion using this equivalent hydraulic impedance and identifies the attenuation trend of rock strength in the vertical depth. The FSI has preset danger trigger thresholds, mechanical characteristic weighting coefficients, and hydraulic characteristic weighting coefficients. Through a weighted normalization algorithm, the FSI generates a critical breakdown warning index for karst caves by weighted summing the ratio of the normalized gradient to the reference value of the benchmark rock specific energy gradient, and the ratio of the reference value of the intact formation hydraulic impedance to the formation's equivalent hydraulic impedance.

[0015] Furthermore, when the critical breakdown warning index of the karst cave exceeds the danger trigger threshold, the fluid-structure interaction (FSI) decision module determines it as a precursor to critical breakdown. The FSI decision module calculates the compensation mud injection flow rate using the impedance matching principle when outputting flow compensation commands. By establishing a mapping relationship between flow gain and impedance attenuation ratio, and using the baseline injection flow rate before leakage as a basis, the module calculates the ratio of the complete formation hydraulic impedance reference value to the current equivalent formation hydraulic impedance using a logarithmic function as a gain factor. This is then combined with the flow regulation response coefficient to determine the compensation mud injection flow rate capable of offsetting the formation leakage channel's capacity.

[0016] Furthermore, the fluid-structure interaction (FSI) decision module triggers a preventative adjustment logic for the mud pump pressure via a feedforward pressure regulation command, establishing pressure balance before the drill bit actually penetrates the top of the karst cave. The FSI decision module sends feedforward pressure regulation and flow compensation commands to the mud pump frequency converter in the mud circulation pipeline, driving the mud pump drive unit to perform corresponding pumping output adjustments.

[0017] This invention provides an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud control. It has the following beneficial effects: 1. This invention utilizes the synergistic effect of a fluid resistance adaptive calibration module and a parasitic torque decoupling module to strip away the parasitic torque generated by mud viscous resistance in real time during drilling. This eliminates background noise interference from changes in borehole depth and mud properties on geological identification, enabling the net cutting torque to truly reflect the interaction between the drill bit and the rock and soil.

[0018] 2. This invention constructs a dual-physical-field coupling model of mechanical field and fluid field, and uses the precursory changes in fluid impedance to identify the top of the karst cave before the mechanical drilling parameters change abruptly. It also realizes the quantitative early warning of the risk of karst cave breakthrough through the karst cave critical breakdown early warning index.

[0019] 3. This invention establishes a closed-loop response mechanism from the detection end to the control end. When the critical breakdown precursor of the karst cave is detected, a feedforward pressure regulation command is output, and when leakage occurs, a flow compensation command is output. This transforms the traditional passive grouting into an active pressure balance defense, thus maintaining the mechanical stability of the borehole wall. Attached Figure Description

[0020] Figure 1 This is a system structure block diagram of the present invention; Figure 2 This is a flowchart of the system operation logic of the present invention; Figure 3 This is a comparison chart of the mechanical specific energy of the present invention as a function of depth; Figure 4 This is a diagram illustrating the dual-axis linkage process during the cave penetration of the present invention.

[0021] Among them, 10 is a multi-source data synchronous acquisition module; 20 is a fluid resistance adaptive calibration module; 30 is a parasitic torque decoupling module; 40 is a formation energy fingerprint identification module; and 50 is a fluid-structure interaction decision module. 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 This invention provides an intelligent management and control system for pile foundation construction that coordinates karst cave detection and mud regulation. The system is developed based on a modular collaborative management and control platform architecture and adopts a unified data interface specification and control protocol to achieve interconnection and interoperability between heterogeneous devices. Specifically, it includes: a multi-source data synchronous acquisition module 10, a fluid resistance adaptive calibration module 20, a parasitic torque decoupling module 30, a formation energy fingerprint identification module 40, and a fluid-structure interaction decision module 50.

[0024] A smart control system for pile foundation construction that integrates karst cave detection and mud regulation constructs a physical model using preset borehole geometric parameters, achieves data synchronization between modules through full-dimensional state vectors, decouples multiple physical fields using intermediate process variables, and generates final coordinated control commands based on decision index variables.

[0025] The multi-source data synchronous acquisition module 10 is used to establish a data mapping channel between physical equipment and digital computing environment. The multi-source data synchronous acquisition module 10 synchronously reads drilling rig control bus data and mud circulation pipeline sensor data at a set frequency. The multi-source data synchronous acquisition module 10 constructs and outputs a full-dimensional state vector at all times. The full-dimensional state vector includes mechanical components and fluid components. The mechanical components include the total torque measured by the sensor, the drilling pressure applied by the drill bit, the drill rod rotation speed, the mechanical drilling speed, and the current drilling depth. The fluid components include mud injection flow rate, return flow rate, mud pump outlet pressure, mud plastic viscosity, and mud yield value.

[0026] The fluid resistance adaptive calibration module 20 is connected to the multi-source data synchronous acquisition module 10. The fluid resistance adaptive calibration module 20 is used to identify the hover calibration state during the drilling process and calculate the working condition correction coefficient. The fluid resistance adaptive calibration module 20 monitors the drilling pressure and mechanical drilling speed. When the drilling pressure is less than the drilling pressure dead zone threshold, the absolute value of the mechanical drilling speed is less than the drilling speed dead zone threshold, and the drill pipe rotation speed is greater than zero, the fluid resistance adaptive calibration module 20 determines that the drilling rig has entered the hover calibration state.

[0027] The fluid resistance adaptive calibration module 20 pre-stores the inherent mechanical friction constant of the drilling rig under no-load conditions. This mechanical friction constant characterizes the mechanical loss torque of the drilling rig's power head when it is idling without load. Under the hover calibration state, the fluid resistance adaptive calibration module 20 calculates the theoretical fluid resistance torque through the Bingham fluid model, and combines the total torque measured by the sensor with the inherent mechanical friction constant of the drilling rig under no-load conditions. It first calculates the fluid action torque after deducting mechanical losses, and then uses the ratio calculation to invert the working condition correction coefficient.

[0028] The parasitic torque decoupling module 30 is connected to the fluid resistance adaptive calibration module 20 and the multi-source data synchronous acquisition module 10. The parasitic torque decoupling module 30 is used to eliminate the interference of fluid viscous resistance on lithology identification during drilling. The parasitic torque decoupling module 30 calls the stored working condition correction coefficient, combines the real-time mud rheological parameters and motion parameters to calculate the real-time fluid parasitic torque, and separates it from the total torque to output the net cutting torque.

[0029] The formation energy fingerprint identification module 40 is connected to the parasitic torque decoupling module 30. The formation energy fingerprint identification module 40 uses the net cutting torque to calculate and correct the mechanical specific energy. The formation energy fingerprint identification module 40 further calculates the normalized change gradient of the corrected mechanical specific energy with respect to time. The normalized change gradient characterizes the rate of change of formation strength.

[0030] The fluid-structure interaction decision module 50 is connected to the formation energy fingerprint identification module 40 and the multi-source data synchronous acquisition module 10. The fluid-structure interaction decision module 50 is used to construct a formation hydraulic impedance model and generate coordinated control commands. The fluid-structure interaction decision module 50 calculates the formation equivalent hydraulic impedance through the dynamic relationship between pump pressure and flow rate. The fluid-structure interaction decision module 50 is pre-set with a danger trigger threshold, a critical impedance value, a reference value for the specific energy gradient of the benchmark rock, a reference value for the hydraulic impedance of the intact stratum, and weighting coefficients for mechanical and hydraulic characteristics. The fluid-structure interaction decision module 50 logically integrates the normalized gradient and the equivalent hydraulic impedance of the stratum to calculate the critical breakdown warning index of the karst cave. When the critical breakdown warning index of the karst cave exceeds the danger trigger threshold, the fluid-structure interaction decision module 50 determines that it is a precursor to critical breakdown of the karst cave and outputs a feedforward pressure regulation command. When the equivalent hydraulic impedance of the stratum is lower than the critical impedance value, the fluid-structure interaction decision module 50 determines that the leakage channel is connected and outputs a flow compensation command.

[0031] The fluid-structure interaction (FSI) decision module 50 integrates intelligent decision-making algorithms and a library of solutions for different types of karst caves. Based on the amplitude characteristics of the normalized gradient change and the attenuation rate of the formation hydraulic impedance, the FSI decision module 50 classifies the current karst cave conditions into different geological types, including small, bead-like karst caves, filled karst caves, and large cavity karst caves. The FSI decision module 50 establishes a real-time linkage mechanism between detection data and mud control parameters, automatically matching the optimal mud treatment scheme according to the identified karst cave type.

[0032] See attached document Figure 2 The automatic matching process of mud schemes relies on the intelligent decision-making algorithm library integrated within the fluid-structure interaction decision module 50. The intelligent decision-making algorithm library includes a fuzzy logic-based cave classifier and a rule-based reasoning mud formula generator. When the fluid-structure interaction decision module 50 generates mud property control instructions, it does not directly drive the hardware. Instead, it encapsulates the target viscosity value and target sand content into a general control message through the standard control protocol of the collaborative management and control platform and sends it to the local controller of the mud preparation unit, thereby realizing the decoupling and collaboration between upper-level decision-making and lower-level execution.

[0033] When a large cavity cave is identified, the fluid-structure interaction decision module 50 generates a mud property control command while outputting a flow compensation command and sends it to the mud preparation unit to drive it to increase the viscosity of the newly prepared mud and reduce the sand content. The network structure of the high-viscosity mud is used to enhance the ability to carry rock cuttings and the effect of film attachment on the borehole wall. When leakage is detected as a minor crack, the instructions are adjusted to adapt to a low-viscosity, high-sealing mud solution. In this way, the fluid-structure interaction decision module 50 achieves dynamic adaptation of mud viscosity and sand content driven by detection results, improving the adaptability of construction to complex geological conditions and resource utilization. To enable collaborative operation among the modules, this invention predefines several sets of key parameters, namely, borehole geometric parameters, full-dimensional state vector, intermediate process variables, and decision index variables.

[0034] The drilling geometry parameters include the drill rod outer radius, the borehole radius, the drill bit radius, the drill rod length on the drilling machine, and the drill bit cross-sectional area. The drill rod outer radius is used to calculate the boundary conditions of the fluid flow in the annulus, the drill bit radius is used to determine the length of the cutting torque arm, the borehole radius is used to determine the outer boundary of the annulus flow channel, and the drill bit cross-sectional area is used to calculate the mechanical crushing volume and unit energy consumption.

[0035] The full-dimensional state vector is generated by the multi-source data synchronous acquisition module 10 at each discrete time step. The mechanical components of the full-dimensional state vector are the total torque measured by the sensor, the drilling pressure applied by the drill bit, the drill pipe rotation speed, the mechanical drilling speed and the current drilling depth. The fluid components are the mud injection flow rate, the return flow rate, the mud pump outlet pressure, the mud plastic viscosity and the mud yield value. The multi-source data synchronous acquisition module 10 is also connected to a distance sensor for measuring the remaining length on the drill pipe machine and establishes a communication connection with the mud pump frequency converter of the mud circulation pipeline.

[0036] Intermediate process variables are generated and transmitted by the system's calculation module. These intermediate process variables include theoretical fluid resistance torque, working condition correction coefficient, real-time fluid parasitic torque, and net cutting torque. The theoretical fluid resistance torque represents the theoretical viscous resistance experienced by the drill pipe rotation in an ideal concentric circular flow field. The working condition correction coefficient represents the deviation ratio of the actual drilling environment from the ideal model. The real-time fluid parasitic torque represents the fluid resistance component after calibration by the working condition correction coefficient. The net cutting torque represents the rock breaking reaction torque after eliminating the influence of fluid resistance.

[0037] The decision index variables are used by the fluid-structure interaction decision module 50 to determine the geological state. The decision index variables include the corrected mechanical specific energy, the normalized variation gradient, and the formation equivalent hydraulic impedance. The corrected mechanical specific energy characterizes the unit volume breaking work calculated by the net cutting torque. The normalized variation gradient characterizes the relative change rate of the corrected mechanical specific energy over time. The formation equivalent hydraulic impedance characterizes the fluid sealing performance of the borehole formation structure.

[0038] The multi-source data synchronous acquisition module 10 establishes a communication connection with the vehicle-mounted control unit of the pile foundation construction drilling rig through the industrial fieldbus interface. The multi-source data synchronous acquisition module 10 periodically reads the register values ​​of the vehicle-mounted control unit according to a preset time step. The preset time step is set to a value that satisfies the Nyquist sampling theorem to ensure the capture of mechanical transient changes.

[0039] The multi-source data synchronous acquisition module 10 acquires the total torque measured by the sensor. The total torque measured by the sensor is directly measured by the torque sensor installed on the output shaft of the drill rig power head, or it is calculated by the pressure difference between the inlet and outlet of the hydraulic motor of the power head through the torque coefficient. The total torque measured by the sensor represents the total torque required for the drill pipe to drive the drill bit to break rocks and overcome fluid resistance.

[0040] The multi-source data synchronous acquisition module 10 acquires the drilling pressure applied by the drill bit. The drilling pressure applied by the drill bit is obtained by converting the pressure sensor data of the drilling rig pressurizing cylinder or pressurizing winch mechanism through the mechanical balance equation. The drilling pressure applied by the drill bit represents the axial breaking force applied to the rock surface at the bottom of the hole.

[0041] The multi-source data synchronous acquisition module 10 acquires the drill pipe rotation speed, which is obtained by converting the pulse signal frequency acquired by the rotary encoder or Hall sensor installed on the power head spindle. The drill pipe rotation speed represents the cutting linear velocity reference of the drill bit.

[0042] The multi-source data synchronous acquisition module 10 obtains the current drilling depth. The current drilling depth is calculated by combining the wire rope lowering length recorded by the depth encoder of the main winch with the pre-recorded or obtained drill rod remaining length on the machine through the distance sensor. The current drilling depth represents the vertical distance of the bottom face of the drill bit relative to the ground surface reference plane.

[0043] The multi-source data synchronous acquisition module 10 calculates the mechanical drilling speed using time series data of the current borehole depth, that is, by dividing the depth difference between adjacent sampling times by the sampling time step.

[0044] The multi-source data synchronous acquisition module 10 is connected to an electromagnetic flowmeter installed on the outlet pipeline of the mud pump. The multi-source data synchronous acquisition module 10 obtains the mud injection flow rate through the electromagnetic flowmeter. The mud injection flow rate represents the volume of fluid pumped into the borehole per unit time.

[0045] The multi-source data synchronous acquisition module 10 is connected to a flow monitoring device installed at the inlet of the borehole return slurry trough or mud pit. The multi-source data synchronous acquisition module 10 obtains the return slurry flow rate through the flow monitoring device. The flow monitoring device adopts an ultrasonic open channel flow meter or a volumetric flow rate calculation unit based on the liquid level change rate. The return slurry flow rate characterizes the volume of fluid returning from the borehole annulus to the surface per unit time.

[0046] The multi-source data synchronous acquisition module 10 is connected to the pressure transmitter installed at the high-pressure manifold. The multi-source data synchronous acquisition module 10 obtains the mud pump outlet pressure through the pressure transmitter. The mud pump outlet pressure represents the driving head required to overcome the total hydraulic resistance of the circulation pipeline.

[0047] The multi-source data synchronous acquisition module 10 is connected to an online industrial viscometer in series in the mud circulation pipeline. The multi-source data synchronous acquisition module 10 analyzes the rheological parameters of the mud in real time through the online industrial viscometer. The online industrial viscometer fits the mud sampling data through the preset Bingham fluid constitutive equation, analyzes and outputs the plastic viscosity and yield value of the mud.

[0048] The plastic viscosity of mud represents the motion resistance caused by friction between solid particles and fluid interlayer friction during mud flow. The mud yield value represents the minimum shear stress required for mud to start flowing. The multi-source data synchronous acquisition module 10 transmits the plastic viscosity and mud yield value to the fluid resistance adaptive calibration module 20. The fluid resistance adaptive calibration module 20 uses the plastic viscosity and mud yield value to construct a fluid viscous resistance model.

[0049] In addition, based on the unified protocol of the modular collaborative management and control platform, the multi-source data synchronous acquisition module 10 also establishes bidirectional communication with the mud preparation unit on site through the industrial Ethernet interface. The multi-source data synchronous acquisition module 10 reads the raw material inventory status, mixer operating power and sand content monitoring data of the mud preparation unit in real time, and incorporates them as auxiliary decision variables into the full-dimensional state vector, thereby opening up the underlying data link from the detection end to the material preparation end.

[0050] In this embodiment, to ensure strict alignment of heterogeneous data in the time domain, the multi-source data synchronization acquisition module 10 is equipped with a high-precision unified reference clock source. The read drilling rig control bus data is defined as mechanical drilling parameter data packets, and the mud circulation pipeline sensor data is defined as fluid circulation parameter data packets. The unified clock source gives each received data packet a precise timestamp. Considering that the mechanical drilling parameters and fluid circulation parameters come from different hardware interfaces and have different sampling frequencies, the multi-source data synchronization acquisition module 10 uses a linear interpolation algorithm based on timestamps or a zero-order hold strategy to perform resampling processing. Based on a set time step, the high-frequency data is downsampled and averaged, and the low-frequency data is interpolated and fitted, thereby mapping all discrete parameter values ​​to the same time axis coordinate point and constructing a time-synchronized full-dimensional state vector.

[0051] The multi-source data synchronous acquisition module 10 encapsulates the time-aligned parameter values ​​into a full-dimensional state vector. The full-dimensional state vector serves as the standard input format for subsequent processing of the overall architecture. Internally, it encapsulates the mechanical components (including total torque, drilling pressure, rotational speed, drilling speed, and hole depth) and fluid components (including injection flow rate, return flow rate, pump pressure, plastic viscosity, and yield value) at each moment in a preset order. The multi-source data synchronous acquisition module 10 outputs this full-dimensional state vector to the fluid resistance adaptive calibration module 20 and the fluid-structure interaction decision module 50.

[0052] The multi-source data synchronous acquisition module 10 eliminates the data transmission delay differences between different sensors by constructing a full-dimensional state vector. The multi-source data synchronous acquisition module 10 ensures the simultaneity of mechanical action response and fluid pressure response in logical judgment.

[0053] The fluid resistance adaptive calibration module 20 extracts the drill pressure applied by the drill bit, the mechanical drilling speed, and the drill pipe rotation speed from the full-dimensional state vector. The fluid resistance adaptive calibration module 20 has built-in drill pressure dead zone threshold and drill speed dead zone threshold. The drill pressure dead zone threshold is set to a small positive value close to zero to tolerate sensor zero-point drift and drill string suspension fluctuation. The drill speed dead zone threshold is set to the upper limit of speed that can distinguish between drilling footage and hovering micro-motion.

[0054] The fluid resistance adaptive calibration module 20 monitors the drilling rig's operating parameters in real time and determines whether the physical basis for fluid resistance model calibration is available at the current moment through a logical judgment algorithm. In this embodiment, the drill pressure dead zone threshold is set to 1%-3% of the equipment's full-scale drill pressure to tolerate sensor zero-point drift and drill string suspension weight fluctuations; the drill speed dead zone threshold is set to 0.05m / min-0.2m / min. The drill speed dead zone threshold serves as the upper limit of the speed that distinguishes between effective drilling footage and hovering micro-motion. When it is detected that both the actual drill pressure and the mechanical drilling speed are lower than the above thresholds, and the drill rod rotation speed is maintained at the set speed, it is determined that the drilling rig has entered the hovering calibration state.

[0055] When the fluid resistance adaptive calibration module 20 detects that the above three conditions are met simultaneously, it starts a timer and continuously tracks the condition satisfaction status. If the above conditions are met for a longer period than the calibration time threshold, the fluid resistance adaptive calibration module 20 determines that the drilling rig has officially entered the hover calibration state. The calibration time threshold is used to filter out instantaneous operating condition fluctuations or operational jitters to ensure the stability of the collected data.

[0056] After determining that it has entered the hover calibration state, the fluid resistance adaptive calibration module 20 confirms that the drill bit is not in contact with the bottom rock of the hole or has not produced effective cutting, and the drill rod is in a continuous rotating state. The fluid resistance adaptive calibration module 20 confirms that the total torque measured by the sensor at this time is entirely composed of the mechanical friction of the drill rod and the viscous resistance of the mud fluid, and there is no rock breaking reaction torque. The fluid resistance adaptive calibration module 20 triggers the calculation process of the working condition correction coefficient.

[0057] The fluid resistance adaptive calibration module 20 calls pre-stored borehole geometry parameters and real-time data transmitted by the multi-source data synchronous acquisition module 10. Using the principle of a concentric cylinder rotational viscometer, the fluid resistance adaptive calibration module 20 approximates the annulus between the drill pipe and the wellbore as a laminar shear flow field. It then calculates the effective annulus shear rate using the drill pipe outer radius, borehole radius, and drill pipe rotation speed. The calculation formula is: ; in, The drill pipe rotation speed is expressed in r / min. The radius of the borehole is in meters. The outer radius of the drill pipe is in meters (m).

[0058] The fluid resistance adaptive calibration module 20, combined with the mud rheological parameters input from the multi-source data synchronous acquisition module 10 and the current borehole depth, calculates the theoretical fluid resistance torque experienced by the drill pipe rotation throughout the well section. The fluid resistance adaptive calibration module 20 assumes that the mud properties are uniformly distributed in the annulus and that the drill pipe and borehole maintain an ideal concentric state. The fluid resistance adaptive calibration module 20, combined with the mud rheological parameters input from the multi-source data synchronous acquisition module 10 and the current borehole depth, calculates the theoretical fluid resistance torque experienced by the drill pipe rotation throughout the well section.

[0059] ; in: This is the theoretical fluid drag torque, expressed in kN·m. The outer radius of the drill pipe is in meters (m). This represents the current borehole depth, in meters (m). This represents the mud yield strength, in Pa. The value represents the plastic viscosity of the mud, expressed in Pa·s. The effective shear rate of the annulus is expressed in s⁻¹.

[0060] The fluid resistance adaptive calibration module 20 quantifies the reference resistance torque value caused solely by fluid viscosity under the current depth, current rotation speed and current mud performance conditions through the above calculations. The theoretical fluid resistance torque provides a theoretical reference zero point for the subsequent calculation of the working condition correction coefficient.

[0061] Within a defined hovering calibration state time window, the fluid resistance adaptive calibration module 20 acquires the arithmetic mean of the total torque measured by the sensor, and simultaneously acquires the arithmetic mean of the theoretical fluid resistance torque within the hovering calibration state time window.

[0062] The fluid resistance adaptive calibration module 20 calculates the working condition correction coefficient, which characterizes the combined effects of borehole geometric irregularities and mechanical transmission efficiency. Specifically, the fluid resistance adaptive calibration module 20 calculates the arithmetic mean of the total torque measured by the sensor within the hover calibration state time window and the arithmetic mean of the theoretical fluid resistance torque. The net fluid resistance torque is obtained by subtracting the inherent mechanical friction constant of the drilling rig under no-load from the average value of the measured total torque. The ratio of the net fluid resistance torque to the average value of the theoretical fluid resistance torque is calculated and determined as the working condition correction coefficient.

[0063] The fluid resistance adaptive calibration module 20 stores the calculated working condition correction coefficients into the global parameter register. After the drilling rig exits the hover calibration state and resumes the mechanical drilling state, the fluid resistance adaptive calibration module 20 keeps the working condition correction coefficients unchanged until the next hover calibration state is triggered.

[0064] In mechanical drilling mode, the fluid resistance adaptive calibration module 20 calls the stored working condition correction coefficients in real time, and calculates the real-time fluid parasitic torque in combination with the current theoretical fluid resistance torque.

[0065] The fluid resistance adaptive calibration module 20 extracts the real-time fluid parasitic torque from the total torque measured by the sensor and calculates the net cutting torque for geological feature identification. The specific calculation logic is as follows: the product of the current working condition correction coefficient and the theoretical fluid resistance torque is superimposed with the inherent mechanical friction constant of the drilling rig under no-load as the real-time fluid parasitic torque, and then the difference between the total torque and the real-time fluid parasitic torque is calculated to obtain the net cutting torque.

[0066] The fluid resistance adaptive calibration module 20 dynamically eliminates the background resistance interference caused by the increase of drilling depth and changes in mud properties through the above-mentioned inversion and update mechanism. The fluid resistance adaptive calibration module 20 ensures that the net cutting torque only reflects the interaction intensity between the drill bit and the rock and soil. The fluid resistance adaptive calibration module 20 transmits the net cutting torque to the fluid-structure interaction decision module 50 as the core input variable for subsequent formation drillability analysis.

[0067] The parasitic torque decoupling module 30 establishes a data connection with the fluid resistance adaptive calibration module 20. The parasitic torque decoupling module 30 reads the working condition correction coefficient updated by the fluid resistance adaptive calibration module 20 in the most recent hovering calibration state. The parasitic torque decoupling module 30 synchronously receives the real-time drilling depth, drill rod rotation speed and mud rheological parameters transmitted by the multi-source data synchronous acquisition module 10.

[0068] The parasitic torque decoupling module 30 identifies that the drilling rig is currently in a mechanical drilling state. The parasitic torque decoupling module 30 reconstructs the instantaneous fluid parasitic torque under non-ideal working conditions through the current borehole geometry parameters and motion state parameters.

[0069] The instantaneous fluid parasitic torque characterizes the ineffective torque component consumed by mud viscosity effect and mechanical transmission loss at the current moment, current depth and current rotation speed. The parasitic torque decoupling module 30 obtains the corrected fluid resistance component by multiplying the latest working condition correction coefficient by the theoretical fluid resistance torque calculated by the borehole depth and mud rheological parameters at the current moment, and adds the fluid resistance component to the inherent mechanical friction constant of the drilling rig under no-load conditions, thereby reconstructing the instantaneous fluid parasitic torque under non-ideal working conditions.

[0070] The parasitic torque decoupling module 30 achieves real-time tracking of fluid resistance torque by introducing drilling depth and mud rheological parameters that change dynamically over time. The parasitic torque decoupling module 30 ensures that the instantaneous fluid parasitic torque can respond to environmental changes during drilling, rather than simply using the static value at the hovering moment. The parasitic torque decoupling module 30 uses the instantaneous fluid parasitic torque as a reference background noise for subsequent extraction of real rock breaking information from the total torque of the sensor.

[0071] The parasitic torque decoupling module 30 obtains the total torque measured by the sensor from the multi-source data synchronous acquisition module 10. The parasitic torque decoupling module 30 synchronously calls the instantaneous fluid parasitic torque calculated internally. The parasitic torque decoupling module 30 performs vector subtraction operation to subtract the instantaneous fluid parasitic torque from the total torque measured by the sensor.

[0072] The parasitic torque decoupling module 30 introduces non-negative physical constraint logic. The parasitic torque decoupling module 30 identifies negative value anomalies in the subtraction operation results. Since the rock breaking resistance cannot be negative physically, the parasitic torque decoupling module 30 forces the negative value results caused by sensor noise or equipment dynamic response lag to zero.

[0073] The parasitic torque decoupling module 30 calculates the final net cutting torque. The parasitic torque decoupling module 30 calculates the difference between the total torque measured by the sensor and the instantaneous fluid parasitic torque, and performs a non-negative logic judgment on the difference: if the difference is less than zero, it is forcibly set to zero; if the difference is greater than or equal to zero, the difference is retained as the net cutting torque.

[0074] The parasitic torque decoupling module 30 eliminates the interference of mud viscosity resistance on geological feature identification through the above calculations. The net cutting torque obtained by the parasitic torque decoupling module 30 directly reflects the energy consumption of breaking when the drill bit cutting teeth come into contact with the rock and soil medium at the bottom of the hole.

[0075] The parasitic torque decoupling module 30 performs sliding window smoothing on the net cutting torque of the continuous time series. The parasitic torque decoupling module 30 eliminates high-frequency random noise and retains low-frequency trend characteristics that reflect changes in lithology. The parasitic torque decoupling module 30 transmits the processed net cutting torque data stream to the fluid-structure interaction decision module 50 for intelligent identification of geological stratification interfaces.

[0076] The formation energy fingerprint identification module 40 is connected to the parasitic torque decoupling module 30. The formation energy fingerprint identification module 40 receives the net cutting torque after noise reduction and decoupling from the parasitic torque decoupling module 30. The formation energy fingerprint identification module 40 simultaneously obtains the current drill bit applied drilling pressure, drill rod rotation speed, mechanical drilling speed and preset drill bit radius from the multi-source data synchronous acquisition module 10.

[0077] The formation energy fingerprinting module 40 constructs a corrected mechanical specific energy calculation model that integrates net cutting torque through mechanical specific energy theory. The formation energy fingerprinting module 40 uses net cutting torque to replace the traditional ground measurement torque, eliminating the non-fracture energy consumed by drill pipe friction and borehole wall friction and mud viscosity resistance. The formation energy fingerprinting module 40 has a preset minimum calculation drilling speed limit value. When the formation is dense and the measured mechanical drilling speed approaches zero, the minimum calculation drilling speed limit value is used to replace the mechanical drilling speed in the formula to prevent numerical divergence. The formation energy fingerprinting module 40 calculates the corrected mechanical specific energy.

[0078] ; in: To correct for mechanical specific energy, the unit is kPa; The drilling pressure applied to the drill bit, measured in kN; The radius of the drill bit is in meters. The drill pipe rotation speed is expressed in r / min. Net cutting torque, in kN·m; This refers to the mechanical drilling speed or the minimum calculated drilling speed limit, expressed in m / h.

[0079] The corrected mechanical specific energy calculated by the formation energy fingerprinting module 40 represents the true mechanical energy required for the drill bit to break a unit volume of rock. The formation energy fingerprinting module 40 eliminates the interference of fluid resistance caused by the increase in drilling depth, ensuring that the corrected mechanical specific energy value is only related to the formation lithology strength.

[0080] The formation energy fingerprinting module 40 further constructs the gradient characteristics of the modified mechanical specific energy as the borehole depth changes. The formation energy fingerprinting module 40 uses the finite difference method to calculate the normalized change gradient by combining the data of the current sampling time and the previous sampling time. That is, the formation energy fingerprinting module 40 calculates the difference between the modified mechanical specific energy at the current time and the previous time, and divides it by the difference between the borehole depths corresponding to the two times, thereby obtaining the normalized change gradient that characterizes the rate of change of formation strength per unit depth.

[0081] The formation energy fingerprint identification module 40 uses the normalized change gradient to quantify the rate of change of formation drillability. The formation energy fingerprint identification module 40 transmits the calculation results to the fluid-structure interaction decision module 50. The fluid-structure interaction decision module 50 uses the corrected mechanical specific energy and its gradient as key characteristic indicators for identifying the interface of the cave roof and lithological abrupt changes.

[0082] The fluid-structure interaction decision module 50 acquires the mud pump outlet pressure, mud injection flow rate and return flow rate in real time from the multi-source data synchronous acquisition module 10. The fluid-structure interaction decision module 50 monitors the flow balance status of the mud circulation pipeline and identifies the leakage behavior of mud at the interface between the borehole annulus and the formation.

[0083] The fluid-structure interaction decision module 50 constructs a hydraulic impedance calculation model that reflects the formation permeability characteristics using Darcy's law and fluid network theory. The fluid-structure interaction decision module 50 considers the local formation at the depth of the drill bit as a fluid damping unit and calculates the formation hydraulic impedance.

[0084] ; in: This represents the hydraulic impedance of the formation, expressed in MPa·s / L. This represents the outlet pressure of the mud pump, in MPa. The mud injection flow rate is expressed in L / s. The return flow rate is expressed in L / s. To prevent singularity regularization, the value is taken as a very small positive real number, with the unit being L / s.

[0085] The fluid-structure interaction decision module 50 uses an anti-singularity regularization constant to prevent calculation errors with a denominator of zero under ideal conditions where the mud circulation is completely closed and there is no leakage. The fluid-structure interaction decision module 50 quantifies the current formation's resistance to high-pressure mud intrusion by using formation hydraulic impedance.

[0086] The fluid-structure interaction decision module 50 calculates high-value formation hydraulic impedance in intact bedrock strata. When encountering caves, fissures, or fracture zones, the fluid-structure interaction decision module 50 calculates formation hydraulic impedance that drops sharply. The fluid-structure interaction decision module 50 uses formation hydraulic impedance as a fluid physical field characteristic index independent of mechanical parameters. The fluid-structure interaction decision module 50 combines modified mechanical specific energy with formation hydraulic impedance to construct a multi-dimensional cave identification criterion.

[0087] The fluid-structure interaction decision module 50 calls the normalized gradient and formation hydraulic impedance as input variables in real time. The fluid-structure interaction decision module 50 identifies the sharp decline trend of rock strength in vertical depth and the abrupt change characteristics of formation permeability. The fluid-structure interaction decision module 50 determines that when the drill bit approaches the top of the cave, the rock mass develops microcracks due to stress release, resulting in a decrease in mechanical breaking specific work and a decrease in fluid impedance.

[0088] The fluid-structure interaction decision module 50 constructs a calculation model for the critical breakdown early warning index of karst caves through a dual physical field coupling mechanism of mechanical and hydraulic systems. The fluid-structure interaction decision module 50 integrates physical characteristics of different dimensions into a unified risk quantification index through a weighted normalization algorithm. The fluid-structure interaction decision module 50 calculates the critical breakdown early warning index of karst caves.

[0089] ; in: The critical breakthrough warning index for karst caves is dimensionless. The mechanical feature weighting coefficient is dimensionless. This is the normalized gradient, in kPa / m. The reference value for the specific energy gradient of the baseline rock is expressed in kPa / m. is the hydraulic characteristic weighting coefficient, which is dimensionless; The hydraulic impedance of the complete formation is a reference value, in MPa·s / L; This represents the hydraulic impedance of the formation, expressed in MPa·s / L.

[0090] The fluid-structure interaction decision module 50 compares the calculated critical breakthrough warning index of the karst cave with the preset danger triggering threshold. When the fluid-structure interaction decision module 50 detects that the critical breakthrough warning index of the karst cave exceeds the danger triggering threshold, it immediately determines that the current working condition is in a critical instability state before the karst cave breakthrough.

[0091] The fluid-structure interaction decision module 50 generates a feedforward pressure regulation command and transmits it to the mud pump frequency converter. The fluid-structure interaction decision module 50 triggers the preventive adjustment logic of the mud pump pressure through the feedforward pressure regulation command, and establishes pressure balance in advance before the drill bit actually penetrates the top of the karst cave to prevent drill bit falling or hole collapse accidents caused by instantaneous pressure loss. The fluid-structure interaction decision module 50 ensures that the construction control logic changes from passive response to active defense.

[0092] The fluid-structure interaction decision module 50 continuously monitors the formation hydraulic impedance output by the dynamic inversion process of formation hydraulic impedance. When the fluid-structure interaction decision module 50 determines that the formation hydraulic impedance has a step drop and the value is lower than the preset integrity threshold, it confirms that it is a mud loss event caused by cavern breakdown or fracture opening.

[0093] The fluid-structure interaction decision module 50 calculates the compensation mud injection flow rate required to maintain the pressure balance in the borehole by using the impedance matching principle. The fluid-structure interaction decision module 50 determines the mud supply rate that can offset the capacity of the formation leakage channel by establishing a mapping relationship between the flow rate gain and the impedance attenuation ratio.

[0094] ; in: The unit for compensating for mud injection flow rate is L / s; The baseline injection flow rate before leakage occurs, in L / s; The flow regulation response coefficient is dimensionless. The hydraulic impedance of the complete formation is a reference value, in MPa·s / L; This represents the current hydraulic impedance of the formation, expressed in MPa·s / L.

[0095] The fluid-structure interaction decision module 50 uses logarithmic function-based adjustment logic to prevent the calculation of command values ​​that exceed the physical limits of the mud pump under extremely low impedance conditions. The fluid-structure interaction decision module 50 converts the calculated compensated mud injection flow rate into control commands and sends them to the mud pump drive unit and its matching mud pump frequency converter, which are pre-installed in the mud circulation pipeline.

[0096] The fluid-structure interaction decision module 50 drives the mud pump drive unit to increase the pump output according to the compensated mud injection flow rate. The fluid-structure interaction decision module 50 uses the increased fluid dynamics to replenish the pressure loss caused by the reduction of formation hydraulic resistance, and maintains the support pressure of the borehole wall during the transition stage when the leakage channel is not completely sealed. The fluid-structure interaction decision module 50 ensures that the borehole maintains mechanical stability before the subsequent injection of plugging materials.

[0097] At the beginning of each global control cycle, the multi-source data synchronous acquisition module 10 triggers a synchronous sampling command. The multi-source data synchronous acquisition module 10 acquires the mechanical drilling parameters and mud fluid parameters uploaded by the sensor array in parallel. The multi-source data synchronous acquisition module 10 performs filtering and alignment processing on the raw signals to construct the state vector at the current moment. The multi-source data synchronous acquisition module 10 broadcasts the state vector to the fluid resistance adaptive calibration module 20 and the parasitic torque decoupling module 30 at the same time.

[0098] The fluid resistance adaptive calibration module 20 monitors the mechanical drilling speed in real time. When the fluid resistance adaptive calibration module 20 detects that the mechanical drilling speed is lower than the drilling speed dead zone threshold, it activates the hover calibration state logic. The fluid resistance adaptive calibration module 20 updates the working condition correction coefficient using the current sensor data. When the fluid resistance adaptive calibration module 20 detects that the mechanical drilling speed is higher than the drilling speed dead zone threshold, it maintains the value of the working condition correction coefficient stored at the previous moment unchanged and transmits the working condition correction coefficient to the parasitic torque decoupling module 30.

[0099] The parasitic torque decoupling module 30 receives the state vector and the working condition correction coefficient. The parasitic torque decoupling module 30 calculates the instantaneous fluid parasitic torque. The parasitic torque decoupling module 30 performs subtraction and non-negative constraint processing to solve for the net cutting torque. The parasitic torque decoupling module 30 transmits the net cutting torque as a pure rock crushing characteristic quantity to the fluid-structure interaction decision module 50.

[0100] The formation energy fingerprinting module 40 calculates and corrects the mechanical specific energy and its gradient through the net cutting torque and drilling motion parameters, and transmits it to the fluid-structure interaction decision module 50. The fluid-structure interaction decision module 50 simultaneously uses the mud flow rate and pressure parameters to invert the formation hydraulic impedance. The fluid-structure interaction decision module 50 combines the mechanical field and fluid field characteristics to calculate the critical breakdown early warning index of the karst cave.

[0101] The fluid-structure interaction (FSI) decision module 50 compares the critical breakdown warning index of the karst cave with the danger trigger threshold. When the critical breakdown warning index exceeds the limit, the FSI decision module 50 generates a feedforward pressure regulation command. When the formation hydraulic impedance experiences a step drop, the FSI decision module 50 generates a flow compensation command. Simultaneously, based on the karst cave type determined by the intelligent algorithm, the FSI decision module 50 generates mud property control commands (such as increasing viscosity or reducing sand content). The FSI decision module 50 sends the flow and pressure control commands to the mud pump drive unit and the mud property control commands to the mud preparation unit via the platform bus. The FSI decision module 50 then sends control commands to the mud pump drive unit, completing a single control cycle and entering the next sampling cycle.

[0102] To verify the practical application effect of an intelligent management and control system for pile foundation construction that combines karst cave detection and mud regulation, this embodiment selects a high-speed railway bridge pile foundation construction project in a karst development area as the application scenario.

[0103] The construction equipment selected is a rotary drilling rig with a drill bit radius of 0.8m and a drill rod outer radius of 0.25m. The multi-source data synchronous acquisition module 10, the fluid resistance adaptive calibration module 20, the parasitic torque decoupling module 30, the formation energy fingerprint recognition module 40, and the fluid-structure interaction decision module 50 are integrated into the vehicle control unit of the rotary drilling rig.

[0104] Before drilling begins, this invention pre-stores the inherent mechanical friction constant of the drilling rig under no-load as 5 kN·m, the measured value of the mud plastic viscosity as 0.02 Pa·s, and the measured value of the mud yield value as 10 Pa.

[0105] When the drilling depth reaches 20m, the operator performs a drill lifting operation, the drilling pressure applied by the drill bit is reduced to 2kN (less than the drilling pressure dead zone threshold), the mechanical drilling speed is reduced to 0.01m / min (less than the drilling speed dead zone threshold), and the drill rod rotation speed is maintained at 20 rpm.

[0106] The fluid resistance adaptive calibration module 20 identifies the hovering calibration state. At this time, the average total torque measured by the sensor is 35 kN·m, and the fluid resistance adaptive calibration module 20 calculates the theoretical fluid resistance torque as 25 kN·m.

[0107] The fluid resistance adaptive calibration module 20 performs the following calculations: The fluid torque after deducting mechanical losses is calculated as follows: the average total torque measured by the sensor is subtracted from the inherent mechanical friction constant of the drilling rig under no-load conditions, and the result is 30 kN·m.

[0108] Calculate the operating condition correction factor: Divide the fluid action torque after deducting mechanical losses by the theoretical fluid resistance torque to obtain an operating condition correction factor of 1.2. The fluid resistance adaptive calibration module 20 stores the operating condition correction factor in the register.

[0109] The drilling rig resumed mechanical drilling, and the drilling depth increased to 25m. The total torque measured by the sensor rose to 150kN·m. At this time, due to the increase in depth, the theoretical fluid resistance torque was recalculated to 30kN·m. The parasitic torque decoupling module 30 called the working condition correction coefficient and calculated the instantaneous fluid parasitic torque.

[0110] The calculation process is as follows: multiply the working condition correction coefficient by the current theoretical fluid resistance torque to obtain the corrected fluid resistance component, and then add the corrected fluid resistance component to the inherent mechanical friction constant of the drilling rig under no-load conditions to obtain the instantaneous fluid parasitic torque of 41 kN·m.

[0111] Parasitic torque decoupling module 30 calculates net cutting torque: the total torque measured by the sensor is subtracted from the instantaneous fluid parasitic torque to obtain a net cutting torque of 109 kN·m.

[0112] When the drill bit encounters an extremely hard rock layer, the mechanical drilling speed instantly drops to 0.01 m / h. The formation energy fingerprint recognition module 40 detects that the mechanical drilling speed is less than the preset minimum calculated drilling speed limit (set to 0.1 m / h). The formation energy fingerprint recognition module 40 forcibly uses the minimum calculated drilling speed limit to replace the denominator in the formula, calculates and corrects the mechanical specific energy, and prevents numerical calculation divergence.

[0113] When the borehole depth reaches 45m, the normalized gradient of the formation energy fingerprinting module 40 shows a sharp decrease in rock strength. At the same time, the formation hydraulic impedance calculated by the fluid-structure interaction decision module 50 decreases from 50MPa·s / L to 10MPa·s / L.

[0114] The fluid-structure interaction (FSI) decision module 50 calculates the critical breakdown warning index of the karst cave. If the result exceeds the danger trigger threshold, the FSI decision module 50 determines it to be a precursor to critical breakdown of the karst cave. The FSI decision module 50 sends a feedforward pressure regulation command to the mud pump frequency converter controller to increase the mud pump outlet pressure in advance. Subsequently, the formation hydraulic impedance further drops below the critical impedance value. The FSI decision module 50 outputs a flow compensation command to drive the mud pump to automatically increase the injection flow rate from 80L / s to 120L / s. At the same time, the FSI decision module 50 identifies this as a typical large cavity karst cave and automatically triggers a mud treatment plan for large cavity karst caves. At this time, the system displays that the intelligent decision algorithm has identified the feature signal matching the large cavity karst cave model and instructs the mud preparation unit to start the rapid mud preparation mode. The mud preparation unit is instructed to gradually increase the mud viscosity to 0.025Pa·s and strictly control the sand content to be below 4%, successfully maintaining the pressure balance inside the hole and avoiding drill bit loss accidents.

[0115] To verify the effectiveness of a pile foundation construction intelligent control system that integrates karst cave detection and mud regulation, a comparative test was conducted under the same geological conditions. The test subjects were two adjacent pile foundations, named the experimental pile and the control pile, respectively. The experimental pile was controlled by the intelligent control system for pile foundation construction that integrates karst cave detection and mud regulation of the present invention, while the control pile was controlled by the traditional manual experience mode.

[0116] See attached document Figure 3 There is a weak interlayer in the depth range of 38m to 40m. The control group uses the original mechanical specific energy calculation model (without removing fluid resistance and mechanical friction). As the proportion of parasitic torque increases with depth, the original mechanical specific energy curve shows an overall drifting trend, which causes the characteristic peak of the weak interlayer to be masked by background noise and fails to be effectively identified.

[0117] The experimental group used the corrected mechanical specific energy output by the formation energy fingerprinting module 40. The curve clearly showed the stable distribution of the corrected mechanical specific energy with depth, and a low value depression was shown at 38m to 40m, which accurately delineated the location of the weak interlayer. The decoupling algorithm of the parasitic torque decoupling module 30 improved the signal-to-noise ratio.

[0118] See attached document Figure 4In the figure, the solid curve corresponds to the left vertical axis, representing the change in formation hydraulic impedance; the dashed curve corresponds to the right vertical axis, representing the change in mud pump outlet pressure; the horizontal axis represents the time relative to the moment the cave is breached, with 0 representing the instant the cave breach occurs. As shown in the figure, at time -15s (i.e., drilling to 0.5m above the cave top), the solid line shows a clear inflection point in the formation hydraulic impedance and an accelerated decline. At this moment (-15s), the fluid-structure interaction decision module 50 identifies the risk and triggers a feedforward pressure regulation command, and the dashed line shows that the mud pump outlet pressure then begins to rise linearly. Before the leakage occurs at time 0s, the mud pump outlet pressure has completed pre-pressurization. When the cave is breached and the leakage channel is formed at time 0s, although the formation hydraulic impedance drops to a low value (approximately 10MPa·s / L), thanks to the feedforward pressure regulation and flow compensation commands, the bottom hole pressure only fluctuates briefly before quickly stabilizing, and the fluctuation amplitude is effectively controlled within a safe range.

[0119] The control pile mainly relies on the operator to observe the grout return situation. Data shows that manual grouting was only carried out 2 minutes after the complete leakage occurred, which led to local collapse of the borehole wall.

[0120] Effect Comparison Statistics Table The comparative statistics show that the fluid resistance adaptive calibration module 20 improves the accuracy of net cutting torque calculation under complex working conditions by subtracting and re-superimposing the mechanical friction constant. The formation energy fingerprint recognition module 40 introduces a minimum calculation drilling speed limit to ensure the robustness of the algorithm when drilling in extremely hard formations. The fluid-structure interaction decision module 50 realizes active defense against karst cave disasters through the coupling mechanism of mechanical and hydraulic dual physical fields. Compared with traditional methods, a pile foundation construction intelligent management and control system that coordinates karst cave detection and mud control shows significant advantages in terms of safety, identification accuracy and construction efficiency.

Claims

1. A smart management and control system for pile foundation construction that coordinates karst cave detection and mud regulation, characterized in that, include: The multi-source data synchronous acquisition module (10) is used to read data and construct a full-dimensional state vector encapsulating mechanical and fluid components; The fluid resistance adaptive calibration module (20) uses the mechanical component to identify the hovering calibration state, and combines the mechanical friction constant with the theoretical fluid resistance torque calculated using the fluid component and borehole geometric parameters to calculate the working condition correction coefficient. Parasitic torque decoupling module (30) reconstructs real-time fluid parasitic torque using the working condition correction coefficient, and extracts the real-time fluid parasitic torque from the total torque of the full-dimensional state vector to output net cutting torque; The formation energy fingerprinting module (40) uses the net cutting torque to calculate the corrected mechanical specific energy and the normalized change gradient of the corrected mechanical specific energy; The fluid-structure interaction decision module (50) is used to calculate the equivalent hydraulic impedance of the formation, integrate the normalized gradient and the equivalent hydraulic impedance of the formation to calculate the critical breakdown warning index of the karst cave, and output feedforward pressure regulation command and flow compensation command respectively to coordinate the control of mud state when the critical breakdown warning index of the karst cave exceeds the danger trigger threshold and the equivalent hydraulic impedance of the formation is lower than the critical impedance value.

2. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The full-dimensional state vector encapsulates the mechanical component and the fluid component; The mechanical components include the total torque measured by the sensor, the drilling pressure applied by the drill bit, the drill rod rotation speed, the mechanical drilling speed, and the current drilling depth; The fluid components include mud injection flow rate, return flow rate, mud pump outlet pressure, mud plastic viscosity, and mud yield value; The multi-source data synchronous acquisition module (10) adopts a unified reference clock source and maps all parameter values ​​to the same time axis coordinate point by downsampling and averaging high-frequency data and interpolating and fitting low-frequency data.

3. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 2, characterized in that, The fluid resistance adaptive calibration module (20) is preset with a drill pressure dead zone threshold and a drill speed dead zone threshold; When the monitored drilling pressure applied by the drill bit is less than the drilling pressure dead zone threshold, the absolute value of the mechanical drilling speed is less than the drilling speed dead zone threshold, and the drill rod rotation speed is greater than zero, the fluid resistance adaptive calibration module (20) determines that the drilling rig enters the hover calibration state through logical judgment. The fluid resistance adaptive calibration module (20) is used to trigger the calculation process of the working condition correction coefficient after the duration of the hover calibration state exceeds the calibration time threshold.

4. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The fluid resistance adaptive calibration module (20) uses the principle of concentric cylinder rotational viscometer to calculate the effective shear rate of the annulus using the outer radius of the drill rod, the radius of the borehole and the rotation speed of the drill rod, and calculates the theoretical fluid resistance torque by combining the plastic viscosity of the mud and the yield value of the mud. The fluid resistance adaptive calibration module (20) obtains the net fluid resistance torque by subtracting the mechanical friction constant from the total torque measured by the sensor under the hover calibration state, calculates the ratio of the net fluid resistance torque to the theoretical fluid resistance torque, and obtains the working condition correction coefficient that characterizes the combined effect of borehole geometric irregularity and mechanical transmission efficiency.

5. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The parasitic torque decoupling module (30) is used to calculate the current theoretical fluid resistance torque in real time when the drilling rig is in mechanical drilling state, multiply the current theoretical fluid resistance torque by the working condition correction coefficient to obtain the corrected fluid resistance component, and add the fluid resistance component to the mechanical friction constant to reconstruct the real-time fluid parasitic torque. The parasitic torque decoupling module (30) calculates the difference between the total torque measured by the sensor and the real-time fluid parasitic torque, and performs a non-negative logic judgment on the difference: If the difference is less than zero, it is forcibly set to zero; if the difference is greater than or equal to zero, the difference is retained as the net cutting torque.

6. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 2, characterized in that, The formation energy fingerprinting module (40) uses the net cutting torque to replace the traditional ground measurement torque, and calculates the corrected mechanical specific energy by combining the drilling pressure applied by the drill bit, the rotation speed of the drill rod, the mechanical drilling speed and the drill bit radius; The formation energy fingerprinting module (40) uses the finite difference method to obtain the normalized gradient of formation intensity per unit depth by dividing the difference of the corrected mechanical specific energy at adjacent sampling times by the difference of the corresponding borehole depth.

7. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 2, characterized in that, The fluid-structure interaction decision module (50) uses Darcy's law and fluid network theory to calculate the equivalent hydraulic impedance of the formation by using the sum of the absolute value of the difference between the mud pump outlet pressure and the mud injection flow rate and the return flow rate and the anti-singularity regularization constant. The fluid-structure interaction decision module (50) uses the formation equivalent hydraulic impedance to quantify the current formation's resistance to high-pressure mud intrusion and identify the attenuation trend of rock strength in the vertical depth.

8. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The fluid-structure interaction decision module (50) is preset with a hazard triggering threshold, a mechanical characteristic weighting coefficient, and a hydraulic characteristic weighting coefficient; The fluid-structure interaction decision module (50) uses a weighted normalization algorithm to weight and sum the ratio of the normalized change gradient to the reference value of the reference rock specific energy gradient, and the ratio of the reference value of the hydraulic impedance of the intact stratum to the equivalent hydraulic impedance of the stratum, to generate the critical breakdown early warning index of the karst cave. When the critical breakdown warning index of the karst cave exceeds the danger trigger threshold, the fluid-structure interaction decision module (50) determines it as a precursor to critical breakdown of the karst cave.

9. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The fluid-structure interaction decision module (50) is used to calculate the compensation mud injection flow rate by means of impedance matching when outputting the flow compensation command; The fluid-structure interaction decision module (50) establishes a mapping relationship between flow gain and impedance attenuation ratio. Based on the baseline injection flow before leakage occurs, it uses a logarithmic function to calculate the ratio of the complete formation hydraulic impedance reference value to the formation equivalent hydraulic impedance at the current moment as the gain factor, and combines the flow regulation response coefficient to determine the compensation mud injection flow that can offset the formation leakage channel capacity.

10. The intelligent control system for pile foundation construction that coordinates karst cave detection and mud regulation according to claim 1, characterized in that, The fluid-structure interaction decision module (50) triggers the preventive adjustment logic of the mud pump pressure through the feedforward pressure adjustment command, and establishes pressure balance in advance before the drill bit actually penetrates the top of the karst cave. The fluid-structure interaction decision module (50) sends the feedforward pressure regulation command and the flow compensation command to the mud pump frequency converter in the mud circulation pipeline, thereby driving the mud pump drive unit to perform the corresponding pumping output adjustment.

Images