Intelligent deep iteration based positive and negative circulation perfusion pile construction information automatic acquisition system
By using an intelligent depth iteration system, combined with multi-source sensor data fusion and adaptive mud circulation control, the problems of depth measurement error and mud circulation mode lag in cast-in-place pile construction have been solved, achieving high-precision hole formation and stable data acquisition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA CONSTR EIGHT ENG DIV CORP LTD
- Filing Date
- 2026-03-02
- Publication Date
- 2026-06-12
Smart Images
Figure CN122190248A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of pile foundation construction monitoring technology, specifically to an automatic information acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration. Background Technology
[0002] As a major foundation construction method, the quality of borehole formation in cast-in-place piles directly determines the bearing capacity of the pile foundation. Precise control of drilling depth, verticality, and mud circulation process is crucial for ensuring project quality during construction.
[0003] However, existing monitoring technologies for cast-in-place pile construction primarily rely on single-sensor acquisition methods for depth measurement, such as simply installing a rotary encoder or proximity switch at the winch end for pulse counting. This approach ignores the nonlinear diameter changes that occur when the wire rope is wound multiple times on the drum, fails to adequately consider the elastic tensile deformation of the wire rope under load during deep hole operations, and lacks effective compensation for the drill rod's tilt. The combination of these physical factors results in the measured depth data being only an estimate of the inclined length along the drill rod axis, rather than the true vertical depth of the hole bottom. Furthermore, the cumulative error increases significantly with increasing drilling depth, making it difficult to meet the requirements of high-precision construction.
[0004] In terms of mud circulation process control, the switching between forward and reverse circulation modes currently relies mainly on the operator's experience and judgment. Operators typically subjectively infer changes in the formation at the bottom of the borehole based on drilling rig vibration, sound, or mud return, and manually operate valve groups to switch modes. This manual intervention method lacks real-time quantitative data support and exhibits a lag in response to sudden changes in formation lithology. When encountering different geological conditions such as clay, quicksand, or hard rock layers, failure to promptly match the optimal cuttings removal mode and mud parameters can easily lead to low cuttings removal efficiency, and even construction accidents such as stuck drill, buried drill, or borehole collapse due to pressure imbalance within the borehole.
[0005] Furthermore, drilling operations are conducted in harsh environments with frequent mechanical vibrations and electromagnetic interference. Existing data acquisition systems often lack robust signal processing mechanisms and fail to establish accurate nonlinear mathematical models of mechanical transmission. They also lack effective smoothing and compensation methods for random noise and system biases in the raw signals acquired by sensors. This results in monitoring data prone to jumps or interruptions under complex working conditions, leading to poor data continuity and stability, and making it difficult to accurately reproduce changes in the physical state during construction. Summary of the Invention
[0006] To address the shortcomings of existing technologies, this invention provides an automatic information acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration. This system solves the problems in existing cast-in-place pile construction where depth measurement accuracy is affected by multiple physical factors such as multi-layer winding of steel wire rope, elastic tension, and drill rod inclination, and where mud circulation mode switching relies on manual experience, leading to response lag and mismatch between process and stratum.
[0007] To achieve the above objectives, the present invention is implemented through the following technical solution: an automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration, including a data acquisition hardware terminal, a mechanical control actuator, an industrial control center, and a remote data management terminal.
[0008] The data acquisition hardware terminal includes various sensor components: a fixed-height antenna rigidly mounted on the top of the drill bit power unit; a Hall sensor mounted on the winch stator support; a current sensor mounted on the drill rig power cable; an inclination sensor mounted on the side of the drill rod or mast bottom; and a mud flow sensor mounted in the mud circulation pipeline. The data acquisition hardware terminal is used to collect data on BeiDou elevation, winch rotation status, motor current, drill rod inclination, and mud flow.
[0009] The mechanical control actuator includes a solenoid valve control device and a mud pump intelligent control system, which are used to perform cycle mode switching and pumping parameter adjustment actions according to the received instructions.
[0010] The industrial control center receives multi-source data collected by the data acquisition hardware terminal and executes intelligent depth iteration logic. Specifically, the industrial control center fuses the elevation data collected by the fixed-elevation antenna with the wire rope displacement data collected by the Hall sensor to generate a borehole depth sequence, and corrects the borehole depth sequence using the tilt angle collected by the tilt sensor. Furthermore, the industrial control center identifies the formation type based on the current data collected by the current sensor and the real-time drilling speed, and sends control commands to the solenoid valve control device based on the identified formation type, thereby controlling the system to automatically switch to forward or reverse circulation mode.
[0011] Preferably, the data acquisition hardware terminal further includes positioning and directional antennas installed at both ends of the drilling rig gantry crossbeam. These positioning and directional antennas are used in conjunction with the elevation-fixed antenna to calculate the planar coordinates and heading angle of the drilling rig body. The phase center of the elevation-fixed antenna is calibrated to the rotation center axis of the drill rod, so that the elevation-fixed antenna can directly sense the vertical displacement of the drill bit.
[0012] Preferably, the Hall sensor integrates a Hall element, which, in conjunction with a multi-turn primary coil, detects changes in the magnetic field generated when the wire rope moves and drives the main shaft to rotate, thereby obtaining the number of rotations and the direction of rotation of the drum. The data acquisition hardware terminal also includes an absolute encoder, which is mechanically coupled to the winch main shaft via a coupling and serves as a verification source for the Hall sensor data.
[0013] Preferably, the specific method by which the industrial control center performs intelligent depth iteration is as follows: A rigid geometric link model is constructed, and the main depth sequence is generated by combining the real-time elevation data acquired by the fixed-elevation antenna with the elevation of the ground calibration point and the total length of the drill bit; simultaneously, a nonlinear winding model is constructed, and the wire rope release length is calculated by combining the data collected by the Hall sensor with the drum layer diameter correction logic; a data fusion engine is run to perform weighted fusion of the main depth sequence and the wire rope release length to output a fused depth sequence; finally, the fused depth sequence is geometrically corrected using the tilt angle acquired by the tilt sensor to generate a hole-forming depth sequence.
[0014] Preferably, the data fusion engine adopts a hierarchical fusion architecture. In the first-level fusion, the industrial control center uses a Kalman filter algorithm to perform joint state estimation and data fusion processing on the two signals from the Hall sensor, generating a displacement pulse sequence fused at the signal level. In the second-level fusion, the industrial control center establishes a state equation describing the vertical motion characteristics of the drill bit, uses the main depth sequence, the wire rope release length, and the absolute encoder data as observation inputs, and compensates for nonlinear errors through state-space estimation and a neural network model, outputting calibrated depth data.
[0015] Preferably, when performing corrections, the industrial control center also performs verticality geometric correction, that is, based on the tilt angle of the drill rod axis relative to the vertical line of gravity measured by the tilt sensor, the oblique depth along the drill rod axis is converted into the true vertical depth perpendicular to the ground plane using trigonometric function relationships.
[0016] Preferably, the current sensor is clamped and installed on the main power supply cable of the drilling rig's main motor. The industrial control center has a built-in physical model of formation resistance. This model is used to convert the main motor current collected by the current sensor into a drilling resistance index, and the drilling resistance index and drilling speed are matched with a preset geological parameter database to identify the formation type.
[0017] Preferably, the solenoid valve control device is connected to and drives the pipeline reversing valve assembly. The industrial control center controls the solenoid valve control device in the following way: when the identification result is either clay or silt layer, a positive circulation mode command is sent to the solenoid valve control device to drive the pipeline reversing valve assembly to connect the mud pump outlet and the drill pipe cavity; when the identification result is either sand or gravel layer, a reverse circulation switching command is sent to the solenoid valve control device to drive the pipeline reversing valve assembly to switch the flow channel to connect the drill pipe cavity and the mud pump inlet, forming a negative pressure slag suction channel.
[0018] Preferably, when the identification result is a hard rock layer, the industrial control center triggers the air lift reverse circulation mode, that is, sends a start command to the mechanical control actuator to activate the matching air compressor unit and inject high-pressure gas into the bottom mixing chamber.
[0019] Preferably, the system further includes a remote data management terminal, used to receive encrypted sensor data and generate hash values for key quality data and write them into a blockchain node; the remote data management terminal is also used to reconstruct the construction process in a visualization interface based on the data from the mud flow sensor, borehole trajectory, and formation information.
[0020] This invention utilizes elevation data from a fixed-height antenna and displacement data from a winch sensor for multi-source fusion, and combines this with tilt angle data for geometric correction, thereby achieving accurate calculation of borehole depth. Simultaneously, by establishing a mapping relationship between current, drilling speed, and formation, it automatically identifies the formation and switches the mud circulation mode, achieving adaptive control of the construction process.
[0021] This invention provides an automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration. It offers the following advantages: 1. This invention constructs an intelligent depth overlay logic, which weights and fuses the absolute elevation data of the fixed-elevation antenna with the relative displacement data of the Hall sensor, and introduces tilt sensor data for verticality geometric correction. The fusion of multi-source heterogeneous data effectively avoids the cumulative error introduced by the single winch counting due to the multi-layer winding of the wire rope, elastic stretching, and drill rod tilting, thereby obtaining high-precision true vertical depth data and ensuring the accuracy of hole quality detection.
[0022] 2. This invention collects the main motor current and real-time drilling speed, identifies the formation type based on the formation resistance physical model, and controls the solenoid valve device to automatically switch between forward and reverse circulation modes. This mechanism changes the traditional lag situation of relying on human experience to judge the formation and manually switch valves, and realizes adaptive matching of mud circulation process for different geological conditions such as clay, sand or rock, which improves the efficiency of slag removal and reduces the risk of stuck drill.
[0023] 3. This invention establishes a nonlinear winding model and a two-level data fusion architecture, uses Kalman filtering to smooth sensor signal noise, and combines an absolute encoder as a verification source. It compensates for the nonlinear diameter variation error caused by the layered winding of the wire rope on the drum through state space estimation, thereby eliminating random interference and system deviation in the mechanical transmission process and ensuring the continuity and stability of borehole depth monitoring data under complex working conditions. Attached Figure Description
[0024] Figure 1 This is a schematic diagram of the overall architecture of the automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration of the present invention. Figure 2 This is a schematic diagram of the fixed-elevation antenna of the present invention; Figure 3 This is a schematic diagram of the Hall element of the present invention; Figure 4 This is a schematic diagram showing the physical installation layout and connection relationship of the sensing layer hardware of the present invention on the drilling rig; Figure 5 This is a schematic diagram illustrating the principles of digital physical model construction and deep incremental calculation in this invention. Figure 6 This is a schematic diagram of the multi-source data fusion engine based on Kalman filtering and neural networks of the present invention. Figure 7 This is a flowchart of the formation identification and adaptive closed-loop control of forward and reverse circulation modes of the present invention.
[0025] Among them, 1. Fixed-altitude antenna; 2. Positioning and directional antenna; 3. Tilt sensor; 4. Hall sensor; 5. Solenoid valve control device; 6. Mud flow sensor; 7. Hall element; 8. Current sensor. Detailed Implementation
[0026] 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.
[0027] See attached document Figure 1 -Appendix Figure 4This invention provides an automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration. The system consists of a data acquisition hardware terminal, an industrial control hub, a mechanical control actuator, and a remote data management terminal. The physical carrier of the system is a forward and reverse circulation drilling and grouting machine. All components are connected via a CAN industrial bus or RS485 bus to form a closed-loop control network capable of data acquisition, real-time calculation, and action execution.
[0028] In the specific implementation of the system hardware architecture, multiple sensor components are arranged on the mechanical structure of the drilling rig. The BeiDou RTK positioning and directional antenna set includes one BeiDou RTK elevation-fixed antenna 1 and two BeiDou RTK positioning and directional antennas 2. The BeiDou RTK elevation-fixed antenna 1 is rigidly fixed to the top of the drill bit power unit. This mounting position is secured by a rigid bracket, ensuring that the phase center of the BeiDou RTK elevation-fixed antenna 1 maintains a fixed mechanical geometric distance from the bottom surface of the drill bit on the vertical axis. This allows the vertical displacement sensed by the BeiDou RTK elevation-fixed antenna 1 to be directly mapped to the drill bit's depth of cut through geometric conversion. The two BeiDou RTK positioning and directional antennas 2 are respectively installed at both ends of the drilling rig gantry crossbeam. The line connecting the two antennas forms a positioning baseline, used to calculate the planar coordinates and heading angle of the drilling rig body relative to the geodetic coordinate system.
[0029] To achieve redundant measurement of wire rope displacement, the system incorporates a dual detection mechanism at the winch location. The dual-channel Hall sensor 4, equipped with a non-contact probe, is mounted on the stator supports on both sides of the winch's main shaft. The Hall sensor 4 utilizes Hall elements 7, combined with a multi-turn primary coil design, to detect changes in the magnetic field generated when the wire rope moves and drives the main shaft to rotate, thus achieving high-precision measurement of the wire rope's lowering depth. The combination of Hall element 7 and the multi-turn coil ensures that the Hall sensor 4 maintains a measurement accuracy of ±0.5% even under low current and high vibration conditions. The Hall sensor 4, in conjunction with the main shaft gear or magnetic ring, detects the number of rotations and the direction of rotation of the drum. An absolute encoder is directly mechanically coupled to the winch's main shaft via a coupling, serving as the verification source for the Hall sensor 4 data. The power monitoring component consists of dual-channel Hall current sensors 8, clamped and installed on the main power cable of the drilling rig's main motor and the power cable of the mud pump motor, respectively, to collect analog current measurements during motor operation.
[0030] The attitude monitoring component is a dual-axis tilt sensor 3, rigidly mounted on the bottom side of the square drill pipe or mast guide rail. Its sensitive axis is parallel to the drill pipe axis, used to measure the tilt angle of the drill pipe axis relative to the vertical line of gravity in real time. In the mud circulation pipeline, the system is equipped with a mud flow sensor 6 and a circulation mode switching actuator. This actuator includes a pipeline reversing valve assembly installed in the pipeline between the mud pump, the top flange of the drill pipe and the mud pit, as well as a vacuum pump or air lift pipeline control valve matched with the reverse circulation pipeline.
[0031] The system's computation and control core employs an industrial control hub (i.e., an industrial-grade controller, such as an embedded processing unit integrating an industrial 4G module) based on an edge computing architecture. This industrial control hub acts as the master station of the industrial bus, polling or receiving data frames from various sensors at fixed time intervals (e.g., 100 milliseconds). The industrial control hub is internally configured with a unified clock source for timestamping and interpolating / resampling sensor data at different sampling rates, ensuring strict synchronization of multi-source data in the time dimension.
[0032] As the specific hardware implementation of the mechanical control actuator, the system is equipped with an integrated control execution unit. This unit is independent of the industrial control center and is specifically responsible for action execution. The control execution unit mainly consists of two parts: firstly, a solenoid valve control device 5 installed in the drilling rig's circulation system. This device serves as the physical execution center for switching circulation modes, integrating a signal decoding unit and a power drive circuit. It is specifically designed to receive control algorithm commands from the industrial control center and, through precise drive of hydraulic or pneumatic solenoid valve groups, achieve automatic switching between forward and reverse circulation modes in the pipeline. Secondly, a mud pump intelligent control system installed in the mud pump system. This system comprises a frequency converter, pump body sensors, and an embedded logic controller. It can receive commands from the industrial control center in real time and perform closed-loop control of the mud pump's speed and torque, thereby dynamically adjusting the mud pump's flow rate, pressure, and other operating parameters.
[0033] The overall workflow of the system during operation includes initial calibration, synchronous acquisition of multi-source data, intelligent depth iteration, formation identification and cycle mode decision-making, and closed-loop control execution. After the system is powered on, it waits for the BeiDou RTK signal to converge to a fixed solution state. During the depth zero-point calibration stage before drilling begins, when the operator adjusts the bottom surface of the drill bit to be level with the top surface of the casing or the ground reference point, the system records the current BeiDou antenna elevation value, encoder value, and cumulative pulse count of Hall sensor 4, locking the state at that moment as the initial zero point for depth calculation. All subsequent depth data are incrementally calculated based on this zero point.
[0034] During drilling, the system runs intelligent depth iteration logic to obtain high-precision hole depth. The industrial control center uses real-time elevation data acquired by Beidou RTK fixed-elevation antenna 1, adds the elevation of ground calibration points, and subtracts the current total length of the drill string (including the cumulative length of the drill rod and the fixed deviation from the antenna to the power head) to generate the main depth sequence. Simultaneously, the industrial control center uses the number of drum rotations collected by Hall sensor 4 or encoder, combined with preset drum bottom diameter and wire rope diameter parameters, to calculate the wire rope release length. During the calculation, the system automatically determines the number of layers of wire rope wound on the drum based on the cumulative release length and performs arithmetic compensation for changes in the winding diameter to eliminate nonlinear errors caused by multiple layers of winding.
[0035] To obtain highly reliable depth data, the industrial control center runs a Kalman filter algorithm or a neural network fusion algorithm to weight and fuse the depth values calculated by BeiDou and the depth values calculated by wire rope displacement, outputting a fused depth sequence. Subsequently, the system executes verticality geometric correction logic, using the drill rod inclination angle measured by tilt sensor 3 to convert the slant depth along the drill rod axis into the true vertical depth perpendicular to the ground plane using trigonometric functions, thereby eliminating the positive depth measurement error caused by hole inclination.
[0036] While acquiring precise depth, the system executes formation identification and automatic decision-making processes for circulation modes in parallel. The industrial control center, based on the collected main motor current and real-time calculated drilling speed, uses a pre-set formation resistance physical model to invert the current formation characteristics. This model expresses formation resistance as a linear function of current, including a coefficient representing formation hardness and a compensation constant to offset equipment idle losses. This constant is pre-calibrated through a power head idling test. The system performs feature matching between the calculated resistance index and drilling speed and a pre-set geological parameter database to identify the currently encountered formation type (e.g., clay, sand, rock).
[0037] Based on the identification results of the geological strata type, the industrial control center automatically generates control commands and sends them to the control execution unit to adjust the construction process.
[0038] When the identification result is clay or silt layer and the drilling speed is lower than the preset efficiency threshold, the system determines that the slag removal capacity needs to be improved. The industrial control center sends a positive circulation mode command to the solenoid valve control device 5 in the control execution unit, driving the pipeline reversing valve group to connect the mud pump outlet and the drill pipe cavity. At the same time, the industrial control center sends an acceleration command to the mud pump intelligent control system in the control execution unit. The mud pump intelligent control system dynamically increases the mud pump operating frequency and increases the pumping flow to enhance the bottom scouring.
[0039] When the identification result is a sand layer or pebble layer and drilling is blocked, the system generates a reverse circulation switching command and sends it to the solenoid valve control device 5. The device drives the valve to switch the flow channel according to the time sequence and starts the vacuum pump or air lift pipeline to form a negative pressure slag suction channel. At this time, the mud pump intelligent control system receives fluid load feedback and dynamically adjusts the pump speed to the optimal slag suction condition.
[0040] When the identification result is a high-hardness rock layer, the system, while maintaining the reverse circulation mode, coordinates the pressure setpoint of the drilling rig pressurization system and the speed of the main motor, and instructs the mud pump intelligent control system to maintain a constant mud replenishment pressure, so as to achieve adaptive matching of drilling pressure, speed and mud circulation parameters.
[0041] The system encapsulates the borehole depth, verticality, formation type, circulation mode status, and raw sensor data generated by the above processing through an encrypted communication module. After encryption using algorithms such as AES, the data is transmitted in real time to a remote monitoring platform via a wireless network. The remote monitoring platform parses the received data and reconstructs the borehole trajectory and formation profile in a 3D visualization interface. Simultaneously, it generates hash values for key quality data (such as final borehole depth and rock penetration determination results) and writes them to blockchain nodes, completing the immutable storage and full-process traceability of the data.
[0042] See attached document Figure 1 -Appendix Figure 4 To ensure the system operates stably in construction sites with strong vibrations, high dust levels, and severe electromagnetic interference, the sensing layer hardware employs a specific physical integration method and signal conditioning mechanism.
[0043] The deployment location of the two BeiDou RTK positioning and directional antennas directly determines the geometric reference for depth calculation. BeiDou RTK altitude-fixing antenna 1 is mounted on the central axis of the drill rig's power head via a custom-designed high-strength alloy steel vibration-damping bracket. This bracket is filled with high-damping rubber material, forming a double-layer physical filter structure to filter high-frequency mechanical vibrations generated during drilling and prevent loosening of the antenna's internal RF circuitry or phase center drift. The phase center of BeiDou RTK altitude-fixing antenna 1 is precisely calibrated to the rotational axis of the drill pipe, ensuring that its horizontal displacement reflects only the planar movement of the drill rig, while its vertical displacement directly represents the lifting and lowering motion of the power head. Two BeiDou RTK positioning and directional antennas 2 are respectively positioned at both ends of the drill rig's gantry crossbeam, with a fixed physical distance between them (e.g., 2.5 meters), forming a long baseline structure to improve the accuracy of heading angle calculation. Two Beidou RTK positioning and directional antennas 2 are connected to the GNSS receiver in the cab via low-loss coaxial cables. The receiver is equipped with a carrier phase differential algorithm module, which can output positioning data and heading data with centimeter-level accuracy.
[0044] For displacement detection in the winch section, a dual-channel Hall sensor 4 is encapsulated in a stainless steel sleeve with an IP67 protection rating and fixed to the stator base on the winch drum gear side by threads. The air gap between the sensing end face of the Hall sensor 4 and the gear tooth tip is adjusted to between 2 mm and 4 mm. When the drum rotates, the alternating passage of the gear tooth tip and tooth valley causes a change in the magnetic field. The internal circuit of the Hall sensor 4 converts this into two square wave pulse signals (phase A and phase B) with a 90-degree phase difference. The industrial control center captures the rising and falling edges of these two signals through hardware interrupts and compares their phase lead or lag relationship to determine the rotation direction and number of rotations of the drum. A multi-turn absolute encoder is selected, and its shaft is connected to the winch main shaft end face through a coupling with an elastic buffer pad. The coupling with the elastic buffer pad can absorb the slight axial movement and radial runout of the main shaft under heavy load conditions, protecting the encoder bearing from mechanical stress damage. The SSI or Modbus protocol digital signal output by the absolute encoder is transmitted through shielded twisted pair cable, providing absolute position information of the drum rotation angle. This allows for quick restoration of the winch state after a power outage and restart, without the need for recalibrating the mechanical zero point.
[0045] The dual-channel Hall current sensor 8 in the power monitoring component adopts an open-type magnetic balance design, directly snapping onto any one phase of the three-phase power supply cable of the main motor and the mud pump motor without disconnecting the circuit. The current sensor 8 integrates a signal conditioning circuit, linearly converting the detected large current (typically hundreds of amperes) into a standard 4-20mA analog current signal or a 0-10V voltage signal. This standard signal transmission method has strong anti-attenuation capabilities. To prevent high-order harmonic interference generated by the frequency converter, an RC passive low-pass filter is connected in series at the output of the current sensor 8, filtering out high-frequency noise before connecting to the analog input interface of the industrial control center.
[0046] The attitude monitoring component, namely the dual-axis tilt sensor 3, is mounted on a precision-machined aluminum alloy mounting base. This mounting base is attached to the bottom side reference surface of the mast guide rail or power head via strong magnetic attraction or bolt fastening. During installation, a spirit level is used to calibrate the sensor so that its X-axis is parallel to the drill pipe's forward and backward tilt direction, and its Y-axis is parallel to the drill pipe's left and right tilt direction. The tilt sensor 3 internally uses a MEMS accelerometer chip and is filled with high-viscosity silicone oil damping fluid to suppress signal glitches caused by drilling impacts, enabling it to output real-time tilt angle data of the drill pipe relative to the gravity vector.
[0047] The flow sensor in the mud circulation system uses electromagnetic induction and is installed on a straight section of the mud pump outlet pipeline. A straight pipe section at least five times the pipe diameter is required upstream of the sensor, and at least three times the pipe diameter is required downstream, to eliminate the influence of fluid turbulence on the induced electromotive force. The circulation mode switching actuator, part of the control unit, uses a high-torque electric actuator to drive the pipeline reversing valve assembly. This high-torque electric actuator integrates a position feedback potentiometer, which can provide real-time feedback of the valve core's opening status to the industrial control center, ensuring that the flow channel switching action is executed correctly.
[0048] All sensors distributed throughout the drilling rig are connected to a junction box located in the operator's cab or electrical cabinet via industrial-grade double-shielded cables. The cable shielding is grounded at one end on the junction box side to eliminate electromagnetic interference caused by grounding loops. The junction box transmits pre-processed digital and analog signals to the industrial control center via a CAN industrial bus interface. The processing unit triggers a global synchronization sampling signal using an FPGA or high-precision hardware timer, latching all sensor values at the same moment to provide time-aligned raw data frames for subsequent depth calculations and formation identification.
[0049] See attached document Figure 5 To support subsequent high-precision depth calculations, a digital physical model is pre-installed inside the industrial control center. The digital physical model transforms the mechanical structural dimensions of the drilling rig, the spatial relationship of the sensors, and the winding characteristics of the flexible connectors into mathematical expressions, serving as a benchmark framework for real-time calculations.
[0050] Before drilling operations commence, the system first establishes a rigid geometric link model based on two sets of BeiDou RTK positioning directional antennas. This rigid geometric link model defines the vertical mapping relationship from the satellite positioning phase center to the bottom of the drill bit. The industrial control center reads the mechanical dimensional parameters stored in non-volatile memory and dynamically updates the total length of the drill string assembly based on the connection rod operation records, constructing the following BeiDou depth geometric equation: ; in: Indicates at time The main depth sequence calculated based on BeiDou RTK data; Indicates at time The total length of the drill assembly includes the length of the drill bit, the length of the power head connector, and the sum of the lengths of all currently connected drill pipes. The system dynamically updates this value based on the add / remove pipe commands confirmed on the interface. This indicates the fixed vertical distance from the phase center of the Beidou RTK fixed-elevation antenna 1 to the bottom connecting flange surface of the power head. This distance is determined by the geometry of the antenna mounting bracket and is a fixed constant. Indicates at time Real-time absolute elevation measured by Beidou RTK fixed-elevation antenna 1; This indicates the ground elevation set at the construction site. This value is determined during the initial calibration stage before drilling.
[0051] Meanwhile, for the displacement calculation of the winch wire rope, the system established a nonlinear winding model with a multi-layer winding compensation mechanism. Given that the layering of wire rope on the drum leads to a progressively increasing winding diameter, calculating based solely on a fixed diameter would result in significant cumulative errors. Therefore, the nonlinear winding model introduces layered diameter correction logic. The industrial control center pre-calculates the maximum number of turns that each layer can accommodate based on the effective width of the drum and the wire rope diameter. When the cumulative winding and unwinding length fed back by Hall sensor 4 or the encoder exceeds the single-layer rope capacity of the drum, the nonlinear winding model automatically determines that the number of winding layers has changed, and the calculated diameter is automatically increased or decreased by twice the wire rope diameter.
[0052] To accurately describe this process, the system employs a depth increment model to quantify displacement changes at different layers, the mathematical expression of which is: ; in: This indicates the increase in wire rope depth per revolution of the drum, given the current number of winding layers. Pi is a constant. This indicates the initial diameter (plain drum diameter) of the winch drum, which is preset in the system configuration file; This indicates the current winding layer number of the wire rope, with the bottom layer defined as layer 0 or layer 1 (depending on the specific algorithm index), and increasing sequentially outwards; This indicates the nominal diameter of the wire rope.
[0053] Based on the aforementioned depth increment model, the industrial control center obtains the total displacement of the wire rope through real-time integration calculations. Simultaneously, the system also establishes a geometric correction logic for verticality deviation, which physically defines the projection relationship between the drill rod tilt angle and the depth measurement. That is, when the tilt sensor 3 detects a tilt angle in the drill rod... At this time, based on the principles of trigonometric geometry, the system maps the tilt measurement length along the drill rod axis to a true depth component perpendicular to the ground, ensuring that even under non-ideal working conditions where the drill rod tilts, the calculated depth data still accurately reflects the vertical depth of the borehole. This series of pre-defined physical models and parameter definitions lays the mathematical foundation for the subsequent fusion of multi-source sensor data into highly reliable depth information.
[0054] Based on the establishment of the digital physical model, the system enters the real-time solution stage, which aims to extract high-confidence borehole depth values from the raw satellite signals containing high-frequency noise and environmental interference.
[0055] The industrial control center continuously monitors the NMEA-0183 standard format data stream output by the BeiDou receiver via a serial communication interface, specifically locking onto GPGGA or GNGGA statements. The system first verifies the validity of the quality indicator field in the data frame; only when the field value is 4 (RTK fixed solution) or 5 (RTK floating-point solution) is the frame data included in the subsequent calculation queue. To eliminate initial system errors, before drilling begins, when the bottom of the drill bit is aligned with the ground reference point (or the top of the casing), the system executes a depth zero-point calibration procedure, marking the BeiDou antenna elevation value at this moment as the initial reference zero point. All subsequent depth calculations are based on differential operations performed from this zero point.
[0056] Considering the inconsistent data sampling rates of multiple sensors (e.g., BeiDou data is typically 1Hz-10Hz, while current and Hall signals are even higher), the industrial control center utilizes a built-in unified clock to resample or linearly interpolate and align the input data at fixed intervals (e.g., 100ms) to ensure strict consistency of the data input to the algorithm in the time dimension. Simultaneously, the system presets measurement noise covariance parameters based on sensor accuracy specifications, such as setting parameters for BeiDou RTK elevation errors. (e.g., 0.0009) to provide statistical weighting basis for subsequent data fusion algorithms.
[0057] After acquiring time-aligned elevation data, the system calculates the main depth sequence based on the BeiDou depth geometry equation. Since BeiDou RTK measures the vertical elevation difference, while the subsequent fusion object, the wire rope displacement, represents the slant length along the borehole trajectory, the system introduces real-time tilt angle data to preprocess the BeiDou data to unify the physical dimensions. The system uses the following formula to convert the vertical depth projection measured by BeiDou into a slant length sequence along the drill pipe axis: ; in: Indicates at time The converted slant length sequence (or Beidou slant length depth) along the drill rod axis represents the actual displacement of the drill bit along the drilling trajectory and serves as the observation input for the data fusion engine. Indicates at time The vertical depth projection (or vertical depth sequence) measured by Beidou RTK fixed-height antenna 1 is calculated by subtracting the real-time antenna elevation from the ground calibration elevation and deducting the drill string length. This represents the geometric scaling factor that projects the length along the drill pipe path onto the direction of gravity's vertical line.
[0058] The transformed sequence represents the displacement of the drill bit along the drilling trajectory, serving as one of the observation inputs for the subsequent data fusion engine. At this stage, the system ensures that all data fed into the fusion engine physically represents the path length along the trajectory, thus maintaining spatial isomorphism with the linear displacement measured by the winch, laying the foundation for generating high-precision fused slant length and depth.
[0059] See attached document Figure 6 After obtaining single-source depth data processed and tilt-corrected by BeiDou, the system activates a data fusion engine to address issues such as signal loss, multipath effects, and elevation jumps that can occur with single satellite signals in deep foundation pits or urban canyons. The data fusion engine employs a layered fusion architecture, including a first-level fusion for similar sensors and a second-level fusion for heterogeneous sensors. Through a combination of state-space estimation and ELM neural network correction, it outputs a final depth sequence with high continuity and accuracy.
[0060] First, the industrial control center performs the first level of fusion. The system reads data from the dual-channel Hall sensors installed on both sides of the winch spindle, and uses a Kalman filter algorithm to weight and smooth the two signals from the same source, eliminating transient pulse fluctuations or electromagnetic interference noise that may occur from a single sensor, thus generating highly stable basic data for wire rope displacement. Simultaneously, the system reads data from the absolute encoder and calculates the auxiliary reference depth. , and Beidou's solution Together they form a multi-source observation set.
[0061] Subsequently, the system enters the second-level core fusion phase. The system establishes a state equation describing the vertical motion characteristics of the drill bit, defining the real-time vertical depth and vertical descent velocity of the drill bit as the system state vector. The industrial control center, based on a discrete-time dynamics model, uses the state estimate from the previous time step to deduce the prior state at the current time step. This state evolution process follows the following state equation: ; in: Indicates at time The system state vector includes drill depth and descent speed; The state transition matrix describes the kinematics of the drill bit within the sampling interval and is constructed based on a uniform velocity or uniform acceleration model. Indicates the previous moment The posterior state estimate; This represents the control matrix, used to incorporate the effects of external control inputs. This represents the control input vector, which corresponds to the adjustment amount of the mud pump flow rate or the drill rig power head speed during the drilling process; The process noise vector is represented by its covariance matrix. This reflects the uncertainty deviation between the system model and the actual motion.
[0062] After obtaining the prior state, the system incorporates observation data including BeiDou RTK depth, Hall sensor 4 depth, and encoder depth to construct observation equations to correct the state estimation. The observation equations are expressed as follows: ; in: Indicates at time The sensor observation vector contains , and ; The observation matrix is used to map the state space to the observation space and establish a linear relationship between state variables and observations. Indicates at time The system state vector includes drill depth and descent speed; The observation noise vector is represented by the measurement noise covariance matrix. describe.
[0063] To achieve adaptive and accurate fusion, the system dynamically adjusts filtering parameters based on the sensor's physical characteristics and real-time operating conditions. The system has a preset measurement noise covariance matrix. The initial values, where the covariance components corresponding to the BeiDou RTK elevation error are... The covariance component of the Hall sensor 4 error is set to 0.0009 (corresponding to an accuracy of approximately ±3cm). Set to 0.0025 (corresponding to approximately ±0.5% accuracy).
[0064] Furthermore, the system introduces a method based on drilling speed. With motor current An adaptive parameter adjustment mechanism. When current is detected... violent fluctuations or drilling speed When a sudden change occurs (indicating a change in formation hardness or increased drill string vibration), the system automatically adjusts the covariance matrix of the process noise vector. Covariance of measurement noise The numerical values are used to balance the system's confidence weights between model predictions and sensor observations. After completing the linear estimation using Kalman filtering, the system inputs the state vector into a pre-trained ELM (Extremely Strong Learning Machine) neural network model to compensate for nonlinear errors using residuals, ultimately generating a fused deep sequence. The fusion depth sequence serves as the sole reliable depth benchmark for subsequent verticality correction and stratigraphic analysis.
[0065] Obtaining a highly reliable fusion depth sequence through a data fusion engine Subsequently, the industrial control center uses the tilt angle data provided by the attitude monitoring component to perform spatial geometric vector correction on the one-dimensional scalar depth and calculate the actual vertical depth and verticality deviation.
[0066] The industrial control center reads the X-axis tilt angle output from the tilt sensor 3 installed on the drill rig's power head or mast guide rail. (Front-back direction) and tilt angle with the Y-axis (Left and right directions). The system first calls the zero bias constant calibrated during the initial installation of the equipment, and then... and Algebraic calculations are performed to subtract installation errors, yielding the calibrated real-time tilt angle. Subsequently, based on the principle of spatial geometric projection, the system synthesizes the components of the two orthogonal directions into the total tilt angle of the drill pipe axis relative to the vertical line of gravity. .
[0067] Based on total tilt angle With fusion depth sequence The system constructs a verticality deviation correction model to eliminate depth measurement errors caused by drill pipe tilting and outputs the true vertical depth of the borehole. The verticality deviation correction model is expressed as follows: ; in: Indicates at time The corrected true vertical depth corresponds to the borehole depth. The final effective measurement value; Indicates at time The fused depth sequence along the drill pipe axis output by the data fusion engine (this depth integrates the winch displacement and the BeiDou trajectory length after tilt projection correction). Indicates at time The total inclination angle resulting from the combined inclination of the drill pipe axis is given by the formula. Or under small angle conditions Approximate calculations show that, Indicates at time The X-axis tilt angle (i.e., the angle of inclination of the drill pipe relative to the vertical line of gravity) is measured by the tilt sensor. Indicates at time The Y-axis tilt angle (i.e., the angle of inclination of the drill pipe in the left and right directions relative to the vertical line of gravity) is measured by the tilt sensor. The inverse cosine function is used to recover the total angle from the synthesized projected components; The term represents the geometric scaling factor that projects the length along the drill pipe path onto the direction of gravity perpendicular.
[0068] Meanwhile, in order to quantify the quality of borehole formation and support the subsequent early warning and grading mechanism, the system calculates the verticality deviation of the drill bit center relative to the pile center on the horizontal plane. The industrial control center uses trigonometric functions to calculate the horizontal eccentricity at the bottom of the drill bit, and defines the ratio of this eccentricity to the current depth as the verticality deviation.
[0069] The calculation logic for horizontal eccentricity is expressed as follows: The system generates verticality deviation based on this. This is used to determine whether the construction specification requirement of verticality being less than 1% is met, and serves as the input trigger signal for real-time correction of verticality deviation. When When the depth exceeds the system's preset threshold, the system marks the location as a quality anomaly and initiates the correction control logic.
[0070] See attached document Figure 7 Using the high-precision true vertical depth obtained after verticality correction Afterward, the system enters the geological parameter inversion and construction method adaptive control stage. The industrial control center uses a timestamp alignment mechanism to use depth data as the primary key and synchronously latches the main motor current of the power head, the mud pump flow rate, and the instantaneous speed of the drill bit at the same moment to construct a multi-dimensional construction state vector.
[0071] To extract the true lithological characteristics of the formation from the electrical signal, the system first performs a differential calculation of the drilling rate. The industrial control center performs first-order difference processing on the continuous depth sequence to calculate the instantaneous vertical downward drilling rate of the drill bit. Subsequently, based on a pre-set physical model, the system converts the motor current signal into a drilling resistance index that characterizes the hardness of the formation.
[0072] The system uses the following linear characteristic equation of formation resistance for real-time calculation: ; in: Indicates at time The drilling resistance characteristic value, which directly reflects the compressive strength and hardness of the current cutting formation; This represents the effective value of the main motor operating current of the drilling rig's power head, which is collected in real time by the Hall current sensor 8, i.e., the motor current. This represents the system's no-load reference compensation constant (takes a negative value). This constant term... Combining equipment mechanical friction and no-load current benchmarks, the calculation is pre-calibrated and automatically derived through a power head idling test. Its negative value attribute deducts the non-cutting current component generated by the equipment's own mechanical losses from the total current during the calculation, ensuring that the calculated drilling resistance characteristic value is accurate even when the drill bit is not in contact with the formation. Reset to zero; It represents the electromechanical conversion gain coefficient, which reflects the efficiency of the power head in converting electrical energy into cutting force.
[0073] Obtain the characteristic value of drilling resistance Then, the industrial control center correlates it with the instantaneous drilling speed. and mud flow rate The common input is the built-in formation identification and inference engine. This engine runs a classification model trained based on the random forest algorithm, and outputs the formation category label at the bottom of the borehole in real time by analyzing the coupling relationship between the frequency domain characteristics (such as variance) of current fluctuations and velocity changes.
[0074] Based on the real-time interpretation results of the formation type, the system automatically takes over control of the mud circulation system and executes the following closed-loop switching strategy: When the system identifies the strata as clay or silt (characteristics shown in...) The values are low and the fluctuations are stable, while (Maintaining a high level), the system determines that the drill cuttings are mainly fine particulate matter. At this time, the industrial control center sends a first state command to the solenoid valve control device 5. After receiving the command, the solenoid valve control device 5 controls the relay to engage, driving the electric three-way valve to switch to straight-through mode, establishing a positive circulation path for grouting directly into the drill pipe cavity from the mud pump. Simultaneously, the industrial control center sends control algorithm commands to the mud pump intelligent control system. The intelligent control system dynamically adjusts the mud pump operating parameters according to the commands, increases the frequency converter frequency to increase the pumping capacity, and uses high-pressure jets to impact the bottom of the hole, improving the mud-making and wall-protecting effect in soft soil layers.
[0075] When the system identifies that the formation has entered a sand or gravel layer (characteristically manifested as...), Significant fluctuations occurred at medium and high frequencies, and (A non-linear descent occurs), and the system determines that there are high-density particles of sediment at the bottom of the borehole. The industrial control center immediately sends a second state command to the solenoid valve control device 5, driving the valve group to switch direction to change the pipeline connection state, connecting the drill pipe cavity to the mud pump inlet. At the same time, the industrial control center instructs the mud pump intelligent control system to adjust the mud pump speed to match the pump suction reverse circulation condition. At this time, the mud flows from the outside to the inside of the borehole, and the velocity multiplication effect formed by the difference in pipeline cross-sectional area draws the pebbles and coarse sand at the bottom of the borehole into the drill pipe cavity and discharges them.
[0076] When the system identifies the strata as hard rock (characteristics are as follows) Continuously exceeding the high threshold setting, and (Extremely low pressure, accompanied by high drill pressure signal) – the system determines that it has entered the rock-cutting stage. The industrial control center triggers the air-lift reverse circulation mode, sends a start command to the control execution unit, activates the matching air compressor unit, and injects high-pressure gas into the bottom mixing chamber through the outer annulus of the double-walled drill pipe. The gas expansion reduces the fluid density inside the pipe, and the huge internal and external pressure difference generates a strong upward flow, ensuring that large pieces of rock cuttings can be effectively carried back under deep hole high-resistance conditions, preventing sediment from getting stuck in the drill bit.
[0077] After the system completes formation identification and adaptive adjustment of the circulation mode, in order to ensure the safety of drilling operations and the legal validity of construction data, the industrial control center starts the real-time monitoring of construction parameters and the encrypted storage of data in parallel.
[0078] The system first performs a multi-parameter coupling analysis to construct a model including borehole depth. Verticality deviation Loop mode state Drilling speed Motor current and mud flow rate The real-time state vector is input into the industrial control center for logical verification. The physical model rule base is established based on the construction physics mechanism, for example: Rule 1: When drilling at a constant speed in a uniform soil layer, the motor current... The mud flow rate should be kept relatively stable. Should be related to drilling speed Proportional relationship; Rule 2: In reverse circulation mode, drilling speed The threshold should not be exceeded to prevent poor slag removal from causing the drill pipe to tilt.
[0079] Based on the verification results of the above physical rules, the industrial control center executes a graded early warning response strategy according to the degree to which each parameter deviates from the preset benchmark threshold: When the fluctuation range of the monitored parameter is within ±5% and the duration is less than 5 minutes, the system determines it to be in a primary warning state. At this time, the industrial control center pushes a prompt message to the management personnel via the APP, requesting them to pay attention to the parameter trend; When the fluctuation range of the monitored parameters reaches ±10% and lasts for more than 10 minutes, the system determines it to be in a medium-level warning state. At this time, the industrial control center pushes a high-priority inspection command, prompting the operator to verify the equipment on-site; When monitored parameters severely exceed limits (such as verticality deviation exceeding the allowable value) or continuously exhibit abnormal fluctuations, the system determines a high-level warning state. The industrial control center immediately triggers a shutdown or correction protection procedure, sending an emergency shutdown command to the control execution unit. The control execution unit quickly cuts off the power input to the power head and mud pump to prevent the accident from escalating. The handling process includes closed-loop management of on-site verification, problem handling, and result confirmation.
[0080] Meanwhile, to ensure the authenticity and immutability of construction data, the industrial control center performs data encryption and blockchain-based evidence storage. The industrial control center uses the AES-256 encryption algorithm to encrypt and package the raw sensor data (such as BeiDou positioning coordinates and Hall effect sensor depth data) and processing results, which include timestamps.
[0081] The system establishes a construction data archive for each pile number, writing key construction parameters into the blockchain network. Specifically, the system maps the drilling depth record, verticality deviation record, and cycle mode switching record at each sampling moment to a unique blockchain hash value. During the completion acceptance or quality audit stage, regulators can verify whether the construction data is complete and unmodified by comparing the hash value stored in the cloud with the original onboard data, thus forming a traceable quality archive.
[0082] After completing deep data fusion, stratum identification, and security encryption in the industrial control center, the system constructs a two-way interactive communication link based on a microservice architecture in order to achieve visualized guidance at the construction site and digital control at the remote management end.
[0083] The industrial control center utilizes a built-in industrial-grade 4G / 5G communication module, employing the AES-256 encryption algorithm to encrypt and encapsulate raw data and processing results, including timestamps, and establishes a long-term connection with the cloud server using the MQTT IoT message transmission protocol. To adapt to the unstable network environments present at field construction sites, the communication module employs a breakpoint resumption mechanism. During network signal interruptions, data is automatically cached in local Flash memory, and the backlogged historical data is uploaded first upon network restoration, ensuring the integrity of the cloud database.
[0084] The human-machine interface terminal located at the construction site uses a high-brightness industrial touchscreen and runs customized visualization monitoring software. The visualization monitoring software interface constructs a dynamic virtual drilling model, and the screen interface is equipped with soft buttons for confirming the addition / reduction of drill rods. When drill rod extension work is carried out on site, the operator clicks this button, and the system automatically adds a drill rod variable of a preset length to the background geometric model, thereby updating the total drill rod length parameter used to calculate the difference between the Beidou antenna elevation and the bottom elevation of the hole, ensuring the continuity of depth calculation.
[0085] The left side of the interactive interface displays a geological bar chart, rendering in real-time the lithological stratification results output by the stratigraphic parameter identification algorithm. Different lithologies (such as clay, sand, and rock) are distinguished by different fill textures and colors. During the preparation phase before drilling operations begin, the interface guides the operator to perform a one-click zeroing calibration. When the bottom of the drill bit is aligned with the top of the casing or the ground reference, the operator clicks the confirmation button on the screen, and the system immediately captures the current BeiDou elevation and encoder values as the calculation origin.
[0086] During drilling, a virtual dashboard is set up on the interface to display the main motor current load rate and verticality deviation in real time. According to the system's built-in hierarchical early warning mechanism, the interface background color and alarm prompts follow the following display logic: When in the primary early warning state (parameter fluctuation ±5% and duration less than 5 minutes), a yellow warning prompt box pops up on the interface; when in the intermediate early warning state (parameter fluctuation ±10% and duration more than 10 minutes), an orange warning prompt box pops up on the interface and prompts for inspection; when in the advanced early warning state (parameters severely exceeding limits), a red alarm box pops up on the interface and is accompanied by a high-frequency buzzer sound.
[0087] The remote monitoring platform adopts a B / S architecture, allowing administrators to access the project management center via a web browser. The platform integrates a BIM (Building Information Modeling) data interface, supporting the spatial overlay of the BIM pile foundation model from the design phase with the returned real-time construction data. The system simultaneously renders the designed pile cylinder and the actual drilling trajectory in a 3D view, intuitively displaying the actual drilling depth. The difference from the design depth, and the verticality deviation Spatial deviation from the design axis. If the actual trajectory exceeds the model boundary of the designed pile body, the system automatically marks the deviation point in the 3D view.
[0088] In addition, the remote platform also features automated report generation. Based on hash-based evidence data stored on the blockchain network, the system automatically extracts the drilling time, drilling completion time, stratum lithology distribution records, and verticality test reports for each pile, generating a digital construction log that meets engineering acceptance standards. The platform supports data indexing on a pile-by-pile basis, allowing administrators to retrieve historical current variation curves and cycle mode switching records for any pile, enabling full-process traceability of construction quality.
Claims
1. An automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration, characterized in that, include: The data acquisition hardware terminal includes a fixed-height antenna (1), a Hall sensor (4), a current sensor (8), an tilt sensor (3), and a mud flow sensor (6). The data acquisition hardware terminal is used to collect drilling operation physical state data including Beidou elevation, winch rotation status, motor current, drill rod tilt angle, and mud flow. The mechanical control actuator includes a solenoid valve control device (5) and a mud pump intelligent control system. The mechanical control actuator is used to perform cycle mode switching and pumping parameter adjustment actions. An industrial control center is used to receive multi-source data collected by the data acquisition hardware terminal and execute intelligent depth iteration logic. The industrial control center is used to use the elevation data collected by the fixed-height antenna (1) and the wire rope displacement data collected by the Hall sensor (4) to perform data fusion to generate a hole depth sequence, and is used to use the tilt angle collected by the tilt sensor (3) to correct the hole depth sequence. The industrial control center is also used to identify the formation type based on the current data collected by the current sensor (8) and the real-time drilling speed, and to send control commands to the solenoid valve control device (5) based on the identified formation type, thereby automatically switching between positive circulation mode and negative circulation mode.
2. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, The fixed-height antenna (1) is rigidly fixed on the top of the drill bit power unit, the Hall sensor (4) is installed on the winch stator support, the current sensor (8) is clamped on the main power supply cable of the drill rig main motor, the tilt sensor (3) is installed on the bottom side of the drill rod or mast, and the mud flow sensor (6) is installed in the mud circulation pipeline. The data acquisition hardware terminal also includes positioning and directional antennas (2) installed at both ends of the drilling rig gantry beam, which are used in conjunction with the fixed-height antenna (1) to calculate the planar coordinates and heading angle of the drilling rig body; The phase center of the fixed-elevation antenna (1) is calibrated to the rotation center axis of the drill rod, and the fixed-elevation antenna (1) is used to sense the vertical displacement of the drill bit.
3. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, The Hall sensor (4) integrates a Hall element (7), which, together with a multi-turn primary coil, is used to detect the change in magnetic field generated when the wire rope moves and drives the main shaft to rotate, thereby detecting the number of rotations and the direction of rotation of the drum. The data acquisition hardware terminal also includes an absolute encoder, which is mechanically coupled to the main shaft of the winch via a coupling. The absolute encoder is used as a verification source for the data of the Hall sensor (4).
4. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 3, characterized in that, The specific method by which the industrial control center performs the intelligent deep iteration is as follows: A rigid geometric link model is constructed, and the real-time elevation data obtained by the fixed-elevation antenna (1) is combined with the elevation of the ground calibration point and the total length of the drill bit to generate the main depth sequence; A nonlinear winding model is constructed, and the release length of the wire rope is calculated by combining the data collected by the Hall sensor (4) with the drum layer diameter correction logic. Run the data fusion engine to perform weighted fusion of the main depth sequence and the wire rope release length to output a fused depth sequence; The tilt angle collected by the tilt sensor (3) is used to perform geometric correction on the fusion depth sequence, thereby generating the final hole depth sequence.
5. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 4, characterized in that, The data fusion engine adopts a layered fusion architecture, including: In the first stage of fusion, the industrial control center uses the Kalman filter algorithm to perform joint state estimation and data fusion processing on the two signals of the Hall sensor (4), thereby generating a displacement pulse sequence after signal-level fusion, and completing the first stage of fusion. In the second stage of fusion, the industrial control center establishes a state equation describing the vertical motion characteristics of the drill bit, and uses the master depth sequence, the wire rope release length, and the absolute encoder data as observation inputs. Through state space estimation and neural network model, nonlinear errors are compensated, thereby outputting calibrated depth data and completing the second stage of fusion.
6. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, When performing the correction, the industrial control center is also used to perform verticality geometric correction. Specifically, the verticality geometric correction is that the industrial control center uses trigonometric functions to convert the inclined length depth along the drill rod axis into the true vertical depth perpendicular to the ground plane based on the tilt angle of the drill rod axis relative to the gravity vertical line measured by the tilt sensor (3).
7. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, The industrial control center has a built-in physical model of formation resistance. The physical model of formation resistance is used to convert the main motor current collected by the current sensor (8) into a drilling resistance index. The industrial control center is also used to perform feature matching between the drilling resistance index, drilling speed and a preset geological parameter database, thereby identifying the formation type.
8. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, The solenoid valve control device (5) is connected to and drives the pipeline reversing valve group. The specific method by which the industrial control center controls the solenoid valve control device (5) is as follows: When the identification result is clay or silt layer, the industrial control center is used to send a positive circulation mode command to the solenoid valve control device (5), thereby driving the pipeline reversing valve group to connect the mud pump outlet and the drill pipe cavity; When the identification result is a sand layer or a pebble layer, the industrial control center is used to send a reverse circulation switching command to the solenoid valve control device (5), thereby driving the pipeline reversing valve group to switch the flow channel, connecting the drill pipe cavity and the mud pump inlet, and thus forming a negative pressure slag suction channel.
9. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 8, characterized in that, When the identification result is a hard rock layer, the industrial control center is also used to trigger the air lift reverse circulation mode. The industrial control center is used to send a start command to the mechanical control actuator to activate the matching air compressor unit and inject high-pressure gas into the bottom mixing chamber.
10. The automatic data acquisition system for forward and reverse circulation cast-in-place pile construction based on intelligent depth iteration as described in claim 1, characterized in that, The system also includes a remote data management terminal, which is used to receive encrypted sensor data and generate hash values for key quality data and write them into the blockchain node. The remote data management terminal is also used to reconstruct the construction process in the visualization interface based on the data of the mud flow sensor (6), the borehole trajectory, and the formation information.