Intelligent control system and method for mechanical method of removing obstacles of deep stratum steel reinforced concrete structure

Abstract

The application discloses a deep stratum type steel reinforced concrete structure mechanical method obstacle clearing intelligent control system and method; the system comprises a data sensing layer, a data transmission layer, a data processing layer and an execution control layer, and constitutes a closed loop control link; the data sensing layer collects multi-dimensional working condition data such as drilling pressure, torque, footage rate, rotary speed, mud parameters, pore water pressure and stratum displacement in real time through a multi-source sensing unit; the data processing layer is internally provided with a digital twin model and an intelligent decision engine, the model is updated in real time, and the drilling progress specific energy monitoring model, the mud-soil pressure dynamic balance model and the grouting amount prediction model are monitored according to the drilling progress specific energy, the grouting medium is automatically identified, and the horizontal drilling machine footage rate and rotary speed, the mud specific gravity and liquid level, and the grouting amount and grouting pressure are dynamically regulated and controlled; the execution control layer executes the regulation and control instructions; the application realizes intelligent sensing of the whole obstacle clearing process, real-time decision and collaborative optimization control driven by a physical model, and effectively improves the obstacle clearing efficiency, precision and safety.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for underground engineering construction, specifically to an intelligent control system and method for mechanically clearing obstacles in deep strata steel-concrete structures. It is particularly suitable for the intelligent control of mechanical clearing operations of deep steel-concrete obstacles such as underground continuous walls of existing stations in newly constructed shield tunnel sections of urban rail transit. Background Technology

[0002] With the continuous and in-depth construction of urban rail transit networks, the engineering conditions requiring newly built shield tunnel sections to pass through deep obstacles such as the underground continuous walls of existing stations are becoming increasingly common. The mechanical removal process for deep strata steel-concrete structures achieves fully mechanical removal of deep obstacles through a collaborative operation method of "vertical pipe sinking with ultra-large diameter steel casing + full-rotation horizontal drilling and core sampling." However, this process involves the coordinated operation of multiple complex procedures, including vertical pipe sinking, horizontal core sampling, synchronous grouting, hole-by-hole backfilling, and casing lifting. The working space is located in deep strata, with complex and variable working conditions, placing extremely high demands on the precise control of the construction process.

[0003] In existing technologies, the control of deep obstacle removal operations mainly suffers from the following problems: First, there is a lack of intelligent sensing methods throughout the entire process. In current obstacle removal operations, the verticality monitoring of steel casings is mostly done manually using theodolites for periodic measurements, the positioning of horizontal drilling rigs relies on traditional surveying and setting out, drilling parameters are recorded manually, and ground response monitoring and construction control are independent of each other. The working condition data of each link are collected in a scattered manner and stored in isolation, making it impossible to form a continuous, real-time, and collaborative perception of the entire obstacle removal operation process.

[0004] Secondly, there is a lack of real-time dynamic decision-making capabilities based on physical models. In existing technologies, the adjustment of construction parameters (such as drilling rig advance rate, rotation speed, and grouting pressure) mainly relies on the experience and judgment of on-site technicians, lacking support from real-time data-driven physical models. Especially when horizontal drilling rigs are cutting reinforced concrete structures with H-shaped steel joints, parameters such as drilling pressure and torque can undergo drastic changes. Traditional experience-based adjustments are slow to respond, easily causing abnormal tool wear or even stuck drill accidents. Furthermore, the stability control of borehole walls near confined aquifers lacks quantitative safety assessment indicators, and the determination of grouting volume lacks precise calculation basis.

[0005] Third, the two major processes of vertical casing installation and horizontal coring lack coordinated control. In the existing process, the vertical casing installation and the horizontal coring and obstacle removal by the drilling rig are controlled independently—vertical casing installation controls the casing's sinking speed and verticality, while horizontal coring controls the drilling parameters and positioning accuracy of the drilling rig. There is a lack of information exchange and coordinated control mechanisms based on a unified data model between the two processes, making it impossible to achieve coordinated control that "guides casing lifting based on coring progress and coordinates grouting adjustment based on formation response."

[0006] Fourth, there is a lack of forward-looking early warning capabilities based on data fusion. Current technologies mostly assess the settlement impact of adjacent existing structures after the fact, failing to achieve real-time prediction and proactive control during clearance operations. By the time settlement exceeds limits, irreversible damage has often already occurred, making it difficult to meet the safety requirements for close-proximity construction in urban core areas.

[0007] Some existing technologies have proposed improvement solutions. For example, Chinese patent CN119877582A discloses an intelligent control method and device for the overall lowering planar position of a steel casing, which realizes intelligent control of the lowering of the steel casing through a control terminal, a monitoring module, and an intelligent adjustment module. Chinese patent CN117514128B discloses an intelligent horizontal directional drilling control method for coal mines, which achieves trajectory correction by acquiring borehole trajectory parameters and comparing them with a reference trajectory.

[0008] Of the aforementioned existing technologies, CN119877582A only addresses the planar positioning and attitude adjustment during the overall lowering of the steel casing, controlling a single vertically moving component and not involving the perception and adaptive control of the cutting state of the horizontal drilling rig in deep strata; CN117514128B only achieves trajectory correction for horizontal directional drilling in coal mines, controlling the spatial coordinate deviation of the borehole trajectory and not involving online identification of the steel-concrete composite medium and dynamic switching of cutting parameters. Neither of these technologies can achieve online identification of the steel-concrete composite medium, provide adaptive control of cutting parameters based on sudden changes in drilling specific energy, or establish a dynamic balance mechanism for mud-water-soil pressure and a collaborative compensation mechanism for core sampling and grouting. Therefore, these existing technologies are not applicable to the multi-process, highly coupled, and high-risk operational scenario of mechanical obstacle removal for deep steel-concrete structures. Summary of the Invention

[0009] The main objective of this invention is to provide an intelligent control system and method for mechanical obstacle removal in deep strata steel-concrete structures, aiming to solve existing technical problems.

[0010] To achieve the above objectives, the present invention provides an intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures, comprising: The data perception layer includes multi-source sensing units deployed in the obstacle clearing operation area, used to collect multi-dimensional working condition data in real time during the obstacle clearing operation, including at least the drilling pressure, torque, advance rate and rotation speed of the horizontal drilling rig, mud specific gravity and the depth of the mud surface from the ground, as well as pore water pressure and formation displacement data. The data transmission layer is communicatively connected to the data perception layer and is used to transmit the multi-dimensional working condition data to the data processing layer in real time. The data processing layer is communicatively connected to the data transmission layer and has a built-in digital twin model and intelligent decision engine. The digital twin model is updated in real time based on the multi-dimensional working condition data to map the real-time physical state of the obstacle removal operation; The intelligent decision engine is configured as follows: Based on the aforementioned drill pressure, torque, drilling rate, and rotation speed, the drilling specific energy real-time monitoring model calculates the cutting energy consumption per unit volume and automatically determines the current cutting medium type according to its deviation from the preset stable range. When it is determined that the cutting medium is in the form of steel cutting, a first control command is generated to reduce the drilling rate and increase the rotation speed. When it is determined that the cutting medium is not in the form of steel cutting and the drilling specific energy is lower than the lower limit of the preset range, a second control command is generated to increase the drilling rate. Based on the specific gravity of the mud, the depth of the mud surface from the ground, and the pore water pressure, the safety factor of the borehole wall stability is calculated using the borehole wall stability safety factor calculation model. When the safety factor is lower than a preset threshold, a third control command is generated to increase the specific gravity of the mud or raise the mud surface. Based on the measured volume of the extracted core samples and the formation loss volume inverted from the digital twin model, the synchronous grouting volume is calculated using the grouting volume prediction model, and a fourth control command is generated accordingly. The execution control layer is communicatively connected to the data processing layer and is used to receive and execute the first to fourth control commands to adjust the drilling speed and rotation speed of the horizontal drilling rig, the mud specific gravity and the depth of the mud surface from the ground, and the output of the grouting system in real time. The data perception layer, data transmission layer, data processing layer, and execution control layer constitute a closed-loop control link.

[0011] Furthermore, the real-time drilling specific energy monitoring model calculates the cutting energy consumption per unit volume SE according to the following formula: ; In the formula, P is the drilling pressure and T is the torque. ω is the drilling rate, A is the rotation speed, and A is the cross-sectional area of ​​the core cylinder of the horizontal drilling rig.

[0012] Furthermore, the hole wall stability safety factor calculation model calculates the hole wall stability safety factor K according to the following formula: ; In the formula, The specific gravity of the mud. The vertical height of the mud column is determined by the current depth of the clearing operation surface and the depth of the mud surface from the ground. Joint decision, Pore ​​water pressure, The active earth pressure at the bottom of the casing horizontal core drilling working face is calculated in real time based on Rankine earth pressure theory.

[0013] Furthermore, the grouting volume prediction model calculates the synchronous grouting volume according to the following formula. : ; In the formula, This represents the measured total volume of the core sample that has been removed. The formation loss volume is calculated based on the formation displacement field inversion in the digital twin model, where α is the filling compensation coefficient of the core sample volume and β is the advance grouting compensation coefficient of the formation loss.

[0014] Furthermore, the multi-source sensing unit includes: The vertical tube sinking attitude sensing subunit is deployed on the ultra-large diameter steel casing and is used to collect data on the verticality, planar position and sinking depth of the steel casing. The horizontal drilling rig attitude sensing subunit is deployed on the full-rotation horizontal drilling rig and is used to collect the spatial position, drilling angle and drill bit attitude data of the horizontal drilling rig. The drilling parameter sensing subunit is installed on the drive system of the horizontal drilling rig and is used to collect data on drilling pressure P, torque T, and drilling speed during the core sampling operation. and rotational speed ω data; The formation response sensing subunit is deployed in the surrounding strata and adjacent existing structures of the cleared area to collect data on formation displacement and pore water pressure. and existing structural deformation data; The mud parameter sensing subunit is deployed in the mud circulation system inside the steel casing and is used to collect mud specific gravity. Depth of mud surface from the ground And sand content data.

[0015] Furthermore, the vertical tube attitude sensing subunit includes dual theodolites arranged at a 90° angle along the circumference of the steel casing, used to achieve real-time cross-monitoring of the verticality of the steel casing; the data processing layer calculates the real-time horizontal offset E of the center point at the bottom of the casing according to the following formula: ; in Let be the length of the i-th stage casing segment. The angle between the measured axis of the i-th stage casing section and the designed plumb line; the comprehensive plane deviation of the dual theodolite cross-monitoring system. Solve using the following formula: ; in , The real-time planar coordinates of the top center of the casing are obtained based on the forward intersection method. , To design the center coordinates of the borehole; the horizontal drilling rig attitude sensing subunit includes a positioning and orientation instrument installed on the horizontal drilling rig. The positioning and orientation instrument has a built-in inertial measurement unit and a Kalman filter for realizing the autonomous navigation and positioning of the horizontal drilling rig.

[0016] Furthermore, the digital twin model includes: A geological model of the clearing area, constructed based on geological survey data, is used to characterize the stratigraphic distribution and hydrogeological features of the clearing area. The obstacle removal equipment model is constructed based on the geometric and kinematic parameters of the obstacle removal equipment and is used to characterize the spatial pose and motion state of the obstacle removal equipment. The obstacle clearing operation process model is constructed based on the temporal and spatial logic of the obstacle clearing process, and is used to characterize the process progress and spatial coverage of the obstacle clearing operation. An existing structural model, constructed based on the structural parameters of adjacent existing structures, is used to characterize the boundary conditions of the impact of the obstacle removal operation on the existing structures.

[0017] Furthermore, the intelligent decision engine includes: The obstacle clearing status assessment module is used to assess the completion status, obstacle clearing quality, and equipment health status of the current obstacle clearing operation based on the multi-dimensional working condition data and the digital twin model. The risk warning module is used to identify abnormal working conditions during the obstacle removal operation and issue warning signals. The parameter optimization module is used to dynamically optimize the operating parameters of the obstacle clearing operation based on the output results of the obstacle clearing status assessment module and the risk warning module, including at least optimizing the advance rate based on drill pressure P and torque T. With rotational speed ω, based on formation displacement and pore water pressure Optimize grouting pressure and grouting volume Optimize mud specific gravity based on mud parameters and liquid level depth And generate steel casing correction instructions based on verticality deviation; The path planning module is used to automatically plan the next core drilling position and the drilling rig movement path based on the honeycomb hole layout scheme of the cleared area and the current clearing progress.

[0018] Furthermore, the execution control layer includes: The vertical pipe sinking execution unit is connected to the hydraulic drive system of the pipe rolling machine and is used to adjust the sinking speed, the action of the correction cylinder and the timing of the sleeve extension according to the control command. The horizontal drilling rig execution unit, connected to the drive system of the full-rotation horizontal drilling rig, is used to adjust the rotational speed ω and the feed rate of the horizontal drilling rig according to the control commands. Core sampling depth and grouting pressure; The mud circulation execution unit, connected to the mud preparation and circulation system, is used to adjust the mud specific gravity according to the control command. and the depth of the mud surface from the ground .

[0019] A mechanical obstacle removal intelligent control method for deep strata steel-concrete composite structures, employing the aforementioned system, includes the following steps: S1: Real-time acquisition of multi-dimensional working condition data during the obstacle clearing operation through multi-source sensing units; S2: The multi-dimensional operating condition data is transmitted to the data processing layer in real time through the data transmission layer; S3: Based on the multi-dimensional working condition data, the built-in digital twin model is updated in real time to keep the digital twin model synchronized with the actual state of the construction site; S4: The updated digital twin model is analyzed and calculated using an intelligent decision engine to generate obstacle clearing operation control instructions; the analysis and calculation include at least: calculating the unit volume cutting energy consumption SE based on the real-time drilling energy ratio monitoring model; when SE exceeds the upper limit of the preset stable benchmark range, it is determined to be a steel cutting condition and the drilling rate is reduced. Furthermore, increase the rotational speed ω, and increase the feed rate when SE is below the lower limit. The safety factor K is calculated based on the borehole wall stability safety factor calculation model. When K is less than 1.05, the mud density is increased. Alternatively, raise the mud level; calculate the synchronous grouting volume based on the grouting volume prediction model. ; S5: The control layer receives and executes the control instructions to adjust the operating parameters of the obstacle removal equipment in real time; S6: Repeat steps S1 to S5 to form a closed-loop control until the obstacle removal operation is completed.

[0020] The beneficial effects of this invention are reflected in: This invention utilizes multi-source sensing units deployed in the steel casing, horizontal drilling rig, surrounding strata, and mud circulation system to achieve comprehensive, real-time, and collaborative sensing of vertical casing attitude, horizontal drilling rig position, drilling parameters, strata response, and mud parameters. Compared to existing monitoring methods that rely on decentralized data collection and isolated storage, this invention's multi-source sensing system provides a comprehensive, continuous, and accurate data foundation for intelligent control of obstacle removal operations.

[0021] This invention incorporates five core calculation models into its data processing layer: a real-time monitoring model for drilling specific energy, a dynamic balance calculation model for mud-water-soil pressure, a formation loss compensation and grouting volume prediction model, a cumulative cutting energy index evaluation model, and an improved Peck settlement prediction model. This enables scientific quantitative decision-making based on real-time data. Compared to existing technologies that rely on manual experience for control, this invention offers higher decision-making efficiency, better control accuracy, and faster response speed. When a horizontal drilling rig cuts into an H-beam, the drilling specific energy... Sudden changes can be captured in real time and automatically trigger parameter adjustments, avoiding tool overload damage; when close to confined aquifers, the borehole wall stability safety factor is [not specified]. Real-time tracking and automatic compensation ensure the safety of the working face from a quantitative perspective.

[0022] This invention constructs a four-dimensional digital twin model comprising a geological model, an equipment model, an operational process model, and an existing structural model. Real-time operational data drives the digital twin model to be updated synchronously with the construction site, achieving digital mapping and real-time situational awareness of the entire obstacle removal operation process. In particular, by fusing measured settlement data with Peck formula predictions using Kalman filtering, the amount of ground loss is accurately measured. and settling trough width coefficient Real-time correction enables digital twin models to have self-learning and self-evolution capabilities.

[0023] This invention achieves collaborative control of the vertical tunneling system and the horizontal coring system based on a unified data model through a closed-loop control link of "sensing-transmission-analysis-regulation." This involves using the real-time progress of horizontal coring to guide the timing and height of steel casing lifting, using real-time ground response data to collaboratively adjust grouting pressure and mud parameters, and using a grouting volume prediction model. The precise calculation of the grouting volume for each cycle enabled the coordinated optimization of the two major processes, significantly improving the overall efficiency and safety of the obstacle removal operation.

[0024] This invention, while achieving fully automatic intelligent control, retains the human-machine interaction layer for manual intervention, and prioritizes manual control commands over automatically generated ones. This fully leverages the efficiency and precision of the intelligent control system while preserving the flexibility of human experience in complex working conditions, thus achieving an organic integration of technological rationality and experiential wisdom.

[0025] The system architecture of this invention adopts a modular design, with standardized interfaces and loose coupling between the data perception layer, data transmission layer, data processing layer, and execution control layer, allowing for flexible adaptation to obstacle removal projects of different scales and geological conditions. The digital twin model and intelligent decision engine in the system have excellent configurability and can be quickly customized according to the specific geological conditions, equipment parameters, and process requirements of the project. It is suitable for various deep underground engineering scenarios such as urban rail transit construction, comprehensive underground space development, underground floor additions to existing buildings, and geological disaster management. Attached Figure Description

[0026] Figure 1 This is a diagram illustrating the overall architecture of the intelligent control system for mechanical obstacle removal in deep strata steel-concrete structures according to the present invention. Figure 2 This is a schematic diagram of the multi-source sensing unit layout of the present invention; Figure 3 This is a schematic diagram of the digital twin model structure of the present invention; Figure 4 This is a schematic diagram of the intelligent decision engine module of the present invention; Figure 5 This is a flowchart of the closed-loop control process of "sensing-transmission-analysis-regulation" in this invention. Detailed Implementation

[0027] The technical solutions of 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 a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. 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.

[0028] like Figure 1 As shown, the present invention provides an intelligent control system for mechanical obstacle removal of steel-concrete composite structures in deep strata, which consists of a data perception layer, a data transmission layer, a data processing layer, an execution control layer, and a human-computer interaction layer.

[0029] The data perception layer consists of multi-source sensing units deployed in the obstacle removal operation area, responsible for real-time collection of multi-dimensional operational data during the obstacle removal process. The data transmission layer communicates with the data perception layer, employing a combination of wired and wireless transmission to transmit the multi-dimensional operational data to the data processing layer in real time. The data processing layer, also communicating with the data transmission layer, incorporates a digital twin model and an intelligent decision engine. It is responsible for updating the digital twin model in real time based on the multi-dimensional operational data and analyzing and calculating the updated model through the intelligent decision engine to generate obstacle removal operation control commands. The execution control layer, communicating with the data processing layer, is responsible for receiving and executing control commands, adjusting the operating parameters of the obstacle removal equipment in real time. The data perception layer, data transmission layer, data processing layer, and execution control layer form a closed-loop control link.

[0030] The human-computer interaction layer communicates with the data processing layer, including a visualization unit and a manual intervention unit. The visualization unit displays the real-time update status of the digital twin model and the progress of obstacle removal operations in a 3D visualization format. The manual intervention unit allows operators to input manual control commands, which have higher priority than the control commands automatically generated by the intelligent decision engine. The communication protocols between these layers adopt standardized industrial Ethernet and 5G wireless communication protocols to ensure real-time interconnectivity and data exchange between the system layers.

[0031] like Figure 2 As shown, the multi-source sensing unit includes a vertical pipe-sinking attitude sensing subunit, a horizontal drilling rig attitude sensing subunit, a drilling parameter sensing subunit, a formation response sensing subunit, and a mud parameter sensing subunit. Specifically: The vertical immersed tube attitude sensing subunit is deployed on the ultra-large diameter steel casing to collect data on the casing's verticality, planar position, and sinking depth. Specifically, the subunit includes two theodolites arranged at a 90° angle along the circumference of the steel casing. These theodolites are communicatively connected to the data processing layer to achieve real-time cross-monitoring of the steel casing's verticality. The dual theodolite cross-monitoring system eliminates systematic errors from single-point measurements, and its comprehensive planar deviation Δ xy The solution is obtained by forward intersection method.

[0032] Meanwhile, a high-precision tilt sensor is installed every 3m along the axial direction on the inner wall of the steel casing. The tilt sensors are connected in series via a bus to collect the axial tilt angle of each casing section in real time.

[0033] The horizontal drilling rig attitude sensing subunit is deployed on the full-rotation horizontal drilling rig to collect data on the rig's spatial position, drilling angle, and drill bit attitude. Specifically, the subunit includes a positioning and orientation instrument mounted on the rig. This instrument incorporates a high-precision inertial measurement unit (IMU) and an embedded Kalman filter to enable autonomous navigation and positioning of the horizontal drilling rig in deep geological formations. The Kalman filter integrates IMU data and periodically calibrated data from a ground total station, controlling the cumulative positioning error to within millimeter levels.

[0034] The drilling parameter sensing subunit is installed on the drive system of the horizontal drilling rig and is used to collect data on drilling pressure P, torque T, and drilling speed v during the core sampling operation. f And the rotational speed ω data. Specifically, pressure sensors, torque sensors, and displacement sensors are installed on the hydraulic drive system of the horizontal drilling rig to collect drilling parameters in real time, with a sampling frequency of not less than 100 Hz. Here, ω represents the rotational speed of the horizontal drilling rig, measured in revolutions per second (rpm).

[0035] The formation response sensing subunits are deployed in the surrounding strata and adjacent existing structures of the cleared area to collect data on formation displacement and pore water pressure. w And deformation data of existing structures. Specifically, displacement gauges, earth pressure cells and pore water pressure gauges are deployed in layers in the strata surrounding the cleared area, and static levels and surface strain gauges are deployed on the surface of adjacent existing structures, with a data acquisition frequency of not less than 10 Hz.

[0036] The mud parameter sensing subunit is deployed in the mud circulation system inside the steel casing to collect the mud specific gravity γ. mud The depth h of the mud surface from the ground mud And sand content data. Specifically, a mud hydrometer, a level gauge, and a sand content detector are installed in the mud circulation pipeline, with a data acquisition frequency of not less than 1 Hz.

[0037] The data transmission layer employs a combination of wired and wireless transmission. Fixed sensors (including most sensors in the formation response sensing subunit, the tilt sensor in the vertical pipe attitude sensing subunit, and the mud parameter sensing subunit) use industrial Ethernet wired transmission with Modbus TCP or PROFINET protocols and a transmission rate of at least 100Mbps. Moving sensors (including the positioning and orientation instrument in the horizontal drilling rig attitude sensing subunit and the drilling parameter sensing subunit) use 5G wireless transmission, equipped with an industrial-grade 5G communication module, supporting SA networking mode, with an uplink rate of at least 50Mbps and data transmission latency controlled within 100ms. The data transmission layer also includes a data aggregation gateway, responsible for protocol conversion, time synchronization, and preliminary preprocessing of the data collected by each sensing subunit, forming a unified data stream before uploading it to the data processing layer.

[0038] The data processing layer is deployed on an edge computing server at the construction site. The data processing layer incorporates a digital twin model and an intelligent decision-making engine, and its specific implementation is as follows.

[0039] like Figure 3 As shown, the digital twin model includes a geological model of the clearing area, a model of the clearing equipment, a model of the clearing operation process, and a model of the existing structure. Specifically: The geological model of the cleared area is constructed based on geological survey data and is used to characterize the stratigraphic distribution and hydrogeological features of the cleared area. Specifically, based on borehole columnar sections, standard penetration test data, laboratory geotechnical test data, and hydrogeological test data, a three-dimensional stratigraphic model of the cleared area is established using three-dimensional geological modeling software. The model includes at least the spatial distribution, physical and mechanical parameters, and hydrogeological parameters of each soil layer. The model uses tetrahedral meshing, with the mesh size refined to 0.2m within a 5m radius of the cleared well and gradually transitioning to 2m in the far field.

[0040] The obstacle removal equipment model is constructed based on the geometric and kinematic parameters of the obstacle removal equipment, and is used to characterize the spatial pose and motion state of the obstacle removal equipment. Specifically, the steel casing model is constructed based on the geometric parameters such as the diameter, wall thickness, and single section length of the steel casing, and the kinematic parameters such as the sinking speed range and maximum correction force of the pipe rolling machine; the horizontal drilling rig model is constructed based on the geometric and kinematic parameters such as the diameter, length, maximum rotation speed, and maximum advance rate of the horizontal drilling rig.

[0041] The obstacle removal operation process model is constructed based on the temporal and spatial logic of the obstacle removal technology, and is used to represent the process progress and spatial coverage of the obstacle removal operation. Specifically, based on the obstacle removal process design documents, the model establishes the temporal logic relationships of processes including vertical pipe sinking, horizontal coring, synchronous grouting, hole-by-hole backfilling, and casing lifting, as well as the honeycomb spatial arrangement and coring sequence logic of each coring hole. After each coring hole is completed, the model automatically updates the completion status and obstacle removal quality level of that hole.

[0042] The existing structure model is constructed based on the structural parameters of adjacent existing structures and is used to characterize the boundary conditions of the impact of the clearing operation on the existing structures. Specifically, based on the as-built drawings and structural inspection reports of the existing structures, a three-dimensional geometric model and a structural parameter model of the adjacent existing structures are established, including structural form, material strength, foundation form, and spatial relationship with the clearing area.

[0043] The specific update methods for the digital twin model are as follows: update the spatial pose of the steel casing in the obstacle removal equipment model based on the vertical immersed tube attitude data; update the spatial pose and drilling trajectory of the horizontal drilling rig in the obstacle removal equipment model based on the horizontal drilling rig pose data; update the completion status and obstacle removal quality of the current core hole in the obstacle removal operation process model based on the drilling parameter data; and update the stress-strain state of the geological model and the existing structural model of the obstacle removal area based on the formation response data.

[0044] like Figure 4 As shown, the intelligent decision-making engine includes a clearing status assessment module, a risk warning module, a parameter optimization module, and a path planning module. Specifically: The obstacle clearing status assessment module is used to assess the completion status, clearing quality, and equipment health status of the current obstacle clearing operation based on multi-dimensional working condition data and digital twin models.

[0045] Completion status assessment: Calculate the ratio of the number of completed core holes to the total number of holes, and calculate the overall progress percentage of the obstacle removal operation; at the same time, assess the drilling progress of the current core hole, including the ratio of the current depth of drilling to the designed core depth.

[0046] Obstacle removal quality assessment: Based on indicators such as the stability of drilling specific energy (SE), core sample integrity rate, and borehole wall stability during core sampling, the obstacle removal quality level of each borehole location is comprehensively assessed.

[0047] Equipment health status assessment: The real-time wear status of the drill bit is assessed based on the cumulative cutting energy index (WI), the working status of the hydraulic drive system is assessed based on the statistical characteristics of drilling pressure (P) and torque (T), and the working efficiency of the correction system is assessed based on the changing trend of perpendicularity deviation.

[0048] The risk warning module is used to identify abnormal working conditions during the obstacle clearing operation and issue warning signals based on multi-dimensional working condition data and digital twin models.

[0049] The warning rules include: (1) When the verticality deviation of the steel sleeve exceeds 1 / 500, or the plane position deviation exceeds 10mm, a yellow warning is triggered; when the deviation continues to increase and correction is ineffective, an orange warning is triggered. (2) A yellow warning is triggered when the positioning deviation of the horizontal drilling rig exceeds the design allowable value; (3) When the stratum displacement exceeds 80% of the safety threshold, a yellow warning is triggered; when it exceeds the safety threshold, a red warning is triggered and the machine is automatically shut down. (4) Pore water pressure u w A yellow alert is triggered when the fluctuation range exceeds ±20% of the stable value; an orange alert is triggered when it exceeds ±30%. (5) Specific gravity of mud γ mud A yellow alert is triggered when the value exceeds the range of 1.10-1.20. (6) When the drilling pressure P or torque T exceeds 90% of the equipment's rated value, a yellow warning is triggered; when it exceeds the rated value, a red warning is triggered and the machine is automatically shut down. (7) When the real-time value of the drilling energy SE exceeds the upper and lower limits of the preset stable reference range, a yellow warning is triggered and the cutting parameters are automatically switched.

[0050] The warning signals are divided into three levels: yellow, orange, and red. Each level of warning signal is presented through a visual display unit in the human-computer interaction layer, using different colors and sound and light alarms.

[0051] The parameter optimization module is used to dynamically optimize the operating parameters of the obstacle clearing operation based on the output results of the obstacle clearing status assessment module and the risk warning module. Specifically, it includes: (1) Based on real-time drilling pressure P and torque T data, the unit volume cutting energy consumption SE is calculated using the real-time drilling specific energy monitoring model. When SE exceeds the upper limit of the preset stable benchmark range, it is determined to be a steel cutting condition, and the drilling speed v of the horizontal drilling rig is optimized. f With rotational speed ω: Automatically reduce the feed rate and increase the rotational speed; when SE is below the lower limit of the preset range, automatically increase the feed rate to improve obstacle clearing efficiency; (2) Based on real-time formation displacement and pore water pressure u w Data was used to optimize the relationship between grouting pressure and grouting volume Q using a grouting volume prediction model. inj The dynamic adjustment range of grouting pressure is 0.2-0.4 MPa, and the specific value is adjusted in real time according to the formation displacement rate and the rate of change of pore water pressure. (3) Based on real-time mud parameters, the mud specific gravity γ is optimized using a borehole wall stability safety factor calculation model. mud and the depth h of the mud surface from the ground mud Maintain the mud specific gravity within the range of 1.10-1.20, and maintain the mud surface depth from the ground within the range of 0.3-0.5m; (4) Generate steel casing correction command based on real-time verticality deviation data: When the increase rate of horizontal offset E exceeds the preset gradient threshold or the plane deviation Δxy>10mm, automatically calculate the compensation thrust required by the correction cylinder according to the direction of the deviation vector.

[0052] The path planning module automatically plans the next coring hole location and the drilling rig's movement path based on the honeycomb-patterned borehole layout and the current progress of the obstacle removal. Specifically, based on the preset spatial coordinates and sequence constraints of the coring holes in the honeycomb-patterned layout, combined with the distribution of currently completed holes and the current spatial pose of the horizontal drilling rig, the RRT algorithm is used to calculate the optimal movement path from the current position to the target hole location, avoiding completed holes and known obstacles in the formation during the planning process. Simultaneously, the path planning module also has dynamic replanning capabilities; when hole location deviations or changes in formation conditions occur during obstacle removal, the coring sequence and movement path can be replanned within seconds.

[0053] The data processing layer also includes five core computational models, used for real-time monitoring of drilling specific energy, drill bit wear assessment, calculation of borehole wall stability safety factor, grouting volume prediction, and prediction of existing structure settlement. The mathematical expressions and physical meanings of each model are as follows.

[0054] The real-time monitoring model for drilling specific energy is used to calculate the cutting energy consumption (SE) per unit volume. The calculation formula is as follows: ; The meanings of each parameter in the formula are as follows: P is the drilling pressure, in kN, which is collected in real time by the pressure sensor of the drilling parameter sensing subunit; T is the torque, in kN·m, which is collected in real time by the torque sensor of the drilling parameter sensing subunit. ω is the drilling speed, in m / s, obtained by differentiating the displacement sensor of the drilling parameter sensing subunit with respect to time; ω is the rotation speed, in revolutions per second, acquired in real time by the rotation speed sensor of the drilling parameter sensing subunit; A is the cross-sectional area of ​​the core barrel of the horizontal drilling rig, in m², calculated based on the outer diameter of the core barrel.

[0055] The physical meaning of drilling specific energy (SE) is the mechanical energy consumed in cutting a unit volume of obstacle, expressed in MPa or J / cm³. SE has different characteristic value ranges for different media: in soft soil, SE is typically 0.5-1.5 MPa; in reinforced concrete, SE is typically 2.0-3.5 MPa; and in structural steel, SE is typically 4.0-6.0 MPa. Therefore, by monitoring the real-time value of SE and its changing trend, the type of cutting medium can be accurately determined.

[0056] The method for determining the preset stable benchmark interval is as follows: Before the obstacle clearing operation begins, test drilling is conducted in representative strata to statistically analyze the average SE and standard deviation of each typical stratum. The stable benchmark interval is set within ±30% of the average SE of each typical stratum. When the real-time SE exceeds the upper limit of this interval, it is determined that the high-strength steel cutting section has been entered; when the real-time SE is lower than the lower limit of this interval, it is determined that the section is a weak stratum.

[0057] The cumulative cutting energy index evaluation model is used to evaluate the real-time wear state of drill bits, and its calculation formula is as follows: ; The meanings of each parameter in the formula are as follows: WI is the dimensionless cumulative wear index; The rated cutting power of the drill bit, in kW, is provided by the drill bit manufacturer. The equivalent wear coefficient for drilling pressure is dimensionless and calibrated through indoor wear tests, with a typical value of 0.3-0.6. The equivalent wear coefficient of torque is dimensionless and is calibrated through indoor wear tests, with a typical value of 0.4-0.7. This is the start time of the current core extraction cycle; This represents the current time of the current core extraction cycle.

[0058] The physical meaning of WI is the ratio of the cumulative cutting energy consumed by the drill bit to the rated cutting energy, reflecting the cumulative wear degree of the drill bit. When WI reaches the preset tool life threshold, it indicates that the drill bit is nearing the end of its service life, and the risk warning module automatically issues a drill bit replacement warning signal.

[0059] The calculation model for the safety factor of borehole wall stability is used to quantitatively assess borehole wall stability. The formula is as follows: ; The meanings of each parameter in the formula are as follows: The specific gravity of the mud is measured in kN / m³ and is collected in real time by the mud hydrometer of the mud parameter sensing subunit. The vertical height of the mud column is in meters (m), determined by the current depth (h) of the clearing operation surface. face The depth h of the mud surface from the ground mud Joint decision: h mud =h face -h mud ; The pore water pressure is measured in kPa and is collected in real time by the pore water pressure gauge of the formation response sensing subunit. The active earth pressure at the bottom of the casing horizontal core drilling face, in kPa, is calculated in real time based on Rankine earth pressure theory. Pa = γ·h·K a 2c ; Where γ is the unit weight of the soil layer, h is the calculation depth, c is the cohesion, and Ka is the active earth pressure coefficient (K_a=tan²(45°-φ / 2), φ is the internal friction angle).

[0060] The physical meaning of the borehole wall stability safety factor K is the ratio of the lateral support force of the mud slurry to the lateral pressure of the formation, reflecting the stability of the borehole wall. When K > 1, the borehole wall is in a stable state; when K = 1, the borehole wall is in a critically stable state; when K < 1, the borehole wall is in an unstable state. The preset safety threshold of this invention is K. threshold =1.05, the target safe range is 1.10-1.25.

[0061] The grouting volume prediction model is used to calculate the synchronous grouting volume. The formula for its calculation is: ; The meanings of each parameter in the formula are as follows: The measured total volume of the core sample taken out is in m³, calculated based on the diameter and length of the measured core sample, including steel sections and reinforced concrete. The volume of formation loss is calculated based on the formation displacement field inversion in the digital twin model, in m³, and is obtained by integrating the formation displacement field in the digital twin model; α is the filling compensation coefficient of the core sample volume, which is taken as 1.1-1.3, and is used to compensate for the volume shrinkage of the grouting material during the solidification process and the micro-expansion of the borehole wall after the core sample is removed; β is the advance grouting compensation coefficient of formation loss, which is taken as 0.6-0.9, and is used to compensate for the formation loss caused by formation disturbance and stress release caused by the core sampling operation.

[0062] The physical meaning of α and β is the volume of grout required to compensate for the physical cavities formed during coring and the losses due to formation disturbance. The specific values ​​of α and β are determined based on geological conditions and the characteristics of the grouting material: higher values ​​are used in soft formations and lower values ​​are used in hard formations; higher values ​​are used when using cement-based grouting materials and lower values ​​are used when using chemical grouts.

[0063] The improved Peck settlement prediction model is used to predict formation loss V based on real-time inversion. loss The formula for predicting the real-time settlement of adjacent existing structures is as follows: ; ; The meanings of each parameter in the formula are as follows: The horizontal distance from the center of the obstacle removal well, in meters; The value is the settlement trough width coefficient, in meters, and is dynamically determined based on the geological model. The value range is generally 0.5h-1.0h. The formation loss is the amount of formation loss retrieved in real time by multi-source sensing units, in m³, and is calculated by integrating the formation displacement field in the digital twin model. The settlement at a horizontal distance x from the center of the clearance well; This represents the maximum settlement.

[0064] Specifically, the intelligent decision engine utilizes Kalman filtering to analyze the measured settlement S. obs Data fusion with the predicted value S(x) is performed to correct formation loss in real time. And the settling trough width coefficient i. The specific implementation of Kalman filtering is as follows: using Peck's formula as the state equation and measured settlement as the observation value, the Kalman filtering recursive formula is used to update the state equation in real time. The posterior estimates of i and i allow the model predictions to gradually approach the measured values. When the predicted maximum settlement... When the system approaches the safety control threshold of the existing structure, it automatically generates an advance grouting compensation command or a core sampling sequence adjustment command.

[0065] like Figure 5 As shown, the execution control layer includes a vertical pipe sinking execution unit, a horizontal drilling rig execution unit, and a mud circulation execution unit.

[0066] The vertical pipe sinking execution unit is connected to the hydraulic drive system of the pipe rolling machine, and is used to adjust the sinking speed of the pipe rolling machine, the action of the correction cylinder, and the timing of the pipe extension according to the control commands. Specifically: Sinking speed adjustment: By adjusting the displacement of the hydraulic main pump of the pipe rolling machine, the sinking speed of the steel casing can be continuously adjusted within the range of 0.5-2.0m / h; Correction cylinder action: Based on the direction and magnitude of the deviation vector output by the parameter optimization module, the pushing force of 4-6 correction cylinders distributed around the circumference of the steel casing is controlled by the electro-hydraulic proportional valve to achieve dynamic correction of the steel casing. Sleeve extension sequence: When the steel sleeve is lowered to near the length of a single sleeve section, the automatic control of the pipe rolling machine stops the lowering, unlocks the sleeve joint, hoists the next sleeve section, connects and locks it, and resumes the lowering operation.

[0067] The horizontal drilling rig's actuator unit is connected to the drive system of the full-rotation horizontal drilling rig and is used to adjust the horizontal drilling rig's rotational speed ω and feed rate v according to control commands. f Core sampling depth and grouting pressure. Specifically: Rotation speed adjustment: The rotation speed can be continuously adjusted within the range of 0.05-0.5 r / s by adjusting the oil supply flow of the horizontal drilling rig's rotary hydraulic motor; Feed rate adjustment: The feed rate can be continuously adjusted within the range of 0.1-1.5m / h by adjusting the oil supply flow of the feed hydraulic cylinder of the horizontal drilling rig; Core depth control: The feed depth is fed back in real time through a displacement sensor. When the preset core depth is reached, the feed is automatically stopped and the core retraction program is started. Grouting pressure control: The grouting pressure can be continuously adjusted within the range of 0.2-0.4MPa by adjusting the opening of the outlet throttle valve of the grouting pump. The grouting flow rate is linked to the grouting pressure control.

[0068] The mud circulation execution unit is connected to the mud preparation and circulation system and is used to adjust the mud specific gravity γ according to the control command. mud and the depth h of the mud surface from the ground mud . Specifically: Mud specific gravity adjustment: By controlling the amount of weighting agent added in the mud preparation station, the mud specific gravity can be continuously adjusted within the range of 1.05-1.25; Mud slurry level depth adjustment: By controlling the replenishment flow rate and return flow rate of the mud circulation pump, the depth of the mud slurry level from the ground can be continuously adjusted within the range of 0-1.0m, with a target control range of 0.3-0.5m.

[0069] Each execution unit is equipped with a local controller, which receives control commands from the data processing layer and achieves precise control of the actuator through a PID control algorithm, with a control cycle of no more than 100ms.

[0070] The present invention also provides an intelligent control method for mechanical obstacle removal of steel-concrete composite structures in deep strata, which is implemented using the above-mentioned system and includes the following steps.

[0071] Step S1: Real-time acquisition of multi-dimensional working condition data during the obstacle removal operation using a multi-source sensing unit. Specifically, this includes: (1) Collect data on the verticality, planar position and sinking depth of the steel casing through the vertical tube attitude sensing subunit; (2) Collect the spatial position, drilling angle and drill bit attitude data of the horizontal drilling rig through the horizontal drilling rig posture sensing subunit; (3) Collect drilling pressure P, torque T, and drilling speed v during the core sampling operation through the drilling parameter sensing subunit. f and rotational speed ω data; (4) Collect formation displacement and pore water pressure u through the formation response sensing subunit. w and existing structural deformation data; (5) Collect mud specific gravity γ through mud parameter sensing subunit mud Data on the depth h_mud of the mud slurry surface from the ground and the sand content.

[0072] The sampling frequency of all sensing units is matched with the rate of change of key parameters in the obstacle removal operation: the sampling frequency of the drilling parameter sensing subunit is not less than 100 Hz to capture the parameter changes during the moment of steel cutting; the sampling frequency of the vertical pipe sinking attitude and the horizontal drilling rig position is not less than 10 Hz; and the sampling frequency of the formation response and mud parameters is not less than 1 Hz.

[0073] Step S2: Transmit multi-dimensional operating condition data to the data processing layer in real time through the data transmission layer. Fixed sensors use wired transmission via industrial Ethernet, while mobile sensors use 5G wireless transmission. Data transmission latency is controlled within 100ms. The data aggregation gateway performs time synchronization and format unification of data from each channel to ensure precise alignment of data on the timeline.

[0074] Step S3: Update the built-in digital twin model in real time based on multi-dimensional working condition data to keep the digital twin model synchronized with the actual state of the construction site. Specific updates include: (1) Update the spatial pose of the steel casing in the obstacle removal equipment model based on the vertical immersed tube attitude data: including the real-time horizontal offset E of the bottom center point of the casing and the comprehensive plane deviation Δ xy and the depth of subsidence; (2) Update the spatial pose and drilling trajectory of the horizontal drilling rig in the obstacle clearing equipment model based on the horizontal drilling rig pose data: including the current position of the drilling rig, drilling angle, and deviation of the drilling trajectory from the design trajectory; (3) Update the completion status and clearance quality of the current core hole in the obstacle clearing operation process model based on drilling parameter data: including the current drilling depth, core integrity rate, and the change history of drilling specific energy SE; (4) Update the stress-strain state of the geological model and the existing structural model of the obstacle clearing area based on the stratum response data: including the stratum displacement field, pore water pressure field and deformation field of the existing structure.

[0075] The digital twin model is updated at a frequency of no less than 1 Hz to ensure that the time delay between the model state and the actual on-site state does not exceed 1 second.

[0076] Step S4: Analyze and calculate the updated digital twin model using the intelligent decision engine to generate control instructions for the obstacle removal operation. The specific analysis and calculation process includes: The obstacle clearing status assessment module evaluates the completion status, clearing quality, and equipment health status of the current obstacle clearing operation.

[0077] Completion status assessment: Calculate the ratio of the number of completed core holes to the total number of holes, and calculate the overall progress percentage of the obstacle removal operation; assess the drilling progress of the current core hole.

[0078] Obstacle removal quality assessment: Based on indicators such as the stability of drilling specific energy (SE), core sample integrity rate, and fluctuation range of borehole wall stability safety factor (K), the obstacle removal quality level of each borehole location is comprehensively assessed.

[0079] Equipment health status assessment: The real-time wear status of the drill bit is assessed based on the cumulative cutting energy index (WI). When the WI reaches the preset tool life threshold, it is marked as "recommended to replace the drill bit". The working status of the hydraulic drive system is assessed based on the statistical characteristics of drilling pressure (P) and torque (T).

[0080] (ii) Real-time calculation of drilling specific energy and identification of cutting medium The cutting energy consumption per unit volume (SE) is calculated in real time based on the drilling specific energy real-time monitoring model. ; The real-time calculated SE value is compared with the preset stable reference range: When the SE value exceeds the upper limit of the preset stable reference range, it is determined that the horizontal drilling rig has entered the high-strength steel cutting section, and the intelligent decision engine automatically generates the first control command: reduce the feed rate. And increase the rotational speed ω; When the SE value is lower than the lower threshold of the preset stable benchmark range, it is determined to be a weak stratum. The intelligent decision engine automatically generates a second control command: increase the drilling rate. To improve the efficiency of obstacle removal work; When the SE value is within the preset stable reference range, the current cutting parameters remain unchanged.

[0081] WI is calculated based on the cumulative cutting energy index evaluation model: ; When the WI reaches the preset tool life threshold, the risk warning module issues a drill bit replacement warning signal, and the parameter optimization module automatically generates a "deceleration and torque reduction" protection strategy instruction.

[0082] The borehole wall stability safety factor K is calculated in real time based on the dynamic balance calculation model of mud-water pressure: ; in The active earth pressure at the bottom of the casing horizontal core drilling working face is calculated in real time based on Rankine earth pressure theory.

[0083] The real-time calculated K value is compared with a preset threshold: When the safety factor K is below 1.05, the intelligent decision engine determines that the borehole wall is in a critical state of instability and automatically generates a third control command: instructing the mud circulation unit to increase the mud density. Or add slurry to raise the liquid level. This allows the K value to recover and remain within the safe range of 1.10-1.25; When the safety factor K is in the range of 1.10-1.25, the current mud parameters remain unchanged; When the safety factor K is higher than 1.25, the mud specific gravity or liquid level can be appropriately reduced to improve the efficiency of obstacle removal operations.

[0084] The synchronous grouting volume is calculated in real time based on the formation loss compensation and grouting volume prediction model. : ; in This represents the measured total volume of the core sample that has been removed. The formation loss volume is calculated based on the formation displacement field inversion in the digital twin model. α is taken as 1.1-1.3 and β is taken as 0.6-0.9.

[0085] The execution control layer calculates the result. The grouting volume and pressure of the grouting system are controlled, with the dynamic adjustment range of grouting pressure being 0.2-0.4 MPa. The grouting process adopts a "fast first, slow later" grouting strategy—the first 50% of the grouting volume is injected in 1 / 3 of the time, and the remaining 50% is injected slowly in 2 / 3 of the time, in order to facilitate the full diffusion of the grout and the expulsion of gas from the borehole.

[0086] Real-time prediction of settlement of adjacent existing structures based on the improved Peck settlement prediction model: ; ; The intelligent decision engine uses Kalman filtering to fuse measured settlement with predicted values ​​and corrects formation loss in real time. and the settling trough width coefficient i. When the predicted maximum settlement... When the system approaches the safety control threshold of the existing structure, it actively generates an advance grouting compensation command or a core sampling sequence adjustment command.

[0087] The risk warning module compares the current multi-dimensional operating condition data with the preset safety thresholds in real time, identifies abnormal operating conditions, and issues graded warning signals.

[0088] The path planning module automatically plans the next coring hole location and drilling rig movement path based on the honeycomb-patterned borehole layout in the cleared area and the current clearing progress. When borehole location deviations or changes in formation conditions occur during the clearing process, the path planning module dynamically replans the coring sequence and movement path.

[0089] Step S5: The control layer receives and executes control commands to adjust the operating parameters of the obstacle removal equipment in real time. Specific adjustments include: (1) Based on real-time verticality deviation data, control the hydraulic cylinder of the pipe rolling machine to perform dynamic correction - when the rate of increase of E exceeds 2mm / m or Δ xyWhen the deviation is greater than 10mm, the compensation thrust required by the correction cylinder is automatically calculated based on the direction of the deviation vector, and the action of the correction cylinder is controlled by the electro-hydraulic proportional valve. (2) Adjust the drilling speed v of the horizontal drilling rig based on real-time drilling pressure P and torque T data. f With rotational speed ω—execute the first or second control command based on the calculation results of SE; (3) Based on real-time formation displacement and pore water pressure u w Based on the data, adjust the grouting pressure within the range of 0.2-0.4 MPa — according to Q inj The calculation results and formation response rate are adjusted in real time; (4) Adjust the mud specific gravity γ based on real-time mud parameters. mud Within the range of 1.10-1.20, adjust the depth h of the mud slurry surface from the ground. mud To maintain its depth from the ground within the range of 0.3-0.5m—the third control command is executed based on the calculation results of K.

[0090] The adjustment response time of each execution unit shall not exceed 500ms to ensure the timely execution of control commands.

[0091] Step S6: Repeat steps S1 to S5 to form a continuous closed-loop control of "sensing-transmission-analysis-regulation". The control cycle does not exceed 1 second until the core extraction operation of all clearing holes is completed.

[0092] The criteria for determining whether obstacle removal is complete are: all designed coring positions have completed coring operations, and the obstacle removal quality assessment level for each position reaches "qualified" or above. After obstacle removal is completed, the system automatically generates a complete obstacle removal operation report, including drilling parameter history curves for each position, formation response monitoring data, grouting records, and the final obstacle removal quality assessment results.

[0093] Specific embodiment: This embodiment takes the obstacle clearing well project of Goujielong Station in the SG18-11 section of Hangzhou Metro Line 18 Phase I project as an example to further illustrate the present invention.

[0094] Project Overview: The existing Goujilong Station on Line 6 is a two-level underground island platform station, and no provisions were made for the shield tunneling of Line 18. Before the shield tunnel passes under Goujilong Station, it is necessary to remove the existing station's diaphragm wall and steel-concrete structures such as H-beams within the shield tunneling area. The diaphragm wall is 800mm thick, with Ф28@300 steel reinforcement, and the joint I-beams are 10mm thick and 360mm wide. The clearance depth is 27.4-34.5m, and the stratum is mainly silty clay, with a confined aquifer approximately 3.2m away.

[0095] Before the obstacle removal operation begins, the deployment of the intelligent control system of this invention shall be completed: (1) Deployment of data sensing layer: Install double theodolite monitoring targets at a 90° angle along the circumference of the outer wall of the 4.0m diameter steel casing; install a high-precision tilt sensor every 3m along the axial direction on the inner wall of the casing; install a positioning and orientation instrument on the 1.3m diameter full-rotation horizontal drilling rig; install pressure sensor, torque sensor and displacement sensor on the hydraulic drive system of the horizontal drilling rig; deploy displacement gauge, earth pressure cell and pore water pressure gauge on the surrounding strata and adjacent existing Line 6 structure in the obstacle clearing area; install mud hydrometer, level gauge and sand content detector in the mud circulation system.

[0096] (2) Data transmission layer deployment: Fixed sensors use industrial Ethernet wired transmission, and mobile sensors use 5G wireless transmission, with transmission delay controlled within 100ms.

[0097] (3) Data processing layer deployment: deploy edge computing servers on the construction site, with built-in digital twin models and intelligent decision engines.

[0098] (4) Deployment of the execution control layer: The vertical pipe sinking execution unit is connected to the hydraulic drive system of the pipe rolling machine; the horizontal drilling rig execution unit is connected to the horizontal drilling rig rotary drive system, feed drive system and grouting system; the mud circulation execution unit is connected to the mud preparation station and mud circulation pump.

[0099] (5) Deployment of human-computer interaction layer: Set up a large screen display system and operating console in the construction site monitoring room.

[0100] The intelligent control process for obstacle removal operations is as follows: Step S1: The multi-source sensing unit collects multi-dimensional operating condition data in real time at a sampling frequency of not less than 10 Hz.

[0101] Step S2: The data transmission layer transmits multi-dimensional operating condition data to the edge computing server in real time with a latency of ≤100ms.

[0102] Step S3: The data processing layer updates the digital twin model in real time based on the operating condition data—the horizontal offset E of the casing bottom center point and the comprehensive plane deviation Δ. xy The spatial position and orientation of the horizontal drilling rig, the core sampling progress, and the stress-strain state of the formation and existing structures are updated synchronously.

[0103] Step S4: The intelligent decision engine analyzes and calculates the updated digital twin model: (1) Calculation of real-time deviation of verticality of steel casing: When the steel casing is sunk to a depth of 35m, the plane deviation Δ measured by the double theodolite xy When the deviation reaches 7mm, the system automatically calculates the compensation thrust of the correction cylinder based on the direction of the deviation vector, starts the cylinder of the pipe rolling machine for fine adjustment, and restores the deviation to 3mm within 3 minutes.

[0104] (2) Real-time monitoring and adaptive parameter control of drilling specific energy: When the horizontal drilling rig cuts to the position of the h-section steel joint of the diaphragm wall, the SE value jumps instantaneously from 2.3MPa to 4.1MPa, an increase of 78%, far exceeding the 30% sudden change threshold. The system quickly determines that it has entered the high-strength steel cutting section, and the intelligent decision engine automatically adjusts the drilling speed v. f The speed was reduced from 0.8 m / h to 0.3 m / h, while the rotational speed ω was increased from 0.2 r / s to 0.3 r / s. Simultaneously, the system accumulated the wear index in real time according to the WI formula, and automatically issued a drill bit replacement warning when the WI reached a preset threshold.

[0105] (3) Dynamic balance calculation and compensation of mud-water and soil pressure: During the core sampling process, u w The pressure fluctuated from 0.25 MPa to 0.31 MPa, and the safety factor K decreased from 1.18 to 1.03. The system immediately instructed the mud circulation unit to inject weighted mud to reduce the gamma ray pressure. mud Upgraded from 1.12 to 1.18, and h mud Increasing the value from 0.4m to 0.1m caused the K value to recover quickly and stabilize at 1.15.

[0106] (4) Formation loss compensation and accurate prediction of synchronous grouting volume: After the single-hole core sample is taken out, V core ≈1.06m³, ΔV inverted by digital twin model strata ≈0.28m³. The system is set according to Q. inj =1.2×1.06+0.8×0.28=1.496m³ The automatic calculation of synchronous grouting volume and the precise completion of grouting and backfilling with a grouting pressure of 0.3MPa are achieved.

[0107] (5) Real-time prediction of settlement of existing structures: The system uses the improved Peck formula to predict settlement in real time. When the predicted settlement is S max When the settlement approached 2.0 mm, the grouting compensation was initiated ahead of schedule, and the final measured settlement was controlled within 1.5 mm.

[0108] Step S5: The control layer receives control commands and adjusts the obstacle clearing equipment in real time through each execution unit.

[0109] Step S6: Repeat steps S1 to S5 to form a continuous closed-loop control until the core extraction operation of all clearing holes is completed.

[0110] In this embodiment, the following effects are achieved through the application of the intelligent control system of the present invention: (1) The verticality deviation of the steel sleeve is controlled within 1 / 500 throughout the entire process, and the planar position deviation is controlled within 10mm, which improves the accuracy by about 40% compared with the traditional manual monitoring method; (2) The core sampling integrity rate of each hole position is 100%, and the obstacle clearing range accurately covers the shield tunneling section; (3) No face instability or piping events occurred during the clearance process; (4) The maximum final settlement deformation of the existing Line 6 Goujielong Station structure was controlled within 1.5mm, and the operation of the existing line was not affected in any way; (5) It achieves on-demand grouting, saving approximately 22% of grouting materials compared to the traditional experience-based estimation method; (6) It reduced the manpower input for on-site monitoring and control by about 60%.

[0111] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

[0112] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0113] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0114] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0115] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0116] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mechanical obstacle removal intelligent control system for deep strata steel-concrete composite structures, characterized in that, include: The data perception layer includes multi-source sensing units deployed in the obstacle clearing operation area, used to collect multi-dimensional working condition data in real time during the obstacle clearing operation, including at least the drilling pressure, torque, advance rate and rotation speed of the horizontal drilling rig, mud specific gravity and the depth of the mud surface from the ground, as well as pore water pressure and formation displacement data. The data transmission layer is communicatively connected to the data perception layer and is used to transmit the multi-dimensional working condition data to the data processing layer in real time. The data processing layer is communicatively connected to the data transmission layer and has a built-in digital twin model and intelligent decision engine. The digital twin model is updated in real time based on the multi-dimensional working condition data to map the real-time physical state of the obstacle removal operation; The intelligent decision engine is configured as follows: Based on the aforementioned drill pressure, torque, drilling rate, and rotation speed, the drilling specific energy real-time monitoring model calculates the cutting energy consumption per unit volume and automatically determines the current cutting medium type according to its deviation from the preset stable range. When it is determined that the cutting medium is in the form of steel cutting, a first control command is generated to reduce the drilling rate and increase the rotation speed. When it is determined that the cutting medium is not in the form of steel cutting and the drilling specific energy is lower than the lower limit of the preset range, a second control command is generated to increase the drilling rate. Based on the specific gravity of the mud, the depth of the mud surface from the ground, and the pore water pressure, the safety factor of the borehole wall stability is calculated using the borehole wall stability safety factor calculation model. When the safety factor is lower than a preset threshold, a third control command is generated to increase the specific gravity of the mud or raise the mud surface. Based on the measured volume of the extracted core samples and the formation loss volume inverted from the digital twin model, the synchronous grouting volume is calculated using the grouting volume prediction model, and a fourth control command is generated accordingly. The execution control layer is communicatively connected to the data processing layer and is used to receive and execute the first to fourth control commands to adjust the drilling speed and rotation speed of the horizontal drilling rig, the mud specific gravity and the depth of the mud surface from the ground, and the output of the grouting system in real time. The data perception layer, data transmission layer, data processing layer, and execution control layer constitute a closed-loop control link.

2. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete structures according to claim 1, characterized in that, The real-time monitoring model for drilling specific energy calculates the cutting energy consumption per unit volume SE according to the following formula: ； In the formula, P is the drilling pressure and T is the torque. ω is the drilling rate, A is the rotation speed, and A is the cross-sectional area of ​​the core cylinder of the horizontal drilling rig.

3. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 1, characterized in that, The calculation model for the hole wall stability safety factor K is based on the following formula: ； In the formula, The specific gravity of the mud. The vertical height of the mud column is determined by the current depth of the clearing operation surface and the depth of the mud surface from the ground. Joint decision, Pore ​​water pressure, The active earth pressure at the bottom of the casing horizontal core drilling working face is calculated in real time based on Rankine earth pressure theory.

4. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 1, characterized in that, The grouting volume prediction model calculates the synchronous grouting volume according to the following formula. : ； In the formula, This represents the measured total volume of the core sample that has been removed. The formation loss volume is calculated based on the formation displacement field inversion in the digital twin model, where α is the filling compensation coefficient of the core sample volume and β is the advance grouting compensation coefficient of the formation loss.

5. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete structures according to claim 1, characterized in that, The multi-source sensing unit includes: The vertical tube sinking attitude sensing subunit is deployed on the ultra-large diameter steel casing and is used to collect data on the verticality, planar position and sinking depth of the steel casing. The horizontal drilling rig attitude sensing subunit is deployed on the full-rotation horizontal drilling rig and is used to collect the spatial position, drilling angle and drill bit attitude data of the horizontal drilling rig. The drilling parameter sensing subunit is installed on the drive system of the horizontal drilling rig and is used to collect data on drilling pressure P, torque T, and drilling speed during the core sampling operation. and rotational speed ω data; The formation response sensing subunit is deployed in the surrounding strata and adjacent existing structures of the cleared area to collect data on formation displacement and pore water pressure. and existing structural deformation data; The mud parameter sensing subunit is deployed in the mud circulation system inside the steel casing and is used to collect mud specific gravity. Depth of mud surface from the ground And sand content data.

6. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 5, characterized in that, The vertical tube attitude sensing subunit includes dual theodolites arranged at a 90° angle along the circumference of the steel casing for real-time cross-monitoring of the verticality of the steel casing; the data processing layer calculates the real-time horizontal offset E of the center point at the bottom of the casing according to the following formula: ； in Let be the length of the i-th stage casing segment. The angle between the measured axis of the i-th stage casing section and the designed plumb line; the comprehensive plane deviation of the dual theodolite cross-monitoring system. Solve using the following formula: ； in , The real-time planar coordinates of the top center of the casing are obtained based on the forward intersection method. , To design the center coordinates of the borehole; the horizontal drilling rig attitude sensing subunit includes a positioning and orientation instrument installed on the horizontal drilling rig. The positioning and orientation instrument has a built-in inertial measurement unit and a Kalman filter for realizing the autonomous navigation and positioning of the horizontal drilling rig.

7. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 1, characterized in that, The digital twin model includes: A geological model of the clearing area, constructed based on geological survey data, is used to characterize the stratigraphic distribution and hydrogeological features of the clearing area. The obstacle removal equipment model is constructed based on the geometric and kinematic parameters of the obstacle removal equipment and is used to characterize the spatial pose and motion state of the obstacle removal equipment. The obstacle clearing operation process model is constructed based on the temporal and spatial logic of the obstacle clearing process, and is used to characterize the process progress and spatial coverage of the obstacle clearing operation. An existing structural model, constructed based on the structural parameters of adjacent existing structures, is used to characterize the boundary conditions of the impact of the obstacle removal operation on the existing structures.

8. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 1, characterized in that, The intelligent decision engine includes: The obstacle clearing status assessment module is used to assess the completion status, obstacle clearing quality, and equipment health status of the current obstacle clearing operation based on the multi-dimensional working condition data and the digital twin model. The risk warning module is used to identify abnormal working conditions during the obstacle removal operation and issue warning signals. The parameter optimization module is used to dynamically optimize the operating parameters of the obstacle clearing operation based on the output results of the obstacle clearing status assessment module and the risk warning module, including at least optimizing the advance rate based on drill pressure P and torque T. With rotational speed ω, based on formation displacement and pore water pressure Optimize grouting pressure and grouting volume Optimize mud specific gravity based on mud parameters and liquid level depth And generate steel casing correction instructions based on verticality deviation; The path planning module is used to automatically plan the next core drilling position and the drilling rig movement path based on the honeycomb hole layout scheme of the cleared area and the current clearing progress.

9. The intelligent control system for mechanical obstacle removal of deep strata steel-concrete composite structures according to claim 1, characterized in that, The execution control layer includes: The vertical pipe sinking execution unit is connected to the hydraulic drive system of the pipe rolling machine and is used to adjust the sinking speed, the action of the correction cylinder and the timing of the sleeve extension according to the control command. The horizontal drilling rig execution unit, connected to the drive system of the full-rotation horizontal drilling rig, is used to adjust the rotational speed ω and the feed rate of the horizontal drilling rig according to the control commands. Core sampling depth and grouting pressure; The mud circulation execution unit, connected to the mud preparation and circulation system, is used to adjust the mud specific gravity according to the control command. and the depth of the mud surface from the ground .

10. A mechanical obstacle removal intelligent control method for deep strata steel-concrete composite structures, employing the system described in any one of claims 1 to 9, characterized in that, Includes the following steps: S1: Real-time acquisition of multi-dimensional working condition data during the obstacle clearing operation through multi-source sensing units; S2: The multi-dimensional operating condition data is transmitted to the data processing layer in real time through the data transmission layer; S3: Based on the multi-dimensional working condition data, the built-in digital twin model is updated in real time to keep the digital twin model synchronized with the actual state of the construction site; S4: The intelligent decision engine analyzes and calculates the updated digital twin model to generate control instructions for obstacle removal operations; The analysis and calculation include at least: calculating the unit volume cutting energy consumption SE based on the real-time monitoring model of drilling specific energy; when SE exceeds the upper limit of the preset stable benchmark range, it is determined to be a steel cutting condition and the drilling rate is reduced. Furthermore, increase the rotational speed ω, and increase the feed rate when SE is below the lower limit. ; The safety factor K is calculated based on the borehole wall stability safety factor calculation model. When K is less than 1.05, the mud density is increased. Alternatively, raise the mud level; calculate the synchronous grouting volume based on the grouting volume prediction model. ; S5: The control layer receives and executes the control instructions to adjust the operating parameters of the obstacle removal equipment in real time; S6: Repeat steps S1 to S5 to form a closed-loop control until the obstacle removal operation is completed.

Images