High-precision bolt tightening control system for tunnel segments based on multi-axis linkage control

Abstract

This invention belongs to the field of robotic arm technology, specifically disclosing a high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control. The system includes a deformation sensing and measurement unit, an adaptive alignment compensation unit, a multi-axis synchronous control unit, and a tightening strategy execution unit. By sensing deformation in real time and dynamically compensating for axial offset, combined with multi-axis torque and rotation angle synchronous control, it achieves uniform stress distribution on the segment connection surface, improving the tunnel structure's sealing performance and long-term stability.

Description

Technical Field

[0001] This invention belongs to the field of robotic arm technology, specifically relating to a high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control. Background Technology

[0002] In the field of tunnel engineering construction, segment assembly and connection are crucial steps determining the safety and long-term stability of the tunnel structure. The tightening of segment bolts directly affects the overall structure's sealing performance and load-bearing capacity. As a core piece of equipment for achieving high-precision assembly, the performance of the bolt tightening control system directly impacts construction quality and efficiency.

[0003] Among them, the tunnel segment bolt tightening system based on multi-axis linkage control is an important technical direction for realizing the coordinated operation of multiple bolts. Its goal is to achieve high-precision and high-efficiency tightening of bolt groups by coordinating the actions of multiple tightening axes, so as to ensure uniform stress distribution on the segment connection surface.

[0004] In existing technologies, multi-axis tightening systems typically employ rigid structural designs, which struggle to accommodate the minute deformations of tunnel segments under tightening loads. This leads to misalignment between the tightening axis and the bolt axis, resulting in significant lateral stress. In severe cases, this can damage bolt threads, tightening sleeves, or even the tunnel segment itself. Furthermore, existing control systems lack high-precision synchronous torque and angle control capabilities during multi-bolt coordinated tightening, making it difficult to achieve synchronous tightening according to a specific sequence and strategy. This results in uneven stress distribution at the segment connection surfaces, affecting the overall sealing and long-term stability of the tunnel structure.

[0005] Therefore, in the complex and ever-changing tunnel construction environment, how to achieve dynamic adaptation of the multi-axis tightening system to the deformation of the tunnel segments and high-precision synchronous control of multiple bolts has become an urgent technical problem to be solved. Summary of the Invention

[0006] The technical problem this invention aims to solve is to overcome the alignment deviation and lateral stress problems caused by the rigid structural design of existing multi-axis bolt tightening systems for tunnel segments, which makes it difficult to adapt to segment deformation. It also addresses the uneven stress distribution at the connection surface caused by insufficient synchronous control precision during multi-bolt coordinated tightening. The purpose of this invention is to provide a high-precision bolt tightening control system for tunnel segments based on multi-axis linkage control. This system can dynamically sense and adaptively compensate for segment deformation and achieve high-precision synchronous control of the torque and rotation angle of multiple bolts, thereby ensuring uniform stress distribution and long-term structural stability at the segment connection surface.

[0007] To achieve the above objectives, the technical solution adopted by this invention is to construct a complete system integrating deformation sensing, dynamic compensation, and multi-axis synchronous control. This system includes a deformation sensing and measurement unit, an adaptive alignment compensation unit, a multi-axis synchronous control unit, and a tightening strategy execution unit.

[0008] The deformation sensing and measurement unit is responsible for acquiring real-time deformation data of the segment surface under preload. This unit includes a laser displacement sensor array and a strain gauge sensor network arranged at the end of the tightening shaft. The laser displacement sensor array measures the three-dimensional spatial offset between the tightening shaft axis and the theoretical bolt hole axis non-contactly at a sampling frequency of 1000Hz. The strain gauge sensor network is directly attached to the key area of ​​the segment connection surface to measure the micro-strain changes in the local area of ​​the connection surface during tightening, with a measurement accuracy of 1με. The deformation sensing and measurement unit transmits the acquired spatial offset and micro-strain data to the adaptive alignment compensation unit in real time.

[0009] The adaptive alignment compensation unit receives data from the deformation sensing and measurement unit and calculates the dynamic compensation required for each tightening axis. This unit incorporates a deformation compensation algorithm, which first constructs a real-time deformation field model of the segment connection surface based on spatial offset and micro-strain data. Further, the algorithm predicts the future deformation trend of each bolt hole position based on this deformation field model. Subsequently, the algorithm calculates the six degrees of freedom compensation displacement and angle adjustment required for each tightening axis to maintain alignment with the bolt hole axis, including three translational degrees of freedom and three rotational degrees of freedom. The adaptive alignment compensation unit then sends the calculated compensation commands to the corresponding electro-hydraulic servo compensation mechanism for each tightening axis.

[0010] The electro-hydraulic servo compensation mechanism is a physical device that performs dynamic compensation. Each tightening shaft is equipped with an independent set of this mechanism. The mechanism consists of a high-precision servo valve, a hydraulic actuator, a ball joint connector, and a high-rigidity cross slide. After receiving the compensation command, the servo valve precisely controls the extension and direction of the hydraulic actuator, driving the cross slide to perform translational motion in a two-dimensional plane. Simultaneously, the ball joint connector provides deflection degrees of freedom around two rotation axes. The electro-hydraulic servo compensation mechanism can complete a maximum displacement compensation of 5mm or a maximum angular deflection compensation of 2° within 50ms, with a repeatability better than 10μm.

[0011] The multi-axis synchronous control unit coordinates the tightening actions of multiple tightening axes to ensure the synchronization of torque and angle control. This unit receives tightening strategy instructions from the central processing unit and collects the output torque and angle values ​​of each tightening axis in real time. The multi-axis synchronous control unit adopts a master-slave synchronous control architecture, designating one tightening axis as the master axis and the others as slave axes.

[0012] This unit incorporates a built-in synchronous control algorithm, which first compares the real-time torque and angle differences between each slave axis and the master axis. When the torque difference between any slave axis and the master axis exceeds a preset threshold of 5 N·m or the angle difference exceeds a preset threshold of 2°, the synchronous control algorithm immediately generates an adjustment command. This adjustment command dynamically adjusts the speed and current of the slave axis servo motors, enabling the torque and angle outputs of the slave axes to quickly track the state of the master axis until the torque and angle differences of all axes return to the threshold range.

[0013] The tightening strategy execution unit is the central decision-making core of the system, internally storing various predefined tightening strategies. These strategies include a two-step tightening method, a three-step tightening method, and a combined torque and angle control method. The tightening strategy execution unit selects and activates a specific tightening strategy based on the segment type, bolt specifications, and design requirements. This unit sends step-by-step tightening commands to the multi-axis synchronous control unit, including the target torque value, target angle value, and dwell time between steps for each step.

[0014] For example, when executing the three-step tightening method, the tightening strategy execution unit sequentially issues initial pre-tightening commands, intermediate tightening commands, and final tightening commands. The target torque values ​​for each step are 30%, 70%, and 100% of the final torque value, respectively. After each step is completed, the system remains stable for 2 seconds. The tightening strategy execution unit also monitors the entire tightening process in real time. If any abnormal torque or angle is detected, the safety protocol is immediately activated, the tightening operation is stopped, and an alarm is issued.

[0015] In a preferred embodiment of the present invention, the laser displacement sensor array in the deformation sensing and measurement unit adopts a structure in which four laser probes are evenly distributed in a 90° ring. Each laser probe independently measures its distance to the surface of the tube segment. The system uses the principle of triangulation to combine the distance data of the four probes and calculates the three-dimensional spatial offset vector of the tightening shaft axis relative to the bolt hole axis, including the lateral offset of the X-axis and Y-axis and the axial offset of the Z-axis.

[0016] Furthermore, the deformation compensation algorithm in the adaptive centering compensation unit adopts a prediction model based on Kalman filtering. This model uses the spatial offset and micro-strain data at the current moment as observations, and combines them with the known elastic modulus and Poisson's ratio parameters of the segment material to make the optimal estimate of the deformation of each bolt hole position at the next moment, thereby realizing feedforward compensation for the deformation trend and improving the system response speed and control accuracy.

[0017] Furthermore, the synchronization control algorithm of the multi-axis synchronous control unit adopts a fuzzy proportional-integral-derivative (FID) control strategy. This strategy takes the synchronization errors of torque and angle, and their rates of change, as inputs, performs rule judgments through a fuzzy inference engine, and outputs the adjustment amount of the control parameters of the slave axis servo motor. The fuzzy proportional-integral-derivative (FID) control strategy can effectively suppress the influence of system nonlinearity and load disturbances on synchronization accuracy.

[0018] Furthermore, the tightening strategy execution unit also integrates a data recording and analysis module. This module records complete data for each tightening operation, including the final torque value, total rotation angle, operation time, and historical synchronization error. This data is stored in the system's non-volatile memory and can be exported via a data interface for construction quality traceability and process optimization analysis.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0020] 1. This invention captures the deformation of the tube segment in real time through a deformation sensing and measurement unit, and achieves dynamic alignment compensation of the tightening shaft through an adaptive alignment compensation unit and an electro-hydraulic servo compensation mechanism. This fundamentally eliminates the problem of misalignment of the axis caused by the deformation of the tube segment, reduces lateral stress, effectively protects the bolt threads, tightening tools and the tube segment structure, and improves the service life of the equipment and the safety of operation.

[0021] 2. This invention achieves high-precision synchronization of torque and rotation of multiple tightening shafts through a master-slave synchronous control architecture and advanced synchronous control algorithm in a multi-axis synchronous control unit. This ensures the uniformity of stress distribution on the segment connection surface, greatly improves the overall sealing performance and long-term load-bearing stability of the tunnel structure, and fundamentally enhances the construction quality of tunnel engineering.

[0022] 3. The tightening strategy execution unit of the present invention provides a variety of optional standardized tightening strategies and has a complete process monitoring and safety protection mechanism, making the entire tightening process more standardized, intelligent and reliable. It not only improves construction efficiency and reduces reliance on manual operation experience, but also provides a solid data foundation for the digital management and traceability of construction quality. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall technical solution architecture of the high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control proposed in this invention;

[0024] Figure 2 This is a schematic diagram of the core principle framework of the deformation sensing and measurement unit in this invention;

[0025] Figure 3 This is a logical flow diagram of the adaptive centering compensation unit in this invention;

[0026] Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow of the multi-axis synchronous control unit in this invention;

[0027] Figure 5 This is a schematic diagram comparing the technical effects and principles of the tightening strategy execution unit in this invention; Detailed Implementation

[0028] The high-precision bolt tightening control system for tunnel segments based on multi-axis linkage control includes a deformation sensing and measurement unit, an adaptive alignment compensation unit, a multi-axis synchronous control unit, and a tightening strategy execution unit. Please refer to the appendix. Figure 1 In the overall system architecture, the deformation sensing and measurement unit collects real-time deformation data of the tube segment surface through a laser displacement sensor array and strain gauge sensor network, and transmits it to the adaptive alignment compensation unit. The adaptive alignment compensation unit calculates the dynamic compensation amount based on the received data and drives the electro-hydraulic servo compensation mechanism to perform displacement and angle adjustments. The multi-axis synchronous control unit receives instructions from the tightening strategy execution unit and coordinates the synchronous control of the torque and rotation angle of multiple tightening axes. The tightening strategy execution unit, as the central decision-making core, stores and executes predefined tightening strategies while monitoring the entire process. Through the collaborative work of these units, the system achieves adaptive compensation for tube segment deformation and high-precision synchronous tightening of multiple bolts.

[0029] The deformation sensing and measurement unit is responsible for collecting real-time deformation data of the segment surface under preload. Please refer to the appendix. Figure 2 The unit includes a laser displacement sensor array and a strain gauge sensor network arranged at the end of the tightening shaft.

[0030] A laser displacement sensor array non-contactly measures the three-dimensional spatial offset between the axis of the tightening shaft and the axis of the theoretical bolt hole at a sampling frequency of 1000Hz.

[0031] The laser displacement sensor array employs four laser probes arranged in a 90° ring, with each probe independently measuring its distance to the tube surface. The system uses triangulation to synthesize the distance data from the four probes, calculating the three-dimensional spatial offset vector of the tightening shaft axis relative to the bolt hole axis, including the X-axis lateral offset, Y-axis lateral offset, and Z-axis axial offset.

[0032] The solution process first converts the distance values ​​of each probe into coordinates in a local coordinate system, then maps them to a global coordinate system using a coordinate transformation matrix. Finally, the spatial equation of the tightening shaft axis is fitted using the least squares method and compared with the theoretical bolt hole axis equation to obtain the offset vector. The strain gauge sensor network is directly attached to the key area of ​​the pipe segment connection surface to measure the micro-strain changes in the local area of ​​the connection surface during tightening.

[0033] The strain gauge employs a full-bridge circuit configuration, achieving a measurement accuracy of 1με. The strain gauge outputs an analog voltage signal, which is sampled by a 24-bit analog-to-digital converter and then subjected to temperature compensation and nonlinearity correction by a digital signal processor.

[0034] The deformation sensing and measurement unit transmits the collected spatial offset and micro-strain data to the adaptive centering compensation unit in real time via a gigabit Ethernet interface. The data transmission protocol adopts a custom binary protocol based on TCP to ensure data integrity and low latency.

[0035] The adaptive alignment compensation unit receives data from the deformation sensing and measurement unit and calculates the required dynamic compensation for each tightening shaft accordingly. Please refer to the appendix. Figure 3 The unit has a built-in deformation compensation algorithm, which first constructs a real-time deformation field model of the segment connection surface based on the spatial offset and micro-strain data.

[0036] The deformation field model was discretized using the finite element method, dividing the segment connection surface into 512 triangular elements. Each element node was assigned the real-time measured displacement and strain values. The model calculated the displacement and stress distribution of the entire connection surface by solving the elastic equilibrium equations.

[0037] Furthermore, the algorithm predicts the future deformation trend of each bolt hole location based on this deformation field model.

[0038] The deformation compensation algorithm adopts a prediction model based on Kalman filtering. This model uses the spatial offset and micro-strain data at the current moment as observations, and combines the known elastic modulus and Poisson's ratio parameters of the segment material to make the optimal estimate of the deformation of each bolt hole position at the next moment.

[0039] The state equation of the Kalman filter describes the dynamic characteristics of deformation evolution over time, while the observation equation correlates sensor measurements with state variables. The predictive model achieves feedforward compensation for deformation trends by recursively calculating prior and posterior estimates of the state vector.

[0040] The core formula of the deformation compensation algorithm is:

[0041] ;

[0042] in, Indicates the first The state prior estimation vector at time t, which includes the three-dimensional displacement and rotation of each bolt hole position;

[0043] The state transition matrix is ​​derived from the elastic modulus and Poisson's ratio parameters of the segment material.

[0044] For the first The posterior state estimate vector at time 1;

[0045] To control the input matrix; The external control input vector mainly contains the preload applied by the tightening shaft.

[0046] The algorithm then calculates the six-degree-of-freedom compensation displacement and angle adjustment required for each tightening axis to maintain alignment with the bolt hole axis, including three translational degrees of freedom and three rotational degrees of freedom. The calculation is achieved through inverse kinematics, decomposing the transformation matrix between the target bolt hole axis coordinate system and the current tightening axis coordinate system into translation vectors and Euler angles. The adaptive alignment compensation unit sends the calculated compensation commands to the corresponding electro-hydraulic servo compensation mechanism of each tightening axis via a real-time Ethernet bus. The command data packet includes the target position, target velocity, acceleration curve, and motion mode fields.

[0047] The electro-hydraulic servo compensation mechanism is a physical device that performs dynamic compensation. Each tightening shaft is equipped with an independent set of this mechanism. The mechanism consists of a high-precision servo valve, a hydraulic actuator, a ball joint connector, and a high-rigidity cross slide. After receiving the compensation command, the servo valve precisely controls the extension amount and direction of the hydraulic actuator.

[0048] The servo valve adopts a three-stage electro-hydraulic servo valve structure, with a frequency response of 200Hz and a resolution better than 0.1μm. A magnetostrictive displacement sensor is integrated inside the hydraulic actuator to provide real-time feedback on the actual position of the piston rod, forming a closed-loop control.

[0049] The cross slide consists of two sets of precision linear guides and ball screw pairs, providing translational degrees of freedom in the X and Y axes, respectively.

[0050] The ball screw has a lead of 5mm, and the servo motor drives the screw to rotate via a planetary reducer, enabling precise movement of the slide. A ball joint connector is installed between the cross slide and the tightening shaft, providing deflection freedom about two rotational axes.

[0051] The ball joint uses a self-lubricating composite material bushing with a friction coefficient of less than 0.01, ensuring smooth and backlash-free rotation.

[0052] The electro-hydraulic servo compensation mechanism can complete a displacement compensation of up to 5mm or an angle deflection compensation of up to 2° within 50ms, with a repeatability better than 10μm. The mechanism integrates an overload protection module, which immediately cuts off the hydraulic circuit and activates the mechanical locking device when the load force exceeds 150% of the rated value.

[0053] The multi-axis synchronous control unit coordinates the tightening actions of multiple tightening axes to ensure synchronization of torque and angle control. Please refer to the appendix. Figure 4 This unit receives tightening strategy instructions from the central processing unit and collects the output torque and rotation angle values ​​of each tightening shaft in real time.

[0054] Torque is measured using a strain gauge torque sensor mounted in the tightening shaft drive chain. The sensor has a measurement range of 0 to 2000 N·m and an accuracy class of 0.5. Angle is measured using an absolute encoder mounted directly on the rear of the servo motor. The encoder has a 24-bit resolution, corresponding to a 0.001-degree angle resolution.

[0055] The multi-axis synchronous control unit adopts a master-slave synchronous control architecture, designating one tightening axis as the master axis and the rest as slave axes.

[0056] The spindle selection strategy is dynamically determined based on the geometric center position of the bolt arrangement or the priority of the preload transmission path. The unit incorporates a synchronization control algorithm, which first compares the real-time torque and rotation angle differences between each slave axis and the spindle.

[0057] The torque difference is calculated by subtracting the absolute value of the driven shaft torque from the master shaft torque value, and the angle difference is calculated by subtracting the absolute value of the driven shaft angle from the master shaft angle value.

[0058] When the torque difference between any slave axis and the master axis exceeds the preset threshold of 5 N·m or the angle difference exceeds the preset threshold of 2°, the synchronous control algorithm immediately generates an adjustment command.

[0059] The synchronization control algorithm adopts a fuzzy proportional-integral-derivative control strategy, which takes the synchronization error of torque and rotation angle and its rate of change as input, and performs rule judgment through a fuzzy inference engine.

[0060] The fuzzy inference engine contains 25 fuzzy rules. The antecedent of each rule is a combination of linguistic variables, namely the error and the rate of change of the error, and the consequent of each rule is a linguistic variable that controls the adjustment amount of the output.

[0061] The output of the fuzzy proportional-integral-derivative controller is defuzzified and converted into adjustment values ​​for the control parameters of the slave axis servo motor.

[0062] The adjustments include the speed loop proportional gain, the integral time constant, and the current loop limiting value. This strategy can effectively suppress the impact of system nonlinearity and load disturbances on synchronization accuracy.

[0063] The multi-axis synchronous control unit communicates with each tightening axis servo drive via real-time industrial Ethernet, with a communication cycle of 1ms, ensuring the timeliness and determinism of control commands.

[0064] The tightening strategy execution unit is the central decision-making core of the system. Please refer to the appendix. Figure 5 It internally stores a variety of predefined tightening strategies. These strategies include a 2-step tightening method, a 3-step tightening method, and a combined torque and angle control method.

[0065] The tightening strategy execution unit selects and activates a specific tightening strategy based on the segment type, bolt specifications, and design requirements. The selection logic is based on a preset process parameter matrix, where the matrix row index is the segment type code, the column index is the bolt specification code, and the matrix elements are recommended tightening strategy identifiers. This unit sends step-by-step tightening commands to the multi-axis synchronous control unit, which include the target torque value, target rotation angle value, and dwell time between steps for each step.

[0066] The instruction data structure includes the steps number, target torque, target angle, maximum allowable overshoot, over-tolerance shutdown threshold, and step timeout fields.

[0067] For example, when executing the three-step tightening method, the tightening strategy execution unit sequentially issues initial pre-tightening commands, intermediate tightening commands, and final tightening commands. The target torque values ​​for each step are 30%, 70%, and 100% of the final torque value, respectively. After each step is completed, the system remains stable for 2 seconds. During the stable holding period, the system continuously monitors the torque decay rate. If the decay rate exceeds 5% per minute, it is determined to be an abnormal relaxation.

[0068] The tightening strategy execution unit also monitors the entire tightening process in real time, monitoring parameters including real-time torque curve, angle curve, synchronization error curve, and system status word. If an abnormality in torque or angle is detected, such as a sudden torque change exceeding 20% ​​of the set value or an angle exceeding the theoretical range by 15%, a safety protocol is immediately activated. The safety protocol includes stopping the tightening operation, releasing hydraulic system pressure, activating the mechanical brake, and issuing an audible and visual alarm. The alarm information is displayed through the human-machine interface, specifically including the fault code, fault level, and suggested handling measures.

[0069] The tightening strategy execution unit also integrates a data recording and analysis module. This module records complete data for each tightening operation, including the final torque value, total rotation angle, operation time, and synchronization error history. Data recording employs a circular buffer mechanism with a buffer capacity of 10,000 tightening operation data points. Each operation record includes a timestamp, bolt position code, tightening strategy identifier, step sequence data, and quality evaluation indicators.

[0070] The quality assessment index is calculated as the root mean square error between the actual torque curve and the ideal torque curve, and the root mean square error threshold is set to 3% of the final torque value.

[0071] This data is stored in the system's non-volatile memory, which uses a solid-state drive architecture and has a storage capacity of 1TB.

[0072] Data can be exported via Gigabit Ethernet or Universal Serial Bus interface, and the export format supports comma-separated value files and binary raw data files.

[0073] The exported data is used for construction quality traceability and process optimization analysis. For example, statistical process control methods are used to analyze the trend and stability of torque distribution, or machine learning algorithms are used to identify the correlation between tightening parameters and segment deformation patterns.

[0074] The system power supply and grounding design comply with industrial equipment electromagnetic compatibility standards. The main power supply uses three-phase 380V AC, which is converted to 24V DC and 5V DC to power each unit after passing through an isolation transformer and a purified power module. The grounding system adopts a single-point grounding method, with signal ground, power ground, and chassis ground wired separately and then connected to the main grounding copper busbar. The system cabinet has an IP54 protection rating, ensuring reliable operation in the humid and dusty environment of tunnel construction.

[0075] The system startup and self-test process includes hardware initialization, sensor calibration, and functional testing phases. During hardware initialization, the microprocessors of each unit load the firmware, and the peripheral interface chips are configured with their operating modes. In the sensor calibration phase, the laser displacement sensor undergoes zero-point and sensitivity calibration using a standard target plate, and the strain gauge sensor is calibrated by applying a known load. In the functional testing phase, the system simulates a tightening process to verify the motion range of the electro-hydraulic servo compensation mechanism, the response time of the multi-axis synchronous control, and the effectiveness of the safety protocol execution. After passing the self-test, the system enters standby mode, awaiting operator input of operating parameters via the human-machine interface.

[0076] This embodiment further optimizes the layout of the deformation sensing and measurement unit and the calculation strategy of the adaptive centering compensation unit for ultra-large tunnel segment connection structures. Based on a standard 4-point laser displacement sensor array, the deformation sensing and measurement unit adds 4 auxiliary laser probes, forming an 8-point ring measurement structure. The auxiliary laser probes are staggered with the main probe at 45 degrees, extending the measurement coverage area to a 150mm range around the bolt holes. The measurement data from the 8 probes are processed through a data fusion algorithm. First, abnormal measurement points caused by surface contamination or temporary obstruction are removed. Then, a weighted least squares method is used to fit the local deformation surface of the segment. The second-order surface fitting accurately characterizes the concave and convex deformation features. The resolution of the fused deformation data is improved to 0.5mm, suitable for measuring the connection surfaces of large tunnel segments with diameters exceeding 2 meters.

[0077] The deformation compensation algorithm for the adaptive centering compensation unit, based on the Kalman filter prediction model, introduces a Long Short-Term Memory (LSTM) network for deformation trend learning. The LTM network's input layer contains a sequence of historical deformation data for 20 consecutive sampling periods, its hidden layer contains 128 neurons, and its output layer predicts the deformation increment for the next 5 sampling periods. The network training data comes from a historical construction database, containing 10,000 sets of deformation records under different segment types, bolt specifications, and environmental conditions.

[0078] When the trained Long Short-Term Memory (LSTM) network is run online, it is computed in parallel with the Kalman filter model. The outputs of the two are fused through adaptive weighting, with the weights dynamically adjusted based on the prediction error covariance. The fused prediction results more accurately reflect the nonlinear deformation process, making it particularly suitable for compensating for slow deformation caused by concrete creep.

[0079] The electro-hydraulic servo compensation mechanism enhances the rigidity and load-bearing capacity of the cross slide based on the standard configuration. The cross slide guide rails are replaced with roller linear guides, increasing the rated static load from 5000 N to 10000 N. The ball screw pair diameter is increased from 20 mm to 32 mm, while the lead remains unchanged at 5 mm. The hydraulic actuator cylinder diameter is enlarged from 50 mm to 80 mm, providing a maximum output force of 15000 N. The ball joint connector adopts a double-row crossed roller bearing structure, expanding the deflection degrees of freedom from 2 to 3, achieving fully spherical motion. The enhanced electro-hydraulic servo compensation mechanism expands the compensation range to a maximum displacement of 8 mm and a maximum angular deflection of 3 degrees, while maintaining a response time within 50 ms.

[0080] For scenarios involving the tightening of numerous bolt groups, the multi-axis synchronous control unit upgrades the master-slave synchronous control architecture to a virtual spindle synchronous control architecture.

[0081] The virtual spindle is not a physically existing tightening shaft, but an ideal reference model generated by an algorithm. The motion trajectory of the virtual spindle is calculated based on the stiffness distribution of the segment connections, ensuring optimal overall stress distribution.

[0082] Each physical tightening axis uses the virtual spindle as the synchronization target, and the synchronization control algorithm adopts a distributed consensus algorithm.

[0083] Each tightening axis controller exchanges status information with its adjacent controllers, and through multiple iterations, the torque and rotation angle outputs of all axes tend to be consistent.

[0084] The consensus algorithm uses a ring network for communication, reducing the communication cycle to 0.5ms.

[0085] The architecture supports synchronous control of up to 32 tightening axes, with synchronization accuracy indicators maintaining a torque difference of no more than 5 N·m and an angle difference of no more than 2°.

[0086] The tightening strategy execution unit is designed for ultra-large tunnel segments, incorporating a stress uniformity optimization tightening strategy. This strategy is based on a finite element analysis model, pre-calculating the stress distribution on the connection surface under different tightening sequences before tightening. The finite element model has a meshing accuracy of 5mm, considering bolt preload, contact nonlinearity, and material creep effects. The optimization algorithm aims to maximize stress distribution uniformity, solving for the optimal tightening sequence and step-by-step torque values. During execution, the system strictly follows the optimized sequence, and after each tightening step, the stress distribution is measured using a strain gauge sensor network, compared with the predicted values, and then fine-tuned. This strategy ensures that the stress concentration factor on the connection surface of ultra-large tunnel segments is below 1.5, improving structural sealing and fatigue life.

[0087] The system data recording and analysis module has been expanded into a cloud-based collaborative analysis platform. Local data recording is compressed and encrypted before being uploaded to the cloud database via a 5G industrial router. The cloud platform integrates a big data analytics engine to aggregate and analyze tightening data from multiple projects and batches. Analysis functions include process parameter sensitivity analysis, equipment performance degradation early warning, and quality anomaly pattern identification. Analysis results are pushed to mobile terminals in the form of visual reports, supporting remote monitoring and decision-making by construction management personnel. The local and cloud data synchronization cycle is configurable, with a minimum synchronization interval of 1 minute.

[0088] The system's reliability design is enhanced, with key units employing a redundant architecture. In the deformation sensing and measurement unit, the laser displacement sensor array and strain gauge sensor network serve as backups for each other; the system automatically switches to the other data source when one fails. The main control module of the multi-axis synchronous control unit adopts a dual-machine hot standby configuration, with the status of the main and backup modules monitored via a heartbeat line, and the switching time is less than 10ms. The power supply system incorporates an uninterruptible power supply, providing at least 30 minutes of backup power in the event of a mains power failure.

[0089] 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 process, method, article, or apparatus.

[0090] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control, characterized in that, include: The deformation sensing and measurement unit is used to collect real-time deformation data of the tunnel segment surface under preload. An adaptive centering compensation unit is used to receive data from the deformation sensing and measurement unit and calculate the dynamic compensation amount required for each tightening shaft. An electro-hydraulic servo compensation mechanism is used to execute compensation commands issued by the adaptive centering compensation unit; A multi-axis synchronous control unit is used to coordinate the tightening actions of multiple tightening axes and ensure the synchronization of torque and angle control; The tightening strategy execution unit is used to store predefined tightening strategies and select and activate specific tightening strategies according to the segment type, bolt specifications and design requirements. The adaptive alignment compensation unit incorporates a deformation compensation algorithm. This algorithm first constructs a real-time deformation field model of the segment connection surface based on spatial offset and micro-strain data. Then, it predicts the future deformation trend of each bolt hole position based on this deformation field model. Subsequently, it calculates the six-degree-of-freedom compensation displacement and angle adjustment required for each tightening axis to maintain alignment with the bolt hole axis. The electro-hydraulic servo compensation mechanism consists of a high-precision servo valve, a hydraulic actuator, a ball joint connector, and a high-rigidity cross slide. After receiving the compensation command, the servo valve controls the extension amount and direction of the hydraulic actuator, driving the cross slide to perform translational motion in a two-dimensional plane. At the same time, the ball joint connector provides the deflection degree of freedom around the rotation axis. The multi-axis synchronous control unit adopts a master-slave synchronous control architecture, designating one tightening axis as the master axis and the rest as slave axes. The unit has a built-in synchronous control algorithm that compares the real-time torque and angle difference between each slave axis and the master axis. When the torque difference between any slave axis and the master axis exceeds a preset value or the angle difference exceeds a preset threshold, an adjustment command is generated. The tightening strategy execution unit sends step-by-step tightening commands to the multi-axis synchronous control unit. The commands include the target torque value, target rotation angle value, and dwell time between steps for each step.

2. The high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The deformation sensing and measurement unit includes a laser displacement sensor array and a strain gauge sensor network arranged at the end of the tightening shaft. A laser displacement sensor array measures the three-dimensional spatial offset between the axis of the tightening shaft and the axis of the theoretical bolt hole in a non-contact manner. The laser displacement sensor array adopts a structure in which laser probes are evenly distributed in a ring. Each laser probe independently measures its distance to the surface of the tube segment; The system uses the principle of triangulation to synthesize the distance data from the probe and calculates the three-dimensional spatial offset vector of the tightening shaft axis relative to the bolt hole axis.

3. The high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The deformation compensation algorithm in the adaptive centering compensation unit adopts a prediction model based on Kalman filtering. The model uses the spatial offset and micro-strain data at the current moment as observations, and combines them with the known elastic modulus and Poisson's ratio parameters of the segment material to make the optimal estimate of the deformation of each bolt hole position at the next moment.

4. The high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The synchronization control algorithm of the multi-axis synchronous control unit adopts a fuzzy proportional-integral-derivative control strategy. This strategy takes the synchronization error of torque and rotation angle and its rate of change as input, performs rule judgment through fuzzy inference engine, and outputs the adjustment amount of the control parameters of the slave axis servo motor.

5. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The predefined tightening strategies stored inside the tightening strategy execution unit include a 2-step tightening method, a 3-step tightening method, and a torque-angle combined control method. When performing the three-step tightening method, the tightening strategy execution unit sequentially issues the initial pre-tightening command, the intermediate tightening command, and the final tightening command.

6. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 5, characterized in that, After the tightening strategy execution unit executes each tightening command, the control system remains stable.

7. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The electro-hydraulic servo compensation mechanism can perform displacement compensation or angle deflection compensation.

8. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The tightening strategy execution unit also integrates a data recording and analysis module; this module records complete data for each tightening operation, including the final torque value, total rotation angle, operation time, and synchronization error history.

9. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 8, characterized in that, The data recording and analysis module stores the recorded data in the system's non-volatile memory and exports it through a data interface.

10. A high-precision tightening control system for tunnel segment bolts based on multi-axis linkage control according to claim 1, characterized in that, The tightening strategy execution unit monitors the entire tightening process in real time. If abnormal torque or angle is detected, the safety protocol is immediately activated to stop the tightening operation and issue an alarm.

Images