A method for intelligent control of a tunnel boring machine cutter changer robot

Abstract

This invention discloses an intelligent control method for a tunnel boring machine (TBM) cutterhead changing robot, belonging to the field of automation control technology. It aims to solve problems such as low efficiency, inaccurate decision-making, and large positioning errors in cutterhead replacement in complex geological formations. By collecting cutterhead state parameters through multiple sensors, a three-level decision threshold is constructed, and fuzzy logic is used for dynamic adjustment to achieve intelligent hierarchical control for wear prediction, early warning, and emergency response. A multi-dimensional weighted model is used to calculate the priority of worn cutterheads, and a dual-objective function is combined to optimize the cutterhead path, generating an efficient execution sequence. In the positioning stage, reverse path planning, PID compensation for inertial slippage, and multispectral imaging and feature matching technology are used to accurately extract cutterhead marker points and correct pose deviations, ensuring that the cutterhead enters the robot's workspace. An integrated dual-station transfer table enables synchronous exchange of old and new cutterheads through hydraulic and servo control. The end effector adopts force-position hybrid control, combined with a torque wrench for staged operation, improving disassembly and assembly stability.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, and more specifically to an intelligent control method for a tunnel boring machine cutter changer robot. Background Technology

[0002] In tunnel construction, the tunnel boring machine (TBM) is a core piece of equipment, and its cutterhead plays a crucial role in breaking rocks when excavating in hard rock formations. Cutterheads are prone to wear and tear from prolonged operation and need to be replaced promptly to ensure tunneling efficiency and construction safety. In existing technologies, cutterhead replacement mainly relies on manual operation within the tunnel boring machine (TBM) or semi-automated devices, which presents significant technical bottlenecks. Manual cutterhead replacement is limited by the confined space of the TBM and complex geological environments, facing safety risks such as high pressure and oxygen deficiency, rock bursts, and collapses. Furthermore, manual wear assessment relies on experience, resulting in low efficiency and insufficient accuracy. While some automated cutterhead replacement systems incorporate sensor monitoring, they generally suffer from simplistic status monitoring, relying only on a few parameters such as pressure or torque to determine wear, failing to form a multi-dimensional comprehensive assessment, leading to delayed or misjudged replacement decisions. In the positioning and alignment stage, traditional systems rely on preset trajectories or simple visual recognition, making it difficult to dynamically compensate for changes in the TBM's attitude and equipment errors. This often results in misalignment between the cutterhead replacement actuator and the target cutterhead, leading to disassembly and installation failures. These problems make it difficult for existing technologies to meet the engineering requirements of long-distance tunneling for large-diameter TBMs in terms of safety, accuracy, and efficiency in cutterhead replacement, and there are still issues to address, such as inaccurate status monitoring, unintelligent decision-making, large positioning and alignment errors, and insufficient coordination in the cutterhead replacement process. Therefore, to overcome these limitations, this invention proposes an intelligent control method for a TBM cutterhead replacement robot. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide an intelligent control method for a tunnel boring machine (TBM) cutterhead changing robot. This method solves the technical problems of low efficiency and poor safety in manual cutterhead replacement in existing TBMs, as well as inaccurate status monitoring, unintelligent decision-making, large positioning and alignment errors, and insufficient coordination in the cutterhead replacement process in automated cutterhead changing systems. The method achieves real-time monitoring of cutterhead wear status, cutterhead replacement decision-making based on replacement needs, precise spatial positioning and collaborative control of the cutterhead and robot, and efficient storage and transmission of cutterheads, thereby improving the automation level, reliability, and construction efficiency of TBM cutterhead replacement operations.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A method for intelligent control of a tunnel boring machine cutter changer robot includes:

[0006] Collect the operating status parameters of the tunnel boring machine cutter head, divide the parameter range based on multi-dimensional decision thresholds, trigger a graded early warning mechanism, make cutter replacement decisions based on cutter wear, and construct a cutter replacement sequence.

[0007] Based on the number of hobbing tools in the tool change sequence, trigger the tool change control operation:

[0008] Identify the hob type of the tool changing sequence, calculate the comprehensive priority index of wear-related hob replacement using a multi-dimensional weighted evaluation model, construct a wear execution sequence by combining the principle of minimizing the tool head rotation path, set a comprehensive priority for forced hob replacement, construct an emergency execution sequence through a dual objective function, and then generate the tool changing instruction set of the tool changing sequence by combining the available time window duration and tool head rotation control.

[0009] Based on the tool change instruction set, coarse positioning of the tool head is achieved through reverse rotation path optimization and dynamic planning of the tool head motion trajectory; inertial slip compensation is used to position the target replacement hobbing tool in the tool change robot's workspace, triggering the tool change operation.

[0010] Extract the target cutter replacement mark points, construct multi-level feature images through multi-scale Gaussian downsampling, obtain the mark point coordinates of the target cutter replacement enhancement image to identify pose deviations, and perform pose correction through a two-level fine-tuning mechanism;

[0011] The integrated dual-station circular changeover platform enables collaborative operation. Through a coordinated hydraulic lifting station and force-position hybrid control strategy, the tool changing robot is controlled to perform tool changing operations on the target hobbing tool.

[0012] Specifically, the steps for constructing the wear execution sequence include:

[0013] When a worn hob is detected and the tool change sequence begins, a multi-dimensional weighted evaluation model is used to update the execution order of the tool change sequence, i.e.:

[0014] Based on the hob wear degree of the tool change sequence, a basic priority is established, a regional threshold is configured, and the local area of ​​the cutter head where each hob is located in the tool change sequence is divided. By calculating the weighted average of the hob wear degree in the local area, the failure chain risk index of the hob in the tool change sequence is generated.

[0015] The hob wear degree and failure cascade risk index are linearly superimposed to output a comprehensive priority index. The bubble sort algorithm is used to dynamically rearrange the tool change sequence and update the execution order of the tool change sequence.

[0016] Obtain the tool turret stop time window and the required buffer time for the tool turret, divide the available time window, and start the multi-batch replacement mode when the available time window duration is longer than the single tool change reference time. Construct the wear execution sequence of the available time window based on the comprehensive priority index and the principle of minimizing the tool turret rotation path.

[0017] Specifically, the steps for constructing the wear execution sequence within the available time window include:

[0018] Based on the single tool change reference time and the available time window duration, calculate the number of tool changes that can be accommodated in each available time window;

[0019] Configure priority thresholds, including upper priority thresholds and lower priority thresholds, and divide the comprehensive priority index into three execution levels according to the priority thresholds;

[0020] The positions of the worn replacement hobs are converted to polar coordinates. The difference in rotation angle between any two worn replacement hobs is calculated. A dual objective function is constructed, which includes maximizing the comprehensive priority index and minimizing the rotation angle. The dual objective function is balanced by weighting coefficients. Based on a greedy strategy, the wear execution sequence for each available time window is generated.

[0021] Specifically, the steps for generating the tool change instruction set of the tool change sequence include:

[0022] Obtain the execution sequence, including the wear execution sequence and the emergency execution sequence; and obtain the target replacement hob according to the execution order.

[0023] When entering the tool turret stop time window, according to the execution sequence of the available time window of the tool turret stop time window, the target replacement hob of the execution sequence is obtained in sequence, and the reference angle for the operation of the tool changing robot is defined.

[0024] Based on the execution sequence of the target tool changer, the angle difference between the target tool changer and the tool changing robot's operating reference angle is calculated sequentially.

[0025] The overall priority index of the hobbing tool, the angle difference between the target hobbing tool replacement and the tool-changing robot's operating reference angle, and the updated execution order in the execution sequence of the available time window are encapsulated into a tool-changing instruction set.

[0026] Specifically, the steps for controlling the tool-changing robot to perform a tool-changing operation on the target hobbing cutter include:

[0027] During the cutter head rotation control phase, the angle of the target cutter replacement is captured in real time, and the initial angle difference is calculated by combining it with the operating reference angle of the tool changing robot. Based on the initial angle difference, it is determined whether to switch to the reverse rotation path.

[0028] The S-shaped acceleration and deceleration curve is used to dynamically plan the tool head motion trajectory. Based on the real-time inertia parameters, the angular velocity and acceleration threshold are adaptively adjusted to achieve rapid coarse positioning of the tool head.

[0029] Configure a buffer angle. When the angle difference between the target hob and the operating reference angle of the tool changing robot is less than the buffer angle, the incremental PID control algorithm is activated to position the target hob to the working space of the tool changing robot by compensating for inertial slippage.

[0030] The target cutter replacement enhanced image is acquired and generated based on multispectral imaging fusion technology. Multi-level feature images are constructed through multi-scale Gaussian downsampling to locate the marker points in the target cutter replacement enhanced image in order to calculate the pose deviation of the target cutter replacement.

[0031] Configure a conflict threshold to determine whether the two-level fine-tuning mechanism is triggered and perform pose correction.

[0032] The end effector of the tool changing robot uses the coordinates of the marked point of the target replacement hob to release the target replacement hob and install the hob to be replaced in stages through a force-position hybrid control strategy. The target replacement hob is recovered and the hob to be replaced is assembled in collaboration with a dual-station ring converter.

[0033] Specifically, the steps for locating the target and replacing the marker points in the hobbing enhancement image include:

[0034] After the target cutter is replaced and enters the workspace, multispectral imaging is initiated to acquire infrared and visible light images of the target cutter replacement.

[0035] A bilateral filter is used to reduce noise in the infrared image and contrast adaptive histogram equalization is performed on the visible light image. The target replacement roller enhancement image is generated by weighted fusion of the infrared and visible light images.

[0036] Based on the theoretical position of the pre-marked points of the hob in the enhanced image of the target hob replacement, a dynamic search window is established with the theoretical coordinates as the center;

[0037] Multi-scale Gaussian downsampling is performed on the image within the dynamic search window to construct multi-level feature images, including a top-level low-resolution image for fast coarse localization, a middle-level medium-resolution image for marker point template matching optimization, and a bottom-level high-resolution image for feature descriptor matching, in order to replace the marker points of the roller-enhanced image for target localization.

[0038] Specifically, the steps for determining whether the two-level fine-tuning mechanism has been triggered and for performing pose correction include:

[0039] Based on the theoretical position of the preset marker point in the enhanced image of the target cutter replacement, calculate the pose deviation of the target cutter replacement marker point coordinates;

[0040] Configure a conflict threshold. If the pose deviation exceeds the conflict threshold, a two-level fine-tuning mechanism is triggered to correct the pose, including:

[0041] In the initial fine-tuning stage, based on the position data of the tool changer end effector, a path planning algorithm is used to plan a compensation path and drive the end effector to perform pose correction compensation. During the correction process, the pose deviation is monitored simultaneously.

[0042] The number of pose correction compensations is counted, and a correction threshold is configured. If the number of pose correction compensations exceeds the correction threshold and the pose deviation exceeds the conflict threshold, a secondary fine-tuning process is triggered to perform secondary fine-tuning; otherwise, pose correction compensation is terminated and the tool head is locked.

[0043] In the secondary fine-tuning stage, an incremental PID algorithm is used to generate a cutter head angle correction command to drive the cutter head to rotate and correct the angle. The circumferential displacement of the cutter head is monitored in real time. After each angle correction, the image is re-acquired and the pose deviation of the target hob replacement mark coordinates is calculated. If the pose deviation is less than or equal to the conflict threshold, the pose correction is terminated and the cutter head is locked.

[0044] Otherwise, count the number of angle corrections. If the number of angle corrections exceeds the correction threshold, then issue an attitude warning.

[0045] Specifically, the steps for completing the target replacement hob recovery and the assembly of the hob to be replaced through the collaborative operation of a dual-station circular transfer table include:

[0046] After the cutter head is locked, the tool changing robot starts the disassembly and assembly process. Based on the coordinates of the target replacement hob, the end effector of the tool changing robot drives the torque wrench mounted on the end effector to align with the target replacement hob mark point through a force-position hybrid control strategy. According to the preset torque curve, the target replacement hob is loosened in stages.

[0047] The end effector of the tool changing robot grabs the target to replace the roller cutter. The hydraulic lifting mechanism of the dual-station ring transfer table drives the recovery side station to rise vertically, controlling the tool changing robot to transfer the target roller cutter to the recovery station.

[0048] Simultaneously, the cutter storage bin pushes the cutters to be replaced to the supply-side station of the dual-station circular transfer table via a chain conveyor.

[0049] When the dual-station circular converter receives the target replacement cutter and the cutter to be replaced, it triggers the conversion operation, controls the servo motor of the dual-station circular converter to drive the rotary mechanism to rotate, and changes the position of the target replacement cutter and the cutter to be replaced.

[0050] The dual-station circular transfer table raises the supply-side station to the working height of the tool changing robot. The end-effector gripper picks up the hob to be replaced and retracts along a mirror path. Using a force-position hybrid control strategy, the hob to be replaced is moved to the installation station for assembly.

[0051] Specifically, the steps for constructing the tool change sequence include:

[0052] The initial multi-dimensional decision thresholds for the hob operating state parameters are set based on the physical properties of the hob. The multi-dimensional decision thresholds include the normal operating range, the early warning range, and the forced replacement range.

[0053] Combining the construction stratum type, tunnel boring machine advance speed and cutterhead rotation speed, a fuzzy logic algorithm is used to update the multi-dimensional decision threshold correction coefficient in real time, dynamically adjust the multi-dimensional decision threshold, and update the decision threshold range.

[0054] The received hobbing cutter running status parameters are compared dimension by dimension with multi-dimensional decision thresholds to obtain the current decision threshold range, triggering a tiered early warning mechanism.

[0055] If the hob's operating status parameters are within the normal operating range, then the hob's operating status parameter values ​​are trend predicted.

[0056] If the hob's operating status parameters are in the warning range, they are marked as abnormal operating status parameters. An abnormal monitoring process is initiated for abnormal hobs with abnormal operating status parameters to count the number of abnormal operating status parameters, the duration of the abnormality, quantify the wear status, and obtain the wear degree of the abnormal hob.

[0057] Configure a tool change threshold. If the wear of an abnormal hob exceeds the tool change threshold, it will be included in the tool change sequence; otherwise, the wear calculation results will be continuously monitored and updated.

[0058] If the hob's operating status parameters are in the forced change range, then the hob will be included in the tool change sequence.

[0059] The beneficial effects of this invention are:

[0060] This invention dynamically divides the hobbing cutter's operating state range and adjusts the tool-changing strategy in real time through multi-dimensional decision thresholds and fuzzy logic algorithms, accurately distinguishing between wear-induced replacement and forced replacement needs, ensuring the scientific and timely nature of tool-changing decisions. By constructing an execution sequence using a multi-dimensional weighted evaluation model and a dual-objective function, it rationally plans tool-changing priorities and cutter head rotation paths, effectively avoiding inefficient operations caused by priority confusion or path redundancy. Through reverse rotation path optimization, S-shaped acceleration / deceleration curve planning, and incremental PID control, it achieves coarse positioning and inertial slip compensation for the cutter head, significantly improving the stability and accuracy of the positioning process and reducing rotational energy consumption and time loss. Utilizing multispectral imaging fusion technology and multi-scale Gaussian downsampling methods, it accurately extracts hobbing cutter marker points in complex environments such as dust, and combines this with a two-stage fine-tuning mechanism to correct pose deviations, solving the problem of insufficient visual positioning and operational accuracy under harsh working conditions. The system integrates a dual-station ring converter and a force-position hybrid control strategy to achieve coordinated operation of old cutter recovery and new cutter assembly. Through the phased operation of the hydraulic lifting station and intelligent torque wrench, the continuity and reliability of the cutter replacement process are ensured, and the overall automation level of the cutter replacement operation is improved. This effectively addresses the construction challenges of complex strata such as hard rock and water-rich areas, and provides stable support for the continuous and efficient tunneling of the tunnel boring machine. Attached Figure Description

[0061] Figure 1This is a structural schematic diagram of an intelligent control method for a tunnel boring machine cutter changer robot according to the present invention;

[0062] Figure 2 A flowchart illustrating the specific steps involved in constructing the tool change sequence according to this invention;

[0063] Figure 3 This is a flowchart illustrating the tool change instruction set for generating a tool change sequence and replacing a worn hob in this invention.

[0064] Figure 4 This is a flowchart illustrating the process of controlling the tool-changing robot to perform a tool-changing operation on a target by replacing the hobbing cutter.

[0065] Figure 5 This is a flowchart illustrating the two-stage fine-tuning mechanism of the present invention for pose correction. Detailed Implementation

[0066] Example 1

[0067] Please see Figure 1 This embodiment introduces an intelligent control method for a tunnel boring machine cutter changer robot, including:

[0068] Step S1: Real-time acquisition of tunnel boring machine cutter operating status parameters through multi-sensor fusion, including cutter wear, cutter speed, cutter temperature, cutter vibration frequency, cutter cutting torque and cutter axial load.

[0069] In this embodiment, a combination of contact and non-contact measurement methods is employed. Contact measurement uses a triaxial accelerometer mounted on the cutter shaft bearing housing to collect the cutter vibration frequency, a torque sensor integrated into the cutter head drive spindle to collect the cutter cutting torque, and a pressure sensor installed at the connection between the cutter support arm and the shield body to collect the cutter axial load. The cutter rotation speed is calculated from the pulse frequency by a magnetic encoder disk at the cutter shaft end in conjunction with a Hall effect speed sensor near the cutter holder. The cutter temperature is monitored by a miniature wireless temperature sensor in the cutter bearing cavity. Non-contact measurement is dynamically triggered based on the contact measurement results. A preliminary anomaly assessment is performed on the contact measurement parameters by configuring an anomaly threshold range. If any contact measurement parameter exceeds the anomaly threshold range, non-contact measurement is triggered. Non-contact measurement acquires images of the cutter using an industrial vision camera mounted on the inner wall of the cutter head chamber. The industrial vision camera is equipped with an automatic purging air path to address the impact of dust in the cutter head chamber on image quality. Wear features of the hob are identified using image recognition algorithms, and the wear amount is calculated by combining distance compensation data from a laser rangefinder. Specifically, based on the cutterhead position information obtained from the tunnel boring machine control system, the industrial camera group is triggered to acquire images when the hob enters the acquisition area pre-calibrated by a 3D model. Camera synchronization control parameters are configured to ensure image capture is completed when the cutterhead is relatively stationary, reducing the impact of motion blur. After image acquisition, the hob contour features are extracted using edge detection algorithms, the hob ring circumferential curve is identified, and key points on the hob edge are located. Based on this, structured light stripes emitted by a laser rangefinder are projected onto the hob surface, and the 3D coordinate data of each feature point is obtained through triangulation. The radial wear depth of the cutter edge and the change in cutter ring thickness are calculated, ultimately yielding a quantitative result of the cutter wear amount.

[0070] Step S2: Receive the hob running status parameters and divide the running status parameter decision threshold range based on multi-dimensional decision thresholds. When the hob running status parameters exceed the corresponding safety threshold, a hierarchical early warning mechanism is automatically triggered to make a tool change decision, construct a tool change sequence, and trigger tool change control operations based on the number of hobs in the tool change sequence, generating a tool change instruction set for the tool change sequence while also supporting manual intervention decision-making.

[0071] In this embodiment, the operating status parameters of the hobbing cutter are received in real time via an industrial bus, including cutter wear, cutter speed, cutter temperature, cutter vibration frequency, cutter cutting torque, and cutter axial load. The multi-dimensional decision threshold is constructed based on the fatigue characteristics of the cutter material, the stress distribution law of the formation, and construction safety specifications, and includes three levels of decision threshold intervals: normal operation interval, warning interval, and forced replacement interval. When any hobbing cutter operating status parameter exceeds the warning threshold, a graded warning mechanism is automatically triggered. During the normal operation interval, a wear trend reminder is pushed through the human-machine interface. During the warning interval, a pre-replacement plan is generated, the path pre-planning of the cutter-changing robot is initiated, and a warning report is sent. During the forced replacement interval, the tunnel boring machine's advance is immediately suspended, and a cutter-changing instruction set is generated. During the tool change instruction set generation process, a comprehensive priority is calculated using the entropy weight method based on the tool wear level and fault cascading risk. The angle difference between the target hob and the tool changer robot is calculated in real-time using the homogeneous transformation matrix between the tool head coordinate system and the tool changer robot's base coordinate system. The target hob refers to the hob determined by the tool change instruction set to be the one requiring a tool change operation. The job timing planning combines the tool head rotation cycle and the tool changer robot's single-cycle operation time, employing a shortest path first algorithm to avoid path conflicts while reserving time redundancy to handle unforeseen circumstances. The manual intervention interface supports priority adjustment and instruction delay execution from a remote control console.

[0072] Please see Figure 2 Preferably, the specific steps for constructing the tool change sequence include:

[0073] Based on the physical properties of the hob, including material fatigue curves and structural strength limits, initial multi-dimensional decision thresholds for the hob's operating state parameters are set. These multi-dimensional decision thresholds include three levels of decision threshold ranges: normal operating range, warning range, and forced replacement range. These ranges are used to quantify the operating boundaries of the hob's operating state parameters under different working conditions. Operating state parameters within the normal operating range indicate that the parameter values ​​are within the design safety range and the tool performance is stable. Operating state parameters within the warning range indicate that the parameter values ​​are approaching the design limits, requiring the initiation of a pre-maintenance process. Operating state parameters within the forced replacement range indicate that the parameter values ​​exceed the safety boundaries, posing a risk of failure.

[0074] Real-time geological data is acquired through stratum sensing sensors to determine the type of stratum under construction. Based on the type of stratum under construction and combined with the tunnel boring machine's advance speed and cutterhead rotation speed, a fuzzy rule database is designed based on historical construction data. Fuzzy logic algorithms are used to update the multi-dimensional decision threshold correction coefficients in real time, dynamically adjust the multi-dimensional decision thresholds, and update the decision threshold range.

[0075] The received hob running status parameters are compared dimension by dimension with the real-time updated multi-dimensional decision thresholds to obtain the decision threshold range in which the hob running status parameters fall. A tiered early warning mechanism is then triggered based on the decision threshold range in which the parameters fall.

[0076] If the hob's operating status parameters are within the normal operating range, then by combining historical operating status parameter data and using time series analysis, the hob's operating status parameter values ​​are trend-predicted in order to assess the risk of operating status parameter drift.

[0077] If the hob's operating status parameters are within the warning range, they are marked as abnormal operating status parameters. An abnormal monitoring process is initiated for hobs with abnormal operating status parameters. This involves statistically analyzing the number of abnormal operating status parameters and their duration (the duration after one or more operating status parameters trigger the warning range and remain in an abnormal state). Combining the number and duration of abnormal operating status parameters, the hob's operating status parameter values ​​are comprehensively evaluated to quantify its wear state and obtain its wear degree. For example, when calculating the wear degree of an abnormal hob, the number of abnormal operating status parameters entering the warning range is first counted and assigned differentiated weights according to parameter type. Then, a time decay or cumulative coefficient is set based on the abnormal duration. Through weighted summation or fuzzy comprehensive evaluation, the number, duration, and degree of abnormality of abnormal parameters are coupled and calculated, ultimately outputting a wear degree value in the range of 0 to 1 to quantify the tool wear state.

[0078] Configure a tool replacement threshold. If the wear of an abnormal hob exceeds the tool replacement threshold, it is determined to be unsustainable and marked as a worn hob to be replaced, and included in the tool replacement sequence. If the wear does not exceed the threshold, abnormal monitoring is continuously performed, and the wear calculation results are updated in real time until the tool replacement conditions are met or the operating status parameters return to normal.

[0079] If the hob's operating status parameters are in the forced replacement range, then mark it as a forced hob replacement and include it in the tool change sequence.

[0080] Preferably, the specific steps for generating the tool change instruction set of the tool change sequence include:

[0081] Operations such as starting and stopping the tunnel boring machine (TBM) cutterhead and initializing the robot have a fixed time consumption. If a cutter wears out and triggers a cutter change, frequent starts and stops will result in an excessively high proportion of fixed time consumption, reducing overall construction efficiency. To address this, a cutter change control threshold is configured. When the number of roller cutters in the cutter change sequence exceeds the threshold, a cutter change is triggered. Based on the type of roller cutters in the sequence, a cutter change instruction set is generated, namely:

[0082] Please see Figure 3When a worn hob is detected and the tool change sequence begins, a multi-dimensional weighted evaluation model is used to update the execution order of the tool change sequence, i.e.:

[0083] Based on the hob wear degree of the tool change sequence, a basic priority is established, and the failure chain risk index is evaluated by analyzing the health status of adjacent tools in the tool turret partition where the hob is located.

[0084] Specifically, by configuring regional thresholds, the cutterhead surface is divided into local areas of each cutterhead according to the radial spacing and the circumferential angle. The weighted average of the cutter wear in the local area of ​​the cutter in the cutter replacement sequence is calculated in real time. Combined with the formation lithology hardness parameters, the data is normalized to generate a fault cascading risk index in the range of 0 to 1.

[0085] By linearly superimposing the hob wear degree and the failure cascade risk index, a comprehensive priority index for hob replacement due to wear is output. The bubble sort algorithm is then used to dynamically rearrange the tool change sequence and update the execution order of the tool change sequence.

[0086] During tunnel boring machine (TBM) construction, the cutterhead cannot be changed without continuous rotation. The TBM's main control system obtains real-time cutterhead operation cycle parameters to determine the cutterhead stopping time window. Due to inertia, the cutterhead has an unstable buffer phase during the stopping process; direct cutter replacement can easily lead to positioning errors. By establishing the correlation between cutterhead angular displacement and time, the required buffer time is obtained, and the available stopping time window is divided. Frequent start-stop of the cutterhead is both time-consuming and detrimental to equipment. When the available time window length exceeds the single cutter replacement reference time, a multi-batch replacement mode is initiated. Considering the relative positions of the cutterhead and the cutter replacement robot, and based on a comprehensive priority index and the principle of minimizing the cutterhead rotation path, a wear execution sequence is constructed for each available time window to minimize the cutter replacement time.

[0087] Based on the single tool change reference time and the available time window duration, calculate the number of tool changes that can be accommodated in each available time window;

[0088] Configure priority thresholds, including upper priority thresholds and lower priority thresholds, and divide the comprehensive priority index into three execution levels according to the priority thresholds;

[0089] The position of the worn replacement cutter is converted to polar coordinates. The difference in rotation angle between any two worn replacement cutters is calculated. A dual objective function is constructed, which includes maximizing the comprehensive priority index and minimizing the rotation angle. The dual objective function is balanced by weighting coefficients. Based on a greedy strategy, the next worn replacement cutter that is closest to the current worn replacement cutter and has the highest priority is selected in turn to generate the wear execution sequence for each available time window.

[0090] When entering the tool turret stop time window, according to the wear execution sequence of its available time window, the wear replacement hob of the wear execution sequence is obtained in sequence, and the tool changing robot operation reference angle is defined, that is, the tool turret angle when the hob rotates to the front of the robot;

[0091] Based on the execution sequence of the wear replacement hob in the wear execution sequence, a bidirectional optimization algorithm is used to select the rotation path. The angle difference between the target replacement hob and the operating reference angle of the tool changing robot is calculated sequentially to ensure that the rotation angle of the tool head does not exceed 180°, thereby reducing rotation energy consumption and time loss.

[0092] The wear execution sequence of each available time window is encapsulated into a tool changing instruction set that conforms to industrial control standards, which includes the hob comprehensive priority index, the angle difference between the target replacement hob and the tool changing robot's operating reference angle, and the updated execution order.

[0093] When the forced replacement of the cutterhead is detected as part of the cutterhead replacement sequence, the overall priority of the forced replacement of the cutterhead is set to the highest level, and an emergency shutdown is initiated. An emergency shutdown signal is sent to the tunnel boring machine control system via the industrial bus to suspend the rotation of the cutterhead and keep the current power system on standby to preserve the cutterhead's fine-tuning capability.

[0094] Obtain the available time window during the current emergency stop. If it is greater than the base time for a single tool change, calculate the number of tool changes that the available time window can accommodate during the emergency stop. Based on the angle difference between the rolling tool and the forced rolling tool replacement in the tool change sequence, and the rolling tool comprehensive priority index, construct an emergency execution sequence including the forced rolling tool replacement through a dual objective function:

[0095] When entering the available time window for emergency stop of the tool turret, the angle difference between the target replacement hob and the tool changing robot's operating reference angle is calculated sequentially according to the emergency execution sequence to generate the tool changing instruction set for the available time window during emergency stop.

[0096] Step S3: Based on the tool change instruction set within the available time window, determine the target hob to be replaced, control the tool change robot to perform the tool change operation on the target hob, obtain the angle difference between the target hob and the tool change robot's operating reference angle in real time, control the tool turret to rotate so that the target hob is in the tool change robot's workspace, when the target hob is in the tool change robot's workspace, control the tool change robot's end effector to complete the hob disassembly and installation operation, and provide a new hob through the dual-station ring tool changer and recycle the old hob;

[0097] In this embodiment, upon receiving the instruction, the current angle of the cutter head is obtained in real time through an absolute encoder. The minimum rotation path to the target reference angle is calculated, and the cutter head is rotated. When it approaches the target reference angle, PID control is activated to compensate for inertial slippage. The tool changing robot picks up the new hob through a dual-station circular converter. The end effector completes the disassembly and installation based on the coordinates calculated by the homogeneous transformation matrix. The entire process triggers a triple safety mechanism, including mechanical locking and force sensor verification.

[0098] Please see Figure 4 Preferably, the specific steps for controlling the tool-changing robot to perform a tool-changing operation on the target hob include:

[0099] During the cutter head rotation control phase, the current angle of the target hob replacement is captured in real time by an absolute encoder, and the initial angle difference is calculated by combining it with the operating reference angle of the tool changing robot. Based on this initial angle difference, it is determined whether to switch to the reverse rotation path.

[0100] If the initial angle difference exceeds 180°, the reverse rotation path will be automatically switched to optimize the actual rotation angle to 360° minus the initial difference, thereby minimizing the travel of the cutter head.

[0101] The tool turret motion trajectory is dynamically planned using an S-shaped acceleration / deceleration curve, and the angular velocity and acceleration thresholds are adaptively adjusted based on real-time inertia parameters to ensure rapid coarse positioning of the tool turret. Real-time inertia parameters are calculated through the fusion of torque sensor and angular accelerometer data.

[0102] Configure a buffer angle. When the angle difference between the target hob and the operating reference angle of the tool changing robot is less than the buffer angle, start the incremental PID control algorithm to control the rotation of the tool head. Use a laser displacement sensor to monitor the inertial slip during the braking process of the tool head in real time. Compensate with reverse micropulses from the servo motor so that when the tool head enters the available time window, the target hob is in the working space of the tool changing robot.

[0103] After the target hob is replaced and enters the workspace, the tool change operation is triggered, and multispectral imaging is started. Near-infrared light and visible light are combined for illumination. Near-infrared light highlights the oxide layer reflection characteristics of the laser-etched mark, and visible light captures the surface texture details of the tool. Infrared and visible light images of the target hob are acquired.

[0104] A bilateral filter is used to reduce noise in the infrared image while preserving the edge sharpness of the marker points; contrast-adaptive histogram equalization is applied to the visible light image to enhance the visibility of features in low-light areas. The infrared and visible light images are then weighted and fused to generate an enhanced image for target replacement of the rolling cutter.

[0105] Based on the theoretical position of the pre-marked points of the hob in the enhanced image of the target hob replacement, a dynamic search window is established with the theoretical coordinates as the center. The dynamic search window dynamically adjusts its range according to historical positioning deviations and environmental disturbances.

[0106] Multi-scale Gaussian downsampling is performed on the image within the dynamic search window to construct multi-level feature images, including a top-level low-resolution image for fast coarse localization, a middle-level medium-resolution image for marker point template matching optimization, and a bottom-level high-resolution image for feature descriptor matching. The marker points of the target localization image are replaced with those of the roller-enhanced image.

[0107] In the top-level low-resolution feature image, background noise is suppressed by morphological top-hat transformation, candidate marker point regions are extracted by adaptive threshold segmentation, and preliminary marker point regions are selected by roundness and area constraints.

[0108] The preliminary marker region in the medium-resolution image is located. The normalized cross-correlation matching algorithm is used to search within the preliminary marker region based on the marker template. The response peak points within the preliminary marker region are identified according to the normalized cross-correlation index. The candidate region of marker is divided by fitting the quadratic surface of the response peak points.

[0109] Locate candidate regions for marker points in the underlying high-resolution image, extract feature descriptors for these candidate regions, compare their similarity with the marker point template, and output the coordinates of the marker points in the target image to be enhanced by replacing the hobbing cutter based on the similarity.

[0110] Based on the theoretical position of the preset marker point in the enhanced image of the target cutter replacement, calculate the pose deviation of the target cutter replacement marker point coordinates;

[0111] Please see Figure 5 Configure a conflict threshold. If the pose deviation exceeds the conflict threshold, a two-level fine-tuning mechanism is triggered to correct the pose, including:

[0112] In the initial fine-tuning stage, based on the position data of the end effector of the tool changer fed back by the six-dimensional force sensor, a path planning algorithm, such as the improved RRT* algorithm, is used to plan the compensation path in real time, driving the end effector to perform pose correction compensation along the tool axis, radial direction, and tangential direction. During the correction process, the pose deviation is monitored simultaneously. The number of pose correction compensations is counted, and a correction threshold is configured. If the number of pose correction compensations exceeds the correction threshold, and the pose deviation exceeds the conflict threshold, a secondary fine-tuning process is triggered to perform secondary fine-tuning; otherwise, the pose correction compensation is terminated, and the tool head is locked to provide a stable reference for subsequent disassembly and assembly operations.

[0113] In the secondary fine-tuning stage, incremental PID algorithm is used to generate micro-angle correction commands for the cutter head, drive the cutter head to rotate, perform angle correction, monitor the circumferential displacement of the cutter head in real time, re-acquire images after each angle correction, and calculate the pose deviation of the target hob replacement mark coordinates.

[0114] If the number of angle corrections is less than or equal to the conflict threshold, the pose correction is terminated; otherwise, the number of angle corrections is counted. If the number of angle corrections exceeds the correction threshold, a pose warning is issued, a manual intervention request is pushed, and the tool turret pose rollback function is activated to restore the most recent stable configuration.

[0115] When the positional deviation is less than or equal to the conflict threshold, the tool head is locked to provide a stable reference for subsequent disassembly and assembly operations.

[0116] After the cutter head is locked, the tool changing robot initiates the disassembly and assembly process. The robot's end effector, based on the coordinates of the marked points of the target replacement hob, integrates multispectral vision and pose sensor data. Through a force-position hybrid control strategy, it controls the alignment of the drive torque wrench. During this process, relying on visual sub-pixel feature matching and pose inertial measurement, it calculates and dynamically compensates for pose residuals in real time, ensuring that the tool and threaded hole coaxiality meet assembly requirements. After alignment, the target replacement hob is loosened in stages according to a preset torque curve. Pre-loosening releases the thread engagement, the transition stage overcomes static friction, and the final stage releases axial constraints.

[0117] The axial pressure and lateral offset of the end effector are monitored in real time to detect any abnormal resistance. If abnormal resistance is detected, a vibration mode is activated to assist in the loosening of the bolt by replacing the hobbing tool. The end effector is subjected to slight axial vibration, and special cutting fluid is injected simultaneously to reduce the coefficient of friction. During the vibration process, the frequency and phase are dynamically adjusted to match the bolt resonance characteristics for efficient loosening.

[0118] The end effector of the tool-changing robot uses a suction gripper to grasp the target and replace the hob. After the end effector gripper replaces the hob, contact pressure and a six-dimensional force sensor jointly verify the stability of the grasping action. The robot then moves along a pre-planned obstacle-avoidance path. This path pre-planning is integrated with the point cloud map of the tool head compartment to avoid fixed obstacles. During transport, a laser scanner updates the local point cloud in real time, and upon encountering a sudden obstacle, it immediately triggers joint trajectory replanning to dynamically maintain a safe distance. Simultaneously, based on the hob's center of gravity data fed back by the force sensor, if a center of gravity shift is detected, the robot automatically switches to a low-inertia attitude trajectory to prevent the robot from falling into dangerous postures such as joint limits or center of gravity instability.

[0119] When the target replacement cutter becomes loose, the hydraulic lifting mechanism of the dual-station circular transfer table drives the recovery station to rise vertically. The tool changing robot transfers the target replacement cutter to the recovery station along a pre-planned obstacle avoidance path. During the transfer, the laser scanner monitors the surrounding obstacles in real time and dynamically maintains a safe distance.

[0120] Simultaneously, the cutter storage bin pushes the cutters to be replaced to the supply-side station of the dual-station circular converter via a chain conveyor. When the dual-station circular converter receives the target cutter and the cutter to be replaced, it triggers a conversion operation, controlling the servo motor of the dual-station circular converter to drive the rotary mechanism to rotate, changing the position of the target cutter and the cutter to be replaced.

[0121] The dual-station circular transfer table raises the supply-side station to the working height of the tool-changing robot. The end effector gripper picks up the hob to be replaced and retracts along a mirror path. The path planning integrates the return load changes to dynamically allocate joint speeds, controlling the hob to be replaced to move to the installation station. During the installation phase, a force-position hybrid control strategy is employed: positioning is achieved through marker point coordinates. When the hob to be replaced is at the installation station, the system switches to impedance control mode, adaptively adjusting the monitoring end effector's pressing force. A cross-sequential loading strategy is used, controlling the torque wrench to incrementally increase force, pausing after each application to release stress, ensuring uniform thread tension to complete the assembly of the hob to be replaced.

[0122] Working principle and its effects:

[0123] This invention uses multi-dimensional decision thresholds and fuzzy logic algorithms to perform real-time dynamic analysis of hobbing cutter operating status parameters, and triggers tool change decisions through a hierarchical early warning mechanism, thereby achieving intelligent closed-loop control from status monitoring to command generation.

[0124] A multi-dimensional weighted evaluation model is adopted, comprehensively considering hob wear and failure cascading risk index, and combining dual objective functions to optimize the tool head rotation path. This ensures that high-priority tools are replaced first, while reducing invalid rotation paths, significantly improving the scientific nature and execution efficiency of tool change decisions. In the positioning stage, reverse rotation path planning, S-shaped acceleration and deceleration curve control, and incremental PID compensation are used to achieve rapid and accurate positioning of the tool head to the tool change workspace.

[0125] By utilizing multispectral imaging fusion technology and multi-scale Gaussian downsampling algorithm, dust interference is effectively overcome, the roller cutter marking points are accurately extracted and the pose deviation is calculated, and a two-level fine-tuning mechanism is used to correct the error, ensuring operational accuracy.

[0126] The collaborative operation design of the dual-station circular changeover platform enables synchronous exchange of old and new cutting tools through hydraulic lifting and servo rotation mechanisms. Combined with a force-position hybrid control strategy and intelligent torque wrench for phased operation, it achieves continuous and automated tool changing processes. This provides adaptive decision-making capabilities under complex geological conditions, effectively coping with harsh working conditions such as hard rock and water-rich environments. It significantly improves the safety, efficiency, and reliability of tool changing operations, providing strong technical support for the continuous and stable tunneling of the tunnel boring machine.

[0127] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent control of a tunnel boring machine cutter changer robot, characterized in that, include: Collect operating status parameters of the tunnel boring machine cutterhead; The parameter range is divided based on multi-dimensional decision thresholds, triggering a hierarchical early warning mechanism. Tool replacement decisions are made based on hob wear, and a tool replacement sequence is constructed. The specific steps for constructing the tool change sequence include: The initial multi-dimensional decision threshold for the hob operating state parameters is set based on the physical properties of the hob. The multi-dimensional decision threshold includes the normal operation range, the early warning range, and the forced replacement range. Combining the construction stratum type, tunnel boring machine advance speed and cutterhead rotation speed, a fuzzy logic algorithm is used to update the multi-dimensional decision threshold correction coefficient in real time, dynamically adjust the multi-dimensional decision threshold, and update the decision threshold range. The received hobbing cutter running status parameters are compared dimension by dimension with the multi-dimensional decision thresholds to obtain the current decision threshold range, triggering a tiered early warning mechanism. If the hob operating status parameters are within the normal operating range, then the hob operating status parameter values ​​are trend predicted; If the hob's operating status parameters are in the warning range, they are marked as abnormal operating status parameters. An abnormal monitoring process is initiated for the abnormal hobs with abnormal operating status parameters to count the number of abnormal operating status parameters, the duration of the abnormality, quantify the wear status, and obtain the wear degree of the abnormal hob. Configure a tool change threshold. If the wear of an abnormal hob exceeds the tool change threshold, it will be included in the tool change sequence; otherwise, the wear calculation results will be continuously monitored and updated. If the hob's operating status parameters are in the forced replacement range, then the hob will be included in the tool change sequence; Based on the number of hobbing tools in the tool change sequence, trigger the tool change control operation: Identify the hob type of the tool changing sequence, calculate the comprehensive priority index of wear-related hob replacement using a multi-dimensional weighted evaluation model, construct a wear execution sequence by combining the principle of minimizing the tool head rotation path, set a comprehensive priority for forced hob replacement, construct an emergency execution sequence through a dual objective function, and then generate the tool changing instruction set of the tool changing sequence by combining the available time window duration and tool head rotation control. Based on the tool change instruction set, coarse positioning of the tool head is achieved through reverse rotation path optimization and dynamic planning of the tool head motion trajectory; inertial slip compensation is used to position the target replacement hobbing tool in the tool change robot's workspace, triggering the tool change operation. Extract the target cutter replacement mark points, construct multi-level feature images through multi-scale Gaussian downsampling, obtain the mark point coordinates of the target cutter replacement enhancement image to identify pose deviations, and perform pose correction through a two-level fine-tuning mechanism; The integrated dual-station circular changeover platform enables collaborative operation. Through a coordinated hydraulic lifting station and force-position hybrid control strategy, the tool changing robot is controlled to perform tool changing operations on the target hobbing tool.

2. The intelligent control method for a tunnel boring machine cutter changer robot as described in claim 1, characterized in that, The specific steps for constructing the wear execution sequence include: When a worn hob is detected and the tool change sequence begins, a multi-dimensional weighted evaluation model is used to update the execution order of the tool change sequence, i.e.: Based on the hob wear degree of the tool change sequence, a basic priority is established, a regional threshold is configured, and a local region of the cutter head where each hob is located in the tool change sequence is divided. By calculating the weighted average of the hob wear degree in the local region, a failure chain risk index of the hob in the tool change sequence is generated. The hob wear degree and failure cascade risk index are linearly superimposed to output a comprehensive priority index. The bubble sort algorithm is used to dynamically rearrange the tool change sequence and update the execution order of the tool change sequence. Obtain the tool turret stop time window and the required buffer time for the tool turret, divide the available time window, and start the multi-batch replacement mode when the available time window duration is longer than the single tool change reference time. Construct the wear execution sequence of the available time window based on the comprehensive priority index and the principle of minimizing the tool turret rotation path.

3. The intelligent control method for a tunnel boring machine cutter changer robot as described in claim 2, characterized in that, The specific steps for constructing the wear execution sequence within the available time window include: Based on the single tool change reference time and the available time window duration, calculate the number of tool changes that can be accommodated in each available time window; Configure priority thresholds, including upper priority thresholds and lower priority thresholds, and divide the comprehensive priority index into three execution levels according to the priority thresholds; Convert the wear replacement cutter position to polar coordinates, calculate the rotation angle difference between any two wear replacement cutters, construct a dual objective function, including maximizing the comprehensive priority index and minimizing the rotation angle. Balance the dual objective function through weighting coefficients, and generate the wear execution sequence for each available time window based on a greedy strategy.

4. The intelligent control method for a tunnel boring machine cutterhead changing robot as described in claim 3, characterized in that, The specific steps for generating the tool change instruction set of the tool change sequence include: Obtain the execution sequence, including the wear execution sequence and the emergency execution sequence; and obtain the target replacement hob for the execution sequence according to the execution order; When entering the tool turret stop time window, according to the execution sequence of the available time window of the tool turret stop time window, the target replacement hob of the execution sequence is obtained in sequence, and the reference angle for the operation of the tool changing robot is defined. Based on the execution sequence of the target tool changer, the angle difference between the target tool changer and the tool changing robot's operating reference angle is calculated sequentially. The overall priority index of the hobbing tool, the angle difference between the target hobbing tool replacement and the tool-changing robot's operating reference angle, and the updated execution order in the execution sequence of the available time window are encapsulated into a tool-changing instruction set.

5. The intelligent control method for a tunnel boring machine cutter changer robot as described in claim 1, characterized in that, The specific steps for controlling the tool-changing robot to perform a tool-changing operation on the target hobbing tool include: During the cutter head rotation control phase, the angle of the target cutter replacement is captured in real time, and the initial angle difference is calculated by combining it with the operating reference angle of the tool changing robot. Based on the initial angle difference, it is determined whether to switch to the reverse rotation path. The S-shaped acceleration and deceleration curve is used to dynamically plan the tool head motion trajectory. Based on the real-time inertia parameters, the angular velocity and acceleration threshold are adaptively adjusted to achieve rapid coarse positioning of the tool head. Configure a buffer angle. When the angle difference between the target hob and the operating reference angle of the tool changing robot is less than the buffer angle, the incremental PID control algorithm is activated to position the target hob to the working space of the tool changing robot by compensating for inertial slippage. The target cutter replacement enhanced image is acquired and generated based on multispectral imaging fusion technology. Multi-level feature images are constructed through multi-scale Gaussian downsampling to locate the marker points in the target cutter replacement enhanced image in order to calculate the pose deviation of the target cutter replacement. Configure a conflict threshold to determine whether the two-level fine-tuning mechanism is triggered and perform pose correction. The end effector of the tool changing robot uses the coordinates of the marked point of the target replacement hob to release the target replacement hob and install the hob to be replaced in stages through a force-position hybrid control strategy. The target replacement hob is recovered and the hob to be replaced is assembled in collaboration with a dual-station ring converter.

6. The intelligent control method for a tunnel boring machine cutter changer robot as described in claim 5, characterized in that, The specific steps for replacing the marker points in the enhanced image of the hobbing cutter for the positioning target include: After the target cutter is replaced and enters the workspace, multispectral imaging is initiated to acquire infrared and visible light images of the target cutter replacement. The infrared image is denoised using a bilateral filter, and the visible light image is subjected to contrast adaptive histogram equalization. The infrared image and the visible light image are then weighted and fused to generate an enhanced image for target replacement of the rolling cutter. Based on the theoretical position of the pre-marked points of the hob in the enhanced image of the target hob replacement, a dynamic search window is established with the theoretical coordinates as the center; Multi-scale Gaussian downsampling is performed on the image within the dynamic search window to construct multi-level feature images, including a top-level low-resolution image for fast coarse localization, a middle-level medium-resolution image for marker point template matching optimization, and a bottom-level high-resolution image for feature descriptor matching, in order to replace the marker points of the roller-enhanced image for target localization.

7. The intelligent control method for a tunnel boring machine cutterhead changing robot as described in claim 5, characterized in that, The specific steps for determining whether the two-level fine-tuning mechanism has been triggered and for performing pose correction include: Based on the theoretical position of the preset marker point in the enhanced image of the target cutter replacement, calculate the pose deviation of the target cutter replacement marker point coordinates; Configure a conflict threshold. If the pose deviation exceeds the conflict threshold, a two-level fine-tuning mechanism is triggered to correct the pose, including: In the initial fine-tuning stage, based on the position data of the tool changer end effector, a path planning algorithm is used to plan a compensation path and drive the end effector to perform pose correction compensation. During the correction process, the pose deviation is monitored simultaneously. The number of pose correction compensations is counted, and a correction threshold is configured. If the number of pose correction compensations exceeds the correction threshold and the pose deviation exceeds the conflict threshold, a secondary fine-tuning process is triggered to perform secondary fine-tuning; otherwise, pose correction compensation is terminated and the tool head is locked. In the secondary fine-tuning stage, an incremental PID algorithm is used to generate a cutter head angle correction command to drive the cutter head to rotate and correct the angle. The circumferential displacement of the cutter head is monitored in real time. After each angle correction, the image is re-acquired and the pose deviation of the target hob replacement mark coordinates is calculated. If the pose deviation is less than or equal to the conflict threshold, the pose correction is terminated and the cutter head is locked. Otherwise, count the number of angle corrections. If the number of angle corrections exceeds the correction threshold, then issue an attitude warning.

8. The intelligent control method for a tunnel boring machine cutter changer robot as described in claim 5, characterized in that, The specific steps for completing the target replacement hob recovery and the assembly of the hob to be replaced through the coordinated use of a dual-station circular transfer table include: After the cutter head is locked, the tool changing robot starts the disassembly and assembly process. Based on the coordinates of the target replacement hob, the end effector of the tool changing robot drives the torque wrench mounted on the end effector to align with the target replacement hob mark point through a force-position hybrid control strategy. According to the preset torque curve, the target replacement hob is loosened in stages. The end effector of the tool changing robot grabs the target to replace the roller cutter. The hydraulic lifting mechanism of the dual-station ring transfer table drives the recovery side station to rise vertically, controlling the tool changing robot to transfer the target roller cutter to the recovery station. Simultaneously, the cutter storage bin pushes the cutters to be replaced to the supply-side station of the dual-station circular transfer table via a chain conveyor. When the dual-station circular converter receives the target replacement cutter and the cutter to be replaced, it triggers the conversion operation, controls the servo motor of the dual-station circular converter to drive the rotary mechanism to rotate, and changes the position of the target replacement cutter and the cutter to be replaced. The dual-station circular transfer table raises the supply-side station to the working height of the tool changing robot. The end-effector gripper picks up the hob to be replaced and retracts along a mirror path. Using a force-position hybrid control strategy, the hob to be replaced is moved to the installation station for assembly.

Images