A method and system for monitoring and remotely controlling the operating state of a shaft sinking machine

Abstract

The application provides a shaft tunneling machine operation state monitoring and remote control method and system, the method comprising: collecting operation data and video data of the shaft tunneling machine; inputting multi-dimensional synchronous data into a preset tunneling machine three-dimensional digital prototype, driving each joint of the preset tunneling machine three-dimensional digital prototype to act and updating a virtual working face scene; and performing simulation and deduction using a preliminary control instruction; performing item-by-item matching analysis on the simulation deduction report and a preset multi-dimensional safety rule library; performing parameter correction or executing a blocking decision on the preliminary control instruction; inputting a final executable instruction into a controller of the shaft tunneling machine to drive equipment to work; the application can generate a full-condition virtual operation environment that is real-time synchronized with the physical equipment and environment by inputting multi-dimensional synchronous data into the preset tunneling machine three-dimensional digital prototype and driving the joints to act and updating the virtual working face scene, and realizes dynamic and accurate mapping from the physical world to the digital space.

Description

Technical Field

[0001] This invention relates to the field of tunneling machine monitoring technology, and in particular to a method and system for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine. Background Technology

[0002] With the continuous development of underground space, the safety and efficiency requirements for shaft excavation operations are increasing, making real-time monitoring and remote control of their operation a key aspect of ensuring project safety. Due to the complex underground geological conditions and the tight coupling of equipment systems, traditional manual on-site monitoring and local operation modes are high-risk and slow to respond, making remote centralized monitoring systems the mainstream configuration.

[0003] Specifically, in existing technologies, such as Figure 1 As shown, Figure 1 This is a schematic diagram of the main monitoring interface of a remote monitoring system for a shaft boring machine (TBM) in the prior art. While this interface can centrally display key operating parameters of various TBM systems (such as the hydraulic system, propulsion system, cutterhead system, and support shoe system), these parameters are all real-time values ​​at the current moment and cannot reflect the evolution trend of the equipment's state over future periods. Operators can only rely on personal experience to judge these current values. When abnormal fluctuations occur in the parameters, the machine is often already in an abnormal operating condition or even a fault state, lacking quantitative pre-simulation and early warning capabilities.

[0004] Figure 2 This is a schematic diagram of the current alarm interface of a remote monitoring system for shaft boring machines in existing technology. Figure 2 As shown, the interface displays the triggered alarm numbers, times, and descriptions in a list format, which is a typical post-event alarm mechanism. Due to the irreversible nature of the shaft excavation process and the strong coupling of multiple systems, the response chain from risk ignition to alarm triggering to manual intervention is too long, making it impossible to intercept risks in advance and difficult to predict potential cascading failures.

[0005] Figure 3 This is a schematic diagram of the parameter setting interface of a remote monitoring system for shaft boring machines in existing technology. (Example:) Figure 3 As shown, operators need to manually input various static thresholds (such as temperature, pressure, rotational speed, penetration depth, etc.). However, during shaft excavation, geological conditions, tool wear, and equipment status are all changing in real time. Static thresholds cannot adapt to this dynamic environment, often leading to conservative operations affecting excavation efficiency or aggressive operations causing safety risks.

[0006] In summary, most existing remote monitoring systems follow a framework of "data acquisition - status display - manual decision-making." These systems deploy various sensors to transmit key parameters of the tunneling machine's cutterhead drive, propulsion system, hydraulic system, and other components, as well as video footage of the working face, to the ground control center in real time, where it is displayed on a comprehensive screen for operator analysis. Despite improving information centralization, existing systems still face fundamental challenges in practical applications: (1) Lag and passivity in risk response: Although the existing system has improved information concentration, it still mainly relies on monitoring and alarms of the current or past states in risk response, lacking quantitative simulation and pre-emptive safety verification of the evolution trend of the system in multiple future states. Operators can only make judgments based on sensor readings and video images at the current moment. When the system issues an over-limit alarm, the equipment is often already in an abnormal or even faulty state. Especially for operations such as shaft excavation that involve multi-system collaboration and irreversible processes, the response chain from risk budding to alarm triggering to manual intervention is too long, making it impossible to achieve pre-emptive risk interception.

[0007] (2) Complex working conditions and cascading risks are difficult to predict: Although the existing monitoring interface can display multiple parameters in parallel, it lacks in-depth analysis of the coupling relationship and dynamic evolution trend between parameters. Operators cannot make predictions based on experience. For example, under the current geological conditions and propulsion speed, will a slight increase in the cutterhead torque lead to a series of problems such as overheating of the main drive and fluctuations in the hydraulic system? There is a lack of effective prediction and visualization early warning methods for potential faults that cross systems and develop in a chain.

[0008] (3) Mismatch between static preset control strategies and dynamic environment: The control logic of the system largely relies on preset fixed thresholds. However, shaft excavation is a dynamic process, with geological conditions, tool wear, and equipment status all changing in real time. Static control parameters cannot adapt to this dynamic environment, which may lead to conservative operations affecting efficiency or aggressive operations causing risks. Frequent manual adjustments by operators rely heavily on their personal experience and lack unified, objective decision support based on future state projections.

[0009] Therefore, it is necessary to propose a method and system for monitoring and remotely controlling the operating status of a shaft tunneling machine to solve the above-mentioned technical problems. Summary of the Invention

[0010] In view of the above, the main objective of this invention is to provide a method and system for monitoring and remotely controlling the operating status of a shaft tunneling machine, so as to solve the above-mentioned technical problems.

[0011] This invention proposes a method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine, the method comprising the following steps: Step 1: Collect the operating data and video data of the shaft boring machine to generate multi-dimensional synchronous data; Step 2: Input the multi-dimensional synchronous data into the preset 3D digital prototype of the tunneling machine, drive the movement of each joint of the preset 3D digital prototype of the tunneling machine and update the virtual working face scene to generate a full-condition virtual operating environment; Step 3: In the full-condition virtual operation environment, based on the current virtual environment state, generate preliminary control commands, and use the preliminary control commands to perform simulation, and generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period. Step 4: Perform item-by-item matching analysis between the simulation report and the pre-set multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, evaluate whether each rule violation event will trigger a chain reaction of deterioration. Identify and mark all rule violation events that are predicted to be triggered to generate a simulation rule verification list containing event type, severity level, and trigger time point. The parameter influence relationship network is adjusted according to the chain reaction of deterioration event pairs in the simulation rule verification list to correct the time delay parameters and weights between parameter nodes in the parameter influence relationship network, or add new associations to achieve dynamic updates. Step 5: According to the simulation rule verification list, perform parameter correction or execution blocking decision on the preliminary control command to obtain the final executable command or control blocking alarm; Step 6: If it is a control blockage alarm, trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, input the final executable command into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data, and repeat step 1 for subsequent monitoring.

[0012] This invention also proposes a monitoring and remote control system for the operating status of a shaft tunneling machine, the system comprising: The multi-source data synchronous acquisition module is used for: Collect operational and video data from the shaft boring machine to generate multi-dimensional synchronous data; The virtual environment generation module is used for: Multidimensional synchronous data is input into a preset 3D digital prototype of a tunneling machine, which drives the movement of each joint of the preset 3D digital prototype of the tunneling machine and updates the virtual working face scene, generating a full-condition virtual operating environment. The simulation and safety decision-making module is used for: In the full-condition virtual operation environment, based on the current virtual environment state, preliminary control commands are generated, and simulation is performed using the preliminary control commands to generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period. The simulation report is matched and analyzed item by item with a pre-built multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, it is evaluated whether each rule violation event will trigger a chain reaction of deterioration. All rule violation events that are predicted to be triggered are identified and marked to generate a simulation rule verification list containing event type, severity level and trigger time point. The parameter influence relationship network is dynamically updated by correcting the time delay parameters and weights between parameter nodes in the parameter influence relationship network according to the chain reaction of deterioration event pairs in the simulation rule verification list. According to the simulation rule verification list, the parameters of the preliminary control command are corrected or an execution blocking decision is made to obtain the final executable command or control blocking alarm. The instruction execution and control module is used for: If the alarm is a control blockage, it will trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, it will be input into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data for subsequent monitoring.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Existing systems rely on static thresholds for alarms. This invention, through a full-condition virtual operating environment and simulation, transforms the monitoring mode from passive response to proactive prediction. Specifically, existing systems can only trigger alarms after parameters exceed limits (e.g., ...). Figure 2 As shown in the figure, this invention, through a full-condition virtual operating environment and simulation, can generate attitude, load and state change curves for future periods before the control command is issued, realizing the transformation from post-event alarm to pre-event prediction.

[0014] 2. The existing system only displays the current values ​​of each parameter in isolation (e.g., ... Figure 1 As shown in the diagram, it is impossible to predict whether an increase in cutter head torque will lead to a chain reaction causing overheating of the main drive or hydraulic fluctuations. This invention uses a network of influence relationships between parameters to query other related parameters affected by the triggering parameter during rule matching, and analyzes whether they show a worsening trend of violating their own safety rules in the future, thereby achieving quantitative assessment and dynamic determination of the severity level of cross-system chain risks.

[0015] 3. The alarm thresholds in the existing system need to be set manually and cannot be adaptive (e.g., Figure 3As shown in the diagram, when geological conditions change abruptly, the optimal adjustment opportunity is often missed. This invention corrects the time delay parameters and weights between parameter nodes or adds new correlations by verifying the cascading deterioration event pairs in the simulation rule checklist. This enables the parameter influence relationship network to learn the actual causal chains online, with the time delay parameters gradually approximating the real physical delay, and the weights reflecting the causal strength. This dual update mechanism allows the system's prediction accuracy to continuously improve and the false alarm rate to be significantly reduced during long-term operation.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by means of embodiments of the invention. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the main monitoring interface of a shaft boring machine operation status monitoring and remote control system in the existing technology.

[0018] Figure 2 This is a schematic diagram of the current alarm interface of the existing vertical shaft tunneling machine operation status monitoring and remote control system.

[0019] Figure 3 This is a schematic diagram of the parameter setting interface for a current-technical-grade shaft boring machine operation status monitoring and remote control system.

[0020] Figure 4 This is a flowchart illustrating the method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine proposed in this invention.

[0021] Figure 5 This is an overall architecture diagram of the vertical shaft tunneling machine operation status monitoring and remote control method proposed in this invention.

[0022] Figure 6 This is a schematic diagram of the structure of the vertical shaft tunneling machine operation status monitoring and remote control system proposed in this invention. Detailed Implementation

[0023] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0024] These and other aspects of the embodiments of the present invention will become clear from the following description and accompanying drawings. In these descriptions and drawings, some specific embodiments of the present invention are specifically disclosed to provide some ways of implementing the principles of the embodiments of the present invention; however, it should be understood that the scope of the embodiments of the present invention is not limited thereto.

[0025] Please see Figure 4 This invention proposes a method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine, which includes the following steps: Step 1: Collect the operating data and video data of the shaft tunneling machine to generate multi-dimensional synchronous data.

[0026] In this embodiment of the invention, operating data reflecting the mechanical state of the equipment are synchronously collected through current and speed sensors deployed on the cutterhead drive system, pressure and displacement sensors installed on the propulsion cylinder and support shoe cylinder, and temperature sensors distributed on the hydraulic station and reducer. Explosion-proof high-definition cameras installed on the tunneling machine head and working face continuously collect video stream data reflecting the rock wall structure and spoil transport. A high-precision network clock protocol is used to assign a unified millisecond-level timestamp to all sensor and camera data. A data frame alignment algorithm is used to time-match and package physical quantity data with key frames of the video stream at different sampling frequencies, generating a multi-dimensional synchronous data packet with strict spatiotemporal synchronization. This eliminates the state distortion problem caused by asynchronous data in constructing a high-fidelity virtual environment.

[0027] Step 2: Input the multi-dimensional synchronous data into the preset 3D digital prototype of the tunneling machine, drive the movement of each joint of the preset 3D digital prototype of the tunneling machine and update the virtual working face scene to generate a full-condition virtual operating environment.

[0028] Please see Figure 5 In step 2, multi-dimensional synchronous data is input into a preset 3D digital prototype of the tunneling machine, driving the movement of each joint of the preset 3D digital prototype of the tunneling machine and updating the virtual working face scene to generate a full-condition virtual operating environment. The specific steps include the following: The device pose data from position and angle sensors, system pressure data from pressure sensors, and motor operation data from the actuators are parsed and separated from the multidimensional synchronous data. From the video data of multidimensional synchronous data, the contour change features of the working face rock wall and the motion vector features of the slag particles between consecutive frames are extracted to obtain the working face change feature data. The device pose data, the system pressure data, and the motor operation data are input into a preset three-dimensional digital prototype of the tunneling machine. The virtual cutterhead spindle, virtual propulsion cylinder, and virtual support shoe joints in the preset three-dimensional digital prototype of the tunneling machine are driven to move to the position and angle corresponding to the vertical shaft tunneling machine, so as to obtain a virtual tunneling machine frame with synchronized joint pose. Input the working face change feature data into the preset virtual tunnel surrounding rock model, update the surface geometry of the virtual rock wall according to the contour change features in the working face change feature data, and update the position and shape of the virtual slag pile according to the motion vector features of the slag particles in the working face change feature data, so as to obtain a virtual working face scene with updated geometry and material. The virtual tunneling machine frame with synchronized joint posture is placed into the corresponding spatial position of the virtual working face scene with updated geometry and materials. Based on the equipment posture data and motor operation data, the corresponding thrust load and torque load are simulated for the virtual propulsion cylinder and the virtual cutterhead spindle to obtain a preliminary virtual operating environment with load attributes. Spatial consistency verification and correction are performed on the initial virtual operating environment with load attributes to generate a full-condition virtual operating environment.

[0029] The process of performing spatial consistency verification and correction on the initial virtual operating environment with load attributes to generate a full-condition virtual operating environment includes the following steps: In the initial virtual operating environment with load attributes, the minimum distance between the virtual cutting tooth tip on the virtual tunneling machine frame and the virtual rock wall surface in the virtual working face scene is calculated. At the same time, the contact polygon area between the virtual support shoe surface and the corresponding virtual tunnel wall is calculated to obtain the virtual contact state quantification data. The virtual contact state quantification data is compared with the predefined operation contact specifications to identify events such as the virtual cutting tooth tip penetrating the virtual rock wall surface and virtual support shoe poor contact events where the contact polygon area is less than the preset safety threshold, so as to obtain the set of virtual space interference events. Based on each interference event in the set of virtual space interference events, local geometric correction is performed on the virtual rock wall surface in the area where the penetration event occurs in the virtual working face scene with updated geometry and materials. The pose of the corresponding virtual support shoe in the virtual tunneling machine frame with synchronized joint pose is translated and adjusted to obtain the corrected virtual tunneling machine frame and the corrected virtual working face scene. The modified virtual tunneling machine frame and the modified virtual working face scene are recombined, and the virtual contact state is recalculated to confirm that the interference events in the set of virtual space interference events have been eliminated, thus obtaining a virtual operating environment that has passed the spatial consistency verification. The timestamps of the equipment joint pose data, the acquisition sequence number of the system pressure data, and the frame number in the working surface change characteristic data are synchronously associated with the virtual operating environment that has passed the spatial consistency verification, so as to generate a full-condition virtual operating environment.

[0030] In this embodiment of the invention, the equipment pose data, system pressure data, and motor operation data in the multi-dimensional synchronous data packet are separated through data parsing and mapped to the corresponding input ports of a preset three-dimensional digital prototype of the tunneling machine. The dynamics calculation in the digital prototype drives the kinematic models of joints such as the virtual cutterhead spindle and virtual propulsion cylinders based on the input pose data, ensuring their poses are consistent with the physical machine. Simultaneously, based on pressure and current data, the load calculation model assigns corresponding torque and thrust loads to the virtual joints.

[0031] For video data, the pixel displacement vectors of the rock wall contours and the texture change features of the slag area between consecutive frames are extracted by computer vision algorithms. These feature data are then input into the geometric update interface of the virtual tunnel surrounding rock model to dynamically correct the surface mesh details of the virtual rock wall and the three-dimensional shape and position of the virtual slag accumulation. This generates a preliminary virtual operating environment with load attributes that is mapped in real time to the physical world in terms of spatial geometry, motion posture, and load status.

[0032] To further ensure the physical plausibility of the virtual environment, a spatial consistency verification step is introduced. By calculating the collision detection distance between the virtual cutting edge tip and the virtual rock wall surface, as well as the contact area between the virtual support shoe surface and the virtual tunnel wall, non-physical penetration or poor contact areas caused by sensor errors or model simplification are automatically identified. Based on the identified set of virtual spatial interference events, micro-protrusion corrections are made to the relevant virtual rock wall surfaces using local triangular facets, or millimeter-level translation adjustments are made to the virtual support shoe pose to eliminate unreasonable geometric interference. Finally, the verified and corrected virtual components are recombined, and contact calculations are performed again to verify that the interference has been eliminated, forming a self-verified, spatially consistent virtual operating environment. Simultaneously, the acquisition timestamps of the equipment data and the video frame numbers are associated with this environment as metadata, generating a full-condition virtual operating environment with a precise spatiotemporal reference.

[0033] Furthermore, compared to existing monitoring interfaces that only display sensor values, the full-condition virtual operating environment constructed in this step is not simply data visualization, but a dynamic digital twin capable of responding in real time to changes in the physical equipment's state and simultaneously updating the virtual rock wall geometry and the morphology of the excavated soil accumulation. This environment provides a virtual test field with load attributes that has passed spatial consistency verification for subsequent simulation and deduction.

[0034] Step 3: In the full-condition virtual operation environment, based on the current virtual environment state, generate preliminary control commands, and use the preliminary control commands to perform simulation simulation to generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period.

[0035] In step 3, within the full-condition virtual operating environment, preliminary control commands are generated based on the current virtual environment state. These commands are then used for simulation to generate a simulation report on the attitude, load, and state changes of the virtual tunneling machine over future periods. This process includes the following steps: The current load of the virtual cutterhead spindle, the current pressure of the virtual propulsion cylinder, and the current contact hardness of the virtual rock wall are extracted from the virtual operating environment under the full working conditions to form the current working condition state vector. Using the current working condition state vector as input, two sets of candidate control commands with differences in cutterhead rotation speed and propulsion speed are generated according to different tunneling efficiencies to obtain a multi-strategy candidate control command set; Each set of instructions in the multi-strategy candidate control instruction set is sequentially input into the full-condition virtual operating environment for short-cycle simulation, and the change curves of virtual cutterhead torque and virtual propulsion pressure under each set of instructions are recorded to obtain the fast simulation response curve of each candidate instruction. Analyze the fast simulation response curves of each candidate instruction, identify the time point when the key parameter in each curve first exceeds the safety threshold, and mark the candidate instruction that causes any parameter to exceed the threshold as a high-risk instruction to obtain the candidate instruction screening results with high-risk marking; From the candidate commands with high-risk markers, the candidate commands that are not marked as high-risk and have the smallest virtual tunneling machine attitude deviation at the end of the simulation are selected and determined as the initial control commands. Simulation is performed based on the initial control commands, and a simulation report is generated.

[0036] The motion parameters of the virtual cutterhead, virtual propulsion cylinder and virtual support shoe are set with preliminary control commands, and the real-time step-length simulation of the complete operation cycle is performed in the full-condition virtual operation environment. During the real-time step-size simulation, the rotation angle of the virtual tunneling machine around its axis and the horizontal and vertical deflection angles relative to the design axis are sampled and recorded at fixed time intervals to form a future attitude time sequence. During the real-time step-size simulation, the torque of the virtual cutterhead spindle, the pressure of the virtual propulsion cylinder, and the current of the virtual main drive motor are sampled and recorded at the same time intervals to form a future load time sequence. During the real-time step-size simulation, the pressure of key valve ports of the virtual hydraulic system and the equivalent load fluctuation coefficient of the virtual transmission system are sampled and recorded at the same time interval to form a future state time sequence. The future attitude time series, future payload time series, and future state time series are aligned by timestamp and fused to form a future multi-dimensional state evolution dataset. Analyze and extract the time periods when the load exceeds the warning threshold, the time periods when the attitude deviates from the permissible range, and the times when the state parameters change abruptly from the future multi-dimensional state evolution dataset to generate a list of future risk events; Based on the future multi-dimensional state evolution dataset and the list of future risk events, a structured document containing future state prediction curves and risk warning information is generated according to a preset format to obtain a simulation report.

[0037] In this embodiment of the invention, the load torque value of the virtual cutterhead spindle, the pressure value of the virtual propulsion cylinder, and the mesh hardness attribute of the virtual rock wall contact area are read in real time from the full-condition virtual operating environment to form a quantified current working condition state vector. Using this state vector as input, a preset instruction planner generates at least two sets of candidate control instructions with differences in cutterhead rotation speed and propulsion speed based on different control objective weights, such as one set biased towards maximizing tunneling speed and another set biased towards minimizing load fluctuation.

[0038] Each set of candidate instructions is injected into a full-condition virtual operating environment for a short period of ultra-real-time simulation. This rapidly simulates the equipment response within the next few seconds and records the changes in key parameters such as virtual cutterhead torque and virtual propulsion pressure, forming a rapid simulation response curve for each candidate instruction. By analyzing these response curves, the system automatically identifies the point in time when key parameters in each curve first exceed their preset safety thresholds. Any candidate instruction that causes parameters to exceed the thresholds during the simulation cycle is immediately marked as a high-risk instruction and excluded from subsequent optimization.

[0039] Among the remaining candidate commands not marked as high-risk, their resulting virtual tunneling machine attitude deviations at the end of the short-cycle simulation—namely, horizontal deflection and vertical settlement—are further compared. The command parameters that cause the smallest attitude deviation are selected as the initial control commands. After determining the initial control commands, the system initiates a high-fidelity, slow-speed, detailed simulation. The motion parameters of the virtual system are set using the initial control commands, and a simulation step size close to the actual physical process is used to simulate a complete future operation cycle.

[0040] In this detailed simulation, the system samples and records the virtual tunneling machine's attitude angles, load values ​​of each drive component, and key system state parameters at a fixed frequency for multiple consecutive future moments, forming three high-density time series: future attitude, load, and state. These time series are aligned and fused along a unified time axis to construct a future evolution dataset covering multi-dimensional state parameters. Through automatic analysis of this dataset, all periods of load exceeding limits, attitude exceeding tolerances, and abnormal abrupt changes in state parameters are extracted, and a structured list of future risk events is generated. Finally, by integrating the dataset and the risk list, a simulation report containing detailed future state prediction curves and clear risk warnings is generated according to a preset report template.

[0041] Furthermore, unlike existing technologies that only generate an alarm list after parameters exceed limits, the simulation report generated in this embodiment includes future attitude time series, future load time series, and future state time series, enabling quantitative prediction of the tunneling machine's state over future periods. This allows operators or the system to receive clear early warning information before risks actually occur, rather than passively responding after an alarm is triggered.

[0042] Step 4: Perform item-by-item matching analysis between the simulation report and the pre-set multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, evaluate whether each rule violation event will trigger a chain reaction of deterioration. Identify and mark all rule violation events that are predicted to be triggered to generate a simulation rule verification list containing event type, severity level, and trigger time point. The parameter influence relationship network is adjusted according to the chain reaction of deterioration event pairs in the simulation rule verification list to correct the time delay parameters and weights between parameter nodes in the parameter influence relationship network, or add new associations to achieve dynamic updates.

[0043] In step 4, the simulation report is matched and analyzed item by item with the pre-set multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, it is evaluated whether each rule violation event will trigger a chain reaction of deterioration. All rule violation events that are predicted to be triggered are identified and marked to generate a simulation rule verification list containing event type, severity level and trigger time point. Specifically, the steps are as follows: S401. Call the rule engine in the preset multidimensional security rule base to perform a time-series scan on the data of the future multidimensional state evolution dataset in the simulation report; wherein, each rule in the preset multidimensional security rule base is bound to at least one target state parameter, one threshold condition and one time window; S402. During the time-series scanning process, when the evolution data of the target's state parameters within the corresponding time window is found to meet the conditions of the target binding rule, a primary rule violation event record is generated; wherein, the primary rule violation event record includes: the rule identifier of the violation, the triggering parameter, the triggering time point, and the initial severity level; S403. For each primary rule violation event record, based on the predefined parameter influence relationship network in the pre-set multidimensional security rule base, query other related parameters affected by the target triggering parameter in a future specified period, and analyze in the future multidimensional state evolution dataset whether all related parameters show a deterioration trend of violating their own security rules in the affected period, and obtain the analysis results. S404. If the analysis results indicate a cascading worsening trend, the severity level of the corresponding primary rule violation event record will be increased by one level; if the analysis results indicate no cascading worsening trend, the initial severity level of the corresponding primary rule violation event record will remain unchanged, thus forming the approved rule violation event. S405. Sort all approved rule violation events according to their corresponding trigger time points and compile them into a simulation rule verification list.

[0044] Specifically, the parameter influence relationship network is used to correct the time delay parameters and weights between parameter nodes in the network based on the cascading deterioration event pairs in the simulation rule verification list, or to add new associations. This includes: S406. From the currently generated simulation rule verification list, extract all rule violation event pairs marked as having a chain reaction deterioration trend to form an event pair set; wherein, each event pair contains a leading event, a related event, and the measured interval time between the leading event and the related event; S407. For each event pair in the event pair set, check whether there is a directed edge in the current parameter influence relationship network from the parameter node corresponding to the leader event to the parameter node corresponding to the associated event. If it exists, classify the event pair as a verification event pair; if it does not exist, classify the event pair as an extension event pair. S408. For confirmatory event pairs, based on the measured interval time in the confirmatory event pair, update the time delay parameters of the corresponding directed edges in the parameter influence relationship network, and add the influence weight attribute of the corresponding directed edges; after processing all confirmatory event pairs, the network with updated parameters is obtained. S409. For extended event pairs, in the network after parameter update, add a directed edge from the parameter node corresponding to the leader event to the parameter node corresponding to the associated event to obtain the corresponding new edge; based on the measured interval time in the extended event pair, set the initial value of the delay parameter for the corresponding new edge, and assign the initial influence weight attribute to the corresponding new edge; after completing the processing of all extended event pairs, the network with expanded structure is obtained. S4010. Use the expanded network as the parameter influence relationship network for the next execution step S403.

[0045] It should be noted that the parameter influence relationship network is maintained by a separate background optimization service; this service starts after each main process generates a simulation rule verification list. Its workflow is as follows: First, the list is parsed, and all rule violation event pairs with a causal sequence are selected, recording their time differences. Second, each event pair is compared with the current network graph; if an edge already exists between the two points, it is classified as a verification pair; otherwise, it is classified as an expansion pair. Then, all verification pairs are traversed, and the theoretical impact delay of the corresponding edge is updated using a moving average algorithm based on its time difference, and the confidence weight of the edge is increased. Next, all expansion pairs are traversed; if their frequency exceeds a preset threshold, a new edge is added to the network, and its delay and weight attributes are initialized with the current time difference. Finally, the updated network graph is persistently stored, and the main process is notified to load it as the parameter influence relationship network in the next risk analysis.

[0046] In this embodiment of the invention, the simulation rule verification process first calls the rule engine in the pre-set multi-dimensional security rule base to perform a time-series scan on the future multi-dimensional state evolution dataset in the simulation simulation report. Each rule is bound to target state parameters, threshold conditions, and a time window.

[0047] When a time-series scan identifies that the evolution data of a certain state parameter within a corresponding time window meets the rule conditions, the system generates a primary rule violation event record, including the rule identifier, triggering parameter, triggering time, and initial severity level. For each primary rule violation event record, the system queries other related parameters that the triggering parameter may affect in the future time period based on a pre-defined network of influence relationships between parameters, and analyzes whether these related parameters show a worsening trend of violating their own safety rules in the future multi-dimensional state evolution dataset. If a cascading worsening trend exists, the severity level of the event is increased by one level; otherwise, the initial level is maintained. All verified rule violation events are sorted by triggering time and compiled to generate a simulation rule verification list.

[0048] Furthermore, after each generation of the simulation rule verification list, the system automatically triggers an update process for the parameter influence relationship network: extracting all rule violation event pairs with a cascading deterioration trend from the list and matching them with the current network. If a corresponding directed edge already exists for the event pair in the network, the theoretical impact delay of the edge is corrected based on the measured interval time, and its impact weight is increased; if no corresponding edge exists, a new directed edge is added, and its delay and weight attributes are initialized with the measured interval time. The updated network will be used in the next risk analysis to achieve adaptive expansion of the network structure and parameter optimization.

[0049] Furthermore, compared to existing static threshold alarms and knowledge graph methods that only allow adding new constraints, this step differs in two aspects: First, during rule matching, instead of checking individual rules in isolation, it queries other parameters associated with the triggering parameter based on the network of influence relationships between parameters and analyzes whether there is a cascading deterioration trend, thereby dynamically determining the severity level; Second, the network update mechanism distinguishes between confirmatory event pairs and expansion event pairs, correcting the delay parameters and increasing the weights for confirmatory event pairs, and adding directed edges for expansion event pairs. This dual update mechanism enables the network to simultaneously achieve parameter refinement and structural expansion.

[0050] Step 5: According to the simulation rule verification list, perform parameter correction or execution blocking decisions on the preliminary control instructions to obtain the final executable instructions or control blocking alarms.

[0051] In step 5, S501, according to the simulation rule verification list, the number of rule violation events marked as the highest severity level in the list is counted, and the tunneling machine subsystems associated with these events are identified, so as to obtain the statistics of the highest level risk events and the list of associated systems; S502. Based on the statistics of the highest-level risk events and the list of related systems, make a decision branch judgment. If the number of the highest-level severity events is greater than zero, proceed to step S506; otherwise, proceed to step S503. S503. Extract all events that are not at the highest severity level from the simulation rule verification list, and analyze the rule violation type, associated control parameters and parameter deviations corresponding to each event to obtain a list of parameters and deviations to be corrected. S504. Taking the list of parameters to be corrected and the deviation as input, query the preset command parameter correction mapping table to generate preliminary parameter correction suggestions; wherein, the preset command parameter correction mapping table defines the correction direction and step size adjustment strategy for different control parameters for various rule violations. S505. Apply the preliminary parameter correction suggestion to the corresponding parameters of the preliminary control instruction to generate the corrected control instruction parameters, and determine whether all parameters in the corrected control instruction parameters are within their respective safe allowable ranges; if yes, compile the corrected control instruction parameters into the final executable instruction; if no, proceed to step S506. S506. Generate a control blocking alarm that includes a summary of all risk events in the simulation rule verification list, the decision basis, and a timestamp.

[0052] In step 5, the simulation rule verification list is first read, and the number of rule violation events marked as the highest severity level is counted. If any highest-severity event exists, the system immediately determines that there is an unacceptable system-level risk, and the decision-making process jumps directly to generating a control blocking alarm. If no highest-severity event exists, the system enters the parameter correction path. In the correction path, the system parses all medium and low severity events in the list, extracts the specific control parameters associated with each event and their deviations and directions relative to the safety threshold, forming a list of parameters to be corrected and their deviations. The system uses the list of parameters to be corrected and their deviations as input to query a preset instruction parameter correction mapping table.

[0053] This mapping table is stored in a rule-knowledge format, explicitly specifying which control parameter (such as cutterhead speed or feed rate) should be increased or decreased when a specific type of rule violation occurs, as well as the recommended single adjustment step size (such as the speed adjustment per minute). Based on this table, preliminary parameter correction suggestions are generated for all current low-to-medium risk events. Subsequently, these correction suggestions are applied one by one to the corresponding parameters of the preliminary control commands, generating a set of corrected candidate control command parameters. The system immediately performs a safety check on this set of new parameters to determine whether all parameters are within their independent safe operating range. If the check passes, the corrected parameters are compiled into the final executable command; if the check fails, it indicates that simple correction cannot resolve all conflicts simultaneously, and a control blockage alarm is triggered to ensure that no commands with potential safety hazards are output.

[0054] Step 6: If it is a control blockage alarm, trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, input the final executable command into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data, and repeat step 1 for subsequent monitoring.

[0055] In this embodiment of the invention, the closed-loop control process is executed according to the above-mentioned decision results. If the decision result is a control interruption alarm, the system immediately triggers an audible and visual alarm on the human-machine interface and pops up an alarm details window, displaying a risk summary and decision basis in the simulation rule verification list. Simultaneously, it sends an instruction lock signal to the underlying controller to ensure that the initial control instruction is absolutely prohibited from being issued. If the decision result is a final executable instruction, the system reliably sends the safety parameter set of the instruction to the programmable logic controller (PLC) of the underground tunneling machine via an industrial real-time Ethernet network. The PLC drives the frequency converter, proportional valve, and other actuators according to the instruction, causing the physical equipment to begin operating according to the verified safety strategy.

[0056] After the equipment performs an action, its state changes, and new operating data and working face video are collected in real time and used as input for the next control cycle. This data is then transmitted back to the initial data acquisition terminal, thus initiating a new round of "perception-simulation-decision-execution" intelligent closed loop. This enables continuous, adaptive, and forward-looking safety monitoring and optimized control of shaft excavation operations.

[0057] Furthermore, unlike existing technologies that rely on manual operators to observe and adjust parameters in real time, this embodiment achieves a closed-loop mechanism by automatically correcting parameters or making blocking decisions, ensuring that control commands are verified and optimized in a virtual environment before execution. When geological conditions change abruptly or equipment status changes, the system can automatically generate corrected control commands or trigger blocking alarms without manual intervention, thus overcoming the shortcomings of static threshold control, such as delayed response and inability to adapt to dynamic working conditions.

[0058] To further illustrate the application process and technical effects of this invention in practical engineering, a specific application example is given below. Example

[0059] The applicant applied this invention to an underground shaft excavation project. The shaft was designed to be 180 meters deep, traversing the following strata: upper clay (0-45 meters), middle gravel layer (45-110 meters), and lower medium-hardness rock layer (110-180 meters). The tunneling machine was equipped with a cutterhead drive power of 450kW, a propulsion cylinder stroke of 1.2 meters, and a total of 8 support shoes across the upper and lower layers.

[0060] During the initial construction phase (0-45 meters clay layer), the system operates according to the following procedure: Step 1: Collect operating data such as cutter head rotation speed, feed pressure, and support shoe displacement at a frequency of 100Hz. At the same time, collect video of the working face at a frequency of 100Hz (i.e., 100 frames per second). After aligning with the timestamps, generate a multi-dimensional synchronization data packet.

[0061] Step 2: Multi-dimensional synchronous data drives the preset 3D digital prototype, synchronizing the poses of the virtual cutterhead spindle, virtual propulsion cylinder, and virtual support shoe with the physical machine; simultaneously, the surface geometry of the virtual rock wall and the morphology of the excavated soil accumulation are updated based on video data. No penetration events were detected during spatial consistency verification, and a full-condition virtual operating environment was directly generated.

[0062] Step 3: Extract the current load (cutterhead torque 215 kN·m, propulsion pressure 18 MPa) and rock wall hardness characteristic value (dimensionless, 25 for clay layer) from the environment to generate preliminary control commands (cutterhead rotation speed 12 rpm, propulsion speed 45 mm / min). After short-cycle rapid simulation screening, it is confirmed that the command has no high-risk markers, and then a fine simulation is performed. The fine simulation outputs the cutterhead torque change curve and propulsion pressure fluctuation curve for the next 60 seconds, predicting that the peak torque will not exceed 250 kN·m and the propulsion pressure fluctuation range will be within ±2 MPa.

[0063] Step 4: Match the simulation report with the multi-dimensional safety rule base. The rule base has a pre-set level 2 alarm for "cutterhead torque exceeding 280 kN·m for more than 3 seconds". The scan found that the predicted torque peak was 250 kN·m, which did not trigger an alarm. However, there is a directed edge in the parameter influence relationship network from "rock wall hardness" to "cutterhead torque", with an initial time delay parameter of 5 seconds. Since the current rock wall hardness changes gradually and no cascading deterioration events have occurred, the simulation rule verification list is empty.

[0064] If the list is empty, the initial control instructions are directly compiled into final executable instructions and sent to the controller to drive the device to work.

[0065] When construction entered the gravel layer (75 meters deep), the rock wall hardness characteristic value suddenly increased (from 25 to 58). During a new cycle, the system's detailed simulation predicted that the cutterhead torque would increase from the current 238 kN·m to 295 kN·m within 15 seconds, triggering a level two alarm. A parameter influence network analysis revealed that "cutterhead torque" affects "main drive current," and the "main drive current" rule itself states that exceeding the rated value by 1.1 times for 5 seconds triggers a level three alarm. Analysis of the predicted data showed that 8 seconds after the cutterhead torque exceeded the limit, the main drive current would exceed the rated value by 1.1 times and remain there for 6 seconds, indicating a cascading deterioration trend. Therefore, the system upgraded the severity level of the cutterhead torque alarm from level two to level one and generated a simulation rule verification list containing this event.

[0066] Step 5: Due to the existence of a Level 1 severe event, the system executes a blocking decision, generates a control blocking alarm, and prevents the issuance of the initial control command (the original recommended feed speed remains at 45 mm / min). Simultaneously, the alarm message suggests "reducing the feed speed to 30 mm / min or reducing the cutter head speed to 10 rpm."

[0067] After operator confirmation, the system issued the revised control command (feed speed reduced to 30 mm / min). The equipment operated according to the new command, and the measured cutterhead torque remained stable between 245-260 kN·m without triggering any alarms.

[0068] During the subsequent construction of the project, the parameter influence relationship network was automatically updated three times based on the actual chain of deterioration events: First, the time delay parameter of the directed edge "rock wall hardness - cutterhead torque" was corrected from 5s to 4.2s; second, the weight of the directed edge "cutterhead torque - main drive current" was increased from 0.6 to 0.8; third, it was found that there were two consecutive accompanying relationships between "increased propulsion speed" and "lateral slippage of the support shoe", and there was no corresponding edge in the network. The system automatically added a directed edge with the time delay parameter initialized to 3.5s and the weight 0.3.

[0069] The entire project involved a total excavation depth of 180 meters. The system generated 320 simulation reports, including 47 triggers of simulation rule verification lists, 12 executions of blocking decisions, and 35 parameter corrections. Compared to the previous vertical shaft project in the same area using a traditional static threshold monitoring system (150-meter excavation depth, 8 unplanned shutdowns, and a false alarm rate of 27%), this project had only one unplanned shutdown (due to an external power supply failure) and a false alarm rate of only 4.2%. The average number of manual interventions per shift decreased from 11 in the traditional system to 2. Especially in sections with abrupt changes in rock strata, this invention provides an early warning 15 to 40 seconds in advance, allowing operators ample response time, while the traditional system can only alarm after exceeding limits, resulting in cutterhead jamming twice.

[0070] This embodiment 1 demonstrates that the present invention not only fully realizes the closed-loop control of steps 1 to 6 in practical applications, but also significantly reduces the false alarm rate and the number of unplanned shutdowns through dynamic updates of the parameter influence relationship network, achieving a significantly better early warning and adaptive control effect than the prior art.

[0071] Please see Figure 6 The present invention also proposes a monitoring and remote control system for the operating status of a vertical shaft tunneling machine, the system comprising: The multi-source data synchronous acquisition module is used for: Collect operational and video data from the shaft boring machine to generate multi-dimensional synchronous data; The virtual environment generation module is used for: Multidimensional synchronous data is input into a preset 3D digital prototype of a tunneling machine, which drives the movement of each joint of the preset 3D digital prototype of the tunneling machine and updates the virtual working face scene, generating a full-condition virtual operating environment. The simulation and safety decision-making module is used for: In the full-condition virtual operation environment, based on the current virtual environment state, preliminary control commands are generated, and simulation is performed using the preliminary control commands to generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period. The simulation report is matched and analyzed item by item with a pre-built multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, it is evaluated whether each rule violation event will trigger a chain reaction of deterioration. All rule violation events that are predicted to be triggered are identified and marked to generate a simulation rule verification list containing event type, severity level and trigger time point. The parameter influence relationship network is dynamically updated by correcting the time delay parameters and weights between parameter nodes in the parameter influence relationship network according to the chain reaction of deterioration event pairs in the simulation rule verification list. According to the simulation rule verification list, the parameters of the preliminary control command are corrected or an execution blocking decision is made to obtain the final executable command or control blocking alarm. The instruction execution and control module is used for: If the alarm is a control blockage, it will trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, it will be input into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data for subsequent monitoring.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0074] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine, characterized in that, The method includes the following steps: Step 1: Collect the operating data and video data of the shaft boring machine to generate multi-dimensional synchronous data; Step 2: Input the multi-dimensional synchronous data into the preset 3D digital prototype of the tunneling machine, drive the movement of each joint of the preset 3D digital prototype of the tunneling machine and update the virtual working face scene to generate a full-condition virtual operating environment; Step 3: In the full-condition virtual operation environment, based on the current virtual environment state, generate preliminary control commands, and use the preliminary control commands to perform simulation, and generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period. Step 4: Perform item-by-item matching analysis between the simulation report and the pre-set multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, evaluate whether each rule violation event will trigger a chain reaction of deterioration. Identify and mark all rule violation events predicted to be triggered to generate a simulation rule verification list containing event type, severity level, and trigger time point. Specifically, the parameter influence relationship network, based on the chain reaction of deterioration event pairs in the simulation rule verification list, corrects the time delay parameters and weights between parameter nodes in the parameter influence relationship network, or adds new associations to achieve dynamic updates. This includes: From the currently generated simulation rule verification list, extract all rule violation event pairs marked as having a chain reaction deterioration trend to form an event pair set; where each event pair contains a leading event, a related event, and the measured interval between the leading event and the related event; For each event pair in the event pair set, check whether there is a directed edge in the current parameter influence relationship network from the parameter node corresponding to the leading event to the parameter node corresponding to the associated event. If it exists, classify the event pair as a verification event pair; if it does not exist, classify the event pair as an extension event pair. For confirmatory event pairs, based on the measured interval time in the confirmatory event pair, the time delay parameters of the corresponding directed edges in the parameter influence relationship network are updated, and the influence weight attribute of the corresponding directed edges is added; after processing all confirmatory event pairs, the parameter-updated network is obtained. For extended event pairs, in the network after parameter update, a directed edge is added from the parameter node corresponding to the leader event to the parameter node corresponding to the associated event, resulting in the corresponding new edge; based on the measured interval time in the extended event pair, the initial value of the delay parameter is set for the corresponding new edge, and the initial influence weight attribute is assigned to the corresponding new edge; after processing all extended event pairs, the network with expanded structure is obtained. Step 5: According to the simulation rule verification list, perform parameter correction or execution blocking decision on the preliminary control command to obtain the final executable command or control blocking alarm; Step 6: If it is a control blockage alarm, trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, input the final executable command into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data, and repeat step 1 for subsequent monitoring.

2. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 1, characterized in that, In step 2, multi-dimensional synchronous data is input into a preset 3D digital prototype of the tunneling machine, driving the movement of each joint of the preset 3D digital prototype of the tunneling machine and updating the virtual working face scene to generate a full-condition virtual operating environment. The specific steps include the following: The device pose data from position and angle sensors, system pressure data from pressure sensors, and motor operation data from the actuators are parsed and separated from the multidimensional synchronous data. From the video data of multidimensional synchronous data, the contour change features of the working face rock wall and the motion vector features of the slag particles between consecutive frames are extracted to obtain the working face change feature data. The device pose data, the system pressure data, and the motor operation data are input into a preset three-dimensional digital prototype of the tunneling machine. The virtual cutterhead spindle, virtual propulsion cylinder, and virtual support shoe joints in the preset three-dimensional digital prototype of the tunneling machine are driven to move to the position and angle corresponding to the vertical shaft tunneling machine, so as to obtain a virtual tunneling machine frame with synchronized joint pose. Input the working face change feature data into the preset virtual tunnel surrounding rock model, update the surface geometry of the virtual rock wall according to the contour change features in the working face change feature data, and update the position and shape of the virtual slag pile according to the motion vector features of the slag particles in the working face change feature data, so as to obtain a virtual working face scene with updated geometry and material. The virtual tunneling machine frame with synchronized joint posture is placed into the corresponding spatial position of the virtual working face scene with updated geometry and materials. Based on the equipment posture data and motor operation data, the corresponding thrust load and torque load are simulated for the virtual propulsion cylinder and the virtual cutterhead spindle to obtain a preliminary virtual operating environment with load attributes. Spatial consistency verification and correction are performed on the initial virtual operating environment with load attributes to generate a full-condition virtual operating environment.

3. The method for monitoring and remotely controlling the operating status of a shaft boring machine according to claim 2, characterized in that, The initial virtual operating environment with load attributes is subjected to spatial consistency verification and correction to generate a full-condition virtual operating environment. The specific steps include the following: In the initial virtual operating environment with load attributes, the minimum distance between the virtual cutting tooth tip on the virtual tunneling machine frame and the virtual rock wall surface in the virtual working face scene is calculated. At the same time, the contact polygon area between the virtual support shoe surface and the corresponding virtual tunnel wall is calculated to obtain the virtual contact state quantification data. The virtual contact state quantification data is compared with the predefined operation contact specifications to identify events such as the virtual cutting tooth tip penetrating the virtual rock wall surface and virtual support shoe poor contact events where the contact polygon area is less than the preset safety threshold, so as to obtain the set of virtual space interference events. Based on each interference event in the set of virtual space interference events, local geometric correction is performed on the virtual rock wall surface in the area where the penetration event occurs in the virtual working face scene with updated geometry and materials. The pose of the corresponding virtual support shoe in the virtual tunneling machine frame with synchronized joint pose is translated and adjusted to obtain the corrected virtual tunneling machine frame and the corrected virtual working face scene. The modified virtual tunneling machine frame and the modified virtual working face scene are recombined, and the virtual contact state is recalculated to confirm that the interference events in the set of virtual space interference events have been eliminated, thus obtaining a virtual operating environment that has passed the spatial consistency verification. The timestamps of the equipment joint pose data, the acquisition sequence number of the system pressure data, and the frame number in the working surface change characteristic data are synchronously associated with the virtual operating environment that has passed the spatial consistency verification, so as to generate a full-condition virtual operating environment.

4. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 3, characterized in that, In step 3, within the full-condition virtual operating environment, preliminary control commands are generated based on the current virtual environment state. These commands are then used for simulation to generate a simulation report on the attitude, load, and state changes of the virtual tunneling machine over future periods. This process includes the following steps: The current load of the virtual cutterhead spindle, the current pressure of the virtual propulsion cylinder, and the current contact hardness of the virtual rock wall are extracted from the virtual operating environment under the full working conditions to form the current working condition state vector. Using the current working condition state vector as input, two sets of candidate control commands with differences in cutterhead rotation speed and propulsion speed are generated according to different tunneling efficiencies to obtain a multi-strategy candidate control command set; Each set of instructions in the multi-strategy candidate control instruction set is sequentially input into the full-condition virtual operating environment for short-cycle simulation, and the change curves of virtual cutterhead torque and virtual propulsion pressure under each set of instructions are recorded to obtain the fast simulation response curve of each candidate instruction. Analyze the fast simulation response curves of each candidate instruction, identify the time point when the key parameter in each curve first exceeds the safety threshold, and mark the candidate instruction that causes any parameter to exceed the threshold as a high-risk instruction to obtain the candidate instruction screening results with high-risk marking; From the candidate commands with high-risk markers, the candidate commands that are not marked as high-risk and have the smallest virtual tunneling machine attitude deviation at the end of the simulation are selected and determined as the initial control commands. Simulation is performed based on the initial control commands, and a simulation report is generated.

5. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 4, characterized in that, Based on the initial control commands, a simulation is performed to generate a simulation report, which includes the following steps: The motion parameters of the virtual cutterhead, virtual propulsion cylinder and virtual support shoe are set with preliminary control commands, and the real-time step-length simulation of the complete operation cycle is performed in the full-condition virtual operation environment. During the real-time step-size simulation, the rotation angle of the virtual tunneling machine around its axis and the horizontal and vertical deflection angles relative to the design axis are sampled and recorded at fixed time intervals to form a future attitude time sequence. During the real-time step-size simulation, the torque of the virtual cutterhead spindle, the pressure of the virtual propulsion cylinder, and the current of the virtual main drive motor are sampled and recorded at the same time intervals to form a future load time sequence. During the real-time step-size simulation, the pressure of key valve ports of the virtual hydraulic system and the equivalent load fluctuation coefficient of the virtual transmission system are sampled and recorded at the same time interval to form a future state time sequence. The future attitude time series, future payload time series, and future state time series are aligned by timestamp and fused to form a future multi-dimensional state evolution dataset. Analyze and extract the time periods when the load exceeds the warning threshold, the time periods when the attitude deviates from the permissible range, and the times when the state parameters change abruptly from the future multi-dimensional state evolution dataset to generate a list of future risk events; Based on the future multi-dimensional state evolution dataset and the list of future risk events, a structured document containing future state prediction curves and risk warning information is generated according to a preset format to obtain a simulation report.

6. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 5, characterized in that, In step 4, the simulation report is matched and analyzed item by item with the pre-set multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, it is evaluated whether each rule violation event will trigger a chain reaction of deterioration. All rule violation events that are predicted to be triggered are identified and marked to generate a simulation rule verification list containing event type, severity level and trigger time point. Specifically, the steps are as follows: S401. Call the rule engine in the preset multidimensional security rule base to perform a time-series scan on the data of the future multidimensional state evolution dataset in the simulation report; wherein, each rule in the preset multidimensional security rule base is bound to at least one target state parameter, one threshold condition and one time window; S402. During the time-series scanning process, when the evolution data of the target's state parameters within the corresponding time window is found to meet the conditions of the target binding rule, a primary rule violation event record is generated; wherein, the primary rule violation event record includes: the rule identifier of the violation, the triggering parameter, the triggering time point, and the initial severity level; S403. For each primary rule violation event record, based on the predefined parameter influence relationship network in the pre-set multidimensional security rule base, query other related parameters affected by the target triggering parameter in a future specified period, and analyze in the future multidimensional state evolution dataset whether all related parameters show a deterioration trend of violating their own security rules in the affected period, and obtain the analysis results. S404. If the analysis results indicate a cascading worsening trend, the severity level of the corresponding primary rule violation event record will be increased by one level; if the analysis results indicate no cascading worsening trend, the initial severity level of the corresponding primary rule violation event record will remain unchanged, thus forming the approved rule violation event. S405. Sort all approved rule violation events according to their corresponding trigger time points and compile them into a simulation rule verification list.

7. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 6, characterized in that, The parameter influence relationship network is adjusted based on the cascading deterioration event pairs in the simulation rule checklist, correcting the time delay parameters and weights between parameter nodes in the network, or adding new associations. Specifically, this includes: The expanded network will be used as the parameter influence relationship network for the next execution step S403.

8. The method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine according to claim 6, characterized in that, In step 5, based on the simulation rule verification list, the parameters of the preliminary control command are corrected or an execution blocking decision is made to obtain the final executable command or control blocking alarm. Specifically, this includes the following steps: S501. Based on the simulation rule verification list, count the number of rule violation events marked as the highest severity level in the list, and identify the tunneling machine subsystems associated with these events to obtain the statistics of the highest level risk events and the list of associated systems. S502. Based on the statistics of the highest level of risk events and the list of related systems, make a decision branch judgment. If the number of the highest severity level events is greater than zero, then proceed to step S506. Otherwise, proceed to step S503; S503. Extract all events that are not at the highest severity level from the simulation rule verification list, and analyze the rule violation type, associated control parameters and parameter deviations corresponding to each event to obtain a list of parameters and deviations to be corrected. S504. Taking the list of parameters to be corrected and the deviation as input, query the preset command parameter correction mapping table to generate preliminary parameter correction suggestions; wherein, the preset command parameter correction mapping table defines the correction direction and step size adjustment strategy for different control parameters for various rule violations. S505. Apply the preliminary parameter correction suggestion to the corresponding parameters of the preliminary control instruction to generate the corrected control instruction parameters, and determine whether all parameters in the corrected control instruction parameters are within their respective safe allowable ranges; if yes, compile the corrected control instruction parameters into the final executable instruction; if no, proceed to step S506. S506. Generate a control blocking alarm that includes a summary of all risk events in the simulation rule verification list, the decision basis, and a timestamp.

9. A monitoring and remote control system for the operating status of a vertical shaft tunneling machine, characterized in that, The system employs the method for monitoring and remotely controlling the operating status of a vertical shaft tunneling machine as described in any one of claims 1-8, and the system includes: The multi-source data synchronous acquisition module is used for: Collect operational and video data from the shaft boring machine to generate multi-dimensional synchronous data; The virtual environment generation module is used for: Multidimensional synchronous data is input into a preset 3D digital prototype of a tunneling machine, which drives the movement of each joint of the preset 3D digital prototype of the tunneling machine and updates the virtual working face scene, generating a full-condition virtual operating environment. The simulation and safety decision-making module is used for: In the full-condition virtual operation environment, based on the current virtual environment state, preliminary control commands are generated, and simulation is performed using the preliminary control commands to generate a simulation report on the attitude, load and state changes of the virtual tunneling machine in the future period. The simulation report is matched and analyzed item by item with a pre-built multidimensional security rule base. Based on the parameter influence relationship network contained in the multidimensional security rule base, it is evaluated whether each rule violation event will trigger a chain reaction of deterioration. All rule violation events that are predicted to be triggered are identified and marked to generate a simulation rule verification list containing event type, severity level and trigger time point. The parameter influence relationship network is dynamically updated by correcting the time delay parameters and weights between parameter nodes in the parameter influence relationship network according to the chain reaction of deterioration event pairs in the simulation rule verification list. According to the simulation rule verification list, the parameters of the preliminary control command are corrected or an execution blocking decision is made to obtain the final executable command or control blocking alarm. The instruction execution and control module is used for: If the alarm is a control blockage, it will trigger the alarm and prevent the issuance of the initial control command; if it is a final executable command, it will be input into the controller of the shaft tunneling machine to drive the equipment to work, and collect the next round of shaft tunneling machine operation data and video data for subsequent monitoring.

Images