A mine unattended intelligent control platform operation control method and system

Abstract

The present application relates to the technical field of control, in particular to a kind of mine unattended intelligent control platform operation control method and system;Its method includes: dynamic configuration control cluster containing man-machine safety collaborative entity;Fusion personnel position, action and other multi-source data to build control situation map;In digital twin, replay equipment avoidance, path guidance and emergency collaborative protection strategy;Arbitration and execution high-priority safety instruction;According to dynamic risk seamless switching control mode;Through learning optimization safety rules and strategy;The present application takes personnel safety as core control target, realizes the intelligent, initiative, depth cooperation of equipment automation operation and personnel inspection maintenance activity, provides from risk early warning, dynamic avoidance to active protection all-around safety guarantee, effectively solves the safety problem of man-machine coexistence in unattended scene.

Description

Technical Field

[0001] This invention relates to the field of control technology, specifically to an operation control method and system for an unmanned intelligent control platform in a mine. Background Technology

[0002] The unmanned intelligent control platform for mines aims to achieve fully automated operation of the production process. However, complete unmanned operation is not absolute; personnel still need to enter the work area for tasks such as equipment inspection and maintenance. Existing technologies have serious deficiencies in personnel safety coordination, typically employing physical isolation, electronic fences, or simple "presence equals emergency stop" strategies. These strategies interrupt production and do not consider the complex risks in the dynamic interaction between personnel and equipment, such as conflicts between mobile equipment (mine cars) and moving personnel. There is a lack of proactive risk assessment and collaborative protection; the system cannot predict environmental risks (such as roof falls) that threaten personnel, nor can it proactively dispatch nearby equipment to provide emergency protection. The control objective switches abruptly between manned and unmanned states, lacking a smooth and intelligent transition. Personnel guidance is lacking: the system only focuses on equipment control and lacks intelligent guidance for personnel entering the area based on real-time global production status. Personnel walking paths and speeds rely on personal experience, posing a risk of accidentally entering high-risk dynamic work areas. Perception and control are disconnected; even with video surveillance, the linkage between alarm information and the control system is usually one-way and delayed, failing to achieve proactive safety intervention. In view of this, in order to overcome the above-mentioned technical problems, the present invention proposes an operation control method and system for an unmanned intelligent control platform for mines. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, this invention proposes an operation control method and system for an unmanned intelligent control platform for mines, which overcomes the defects of existing unmanned mine systems in that their safety strategies are passive, isolated, and lack coordination when personnel are involved.

[0004] The operation control method of the unmanned intelligent control platform for mines according to the present invention includes the following steps: S1. Setting up dynamic entity configuration and safety rules: Based on production tasks and personnel activity plans, the dynamic configuration includes a cluster of control entities for equipment control entities and human-machine safety collaboration entities, and injects dynamic safety rules associated with personnel status into the human-machine safety collaboration entities. S2. Real-time collection of equipment operation data, environmental data, and personnel status data, which are then processed to generate a control situation diagram that simultaneously covers the dynamic information of equipment, environment, and personnel. S3. Synchronize the control situation diagram to the mine digital twin, and have the control entity perform parallel simulation and evaluation of various strategies, including equipment action adjustment, personnel path guidance and emergency collaborative protection, in the twin to generate the optimal collaborative control instruction set; S4. Arbitrate control commands from different entities based on safety priorities, distribute the arbitration commands for execution, and monitor personnel status and equipment response to form a safety closed loop. S5. Calculate the dynamic risk level based on the control situation diagram, and drive the control system to smoothly switch between fully automatic unmanned mode, human-machine collaborative deceleration mode, local protection mode and remote emergency takeover mode; S6. Based on the data and results of historical human-machine collaboration scenarios, optimize and learn the personnel behavior prediction model, safety rules, and collaboration strategies.

[0005] Preferably, the human-machine safety collaborative entity in S1 is a dedicated software intelligent agent whose dynamic safety rules define at least: speed limit rules based on the distance partition between personnel and equipment, equipment response rules based on personnel action posture, and emergency protection rules that dispatch nearby equipment to form a physical barrier between personnel and risk sources when environmental risks threaten personnel.

[0006] Preferably, the personnel status data in S2 includes real-time location coordinates obtained through a precise positioning system, and personnel action categories, walking speeds, and orientations obtained through a visual analysis unit; the control situation diagram is a structured data object, whose data domains include at least the equipment operating parameter domain, the environmental state domain, the personnel dynamic information domain, and the association domain representing the relative position and state relationship between humans, machines, and the environment.

[0007] Preferably, the emergency collaborative protection strategy simulation in S3 includes: simulating the development trajectory of environmental risk events in a digital twin, identifying threatened personnel, retrieving and calculating the feasibility, time, and effect of moving nearby mobile devices to a preset protection location, and incorporating effective equipment scheduling instructions into the optimal collaborative control instruction set.

[0008] Preferably, the safety instruction priority arbitration in S4 follows a preset level rule that personnel safety instructions take precedence over equipment safety instructions, which in turn take precedence over production process maintenance instructions, ensuring that control instructions directly related to human and machine safety can be executed immediately and unconditionally.

[0009] Preferably, the smooth switching in S5 refers to the system maintaining basic control over associated devices and areas during the mode switching process, without losing control, and automatically generating a control handover list containing the current situation, executed actions, system suggestions, and pending matters for reporting to remote monitoring personnel.

[0010] Preferably, the optimization learning in S6 adopts a method based on reinforcement learning or case reasoning, and uses the outcome feedback of collaborative scenarios, such as safety / insecurity and efficiency impact, to adjust the trigger threshold of safety rules, the weight of strategy selection, and the parameters of the personnel behavior prediction model.

[0011] An operation control system for an unmanned intelligent control platform for mines includes: The dynamic configuration module for control entities is used to dynamically instantiate equipment control entities and human-machine safety collaboration entities from the preset model library based on the received production plan and personnel work plan, and to configure a dynamic safety rule library for the latter. The human-machine status fusion module is used to access and fuse multi-source data from equipment sensors, environmental monitoring units, personnel positioning and visual recognition units to generate a control situation diagram. The digital twin collaborative simulation module is used to maintain the mine digital twin, receive the control situation diagram, provide a parallel simulation environment for the control entity, and output the optimal collaborative control instruction set verified by simulation. The instruction execution module is used to receive the optimal collaborative control instruction set, arbitrate and resolve conflicts according to the preset security priority, distribute the final instructions to the execution agency and personnel guidance interface, and monitor the execution feedback. The mode switching module is used to calculate the dynamic risk level based on the real-time control situation diagram and drive the control system to smoothly switch between different operating modes according to predefined logic. The self-learning module is used to collect data throughout the system's operation cycle, optimize personnel behavior prediction models, safety rule parameters, and decision-making strategies through machine learning algorithms, and feed the updates back to relevant modules.

[0012] Preferably, the digital twin collaborative simulation module includes a strategy simulation sandbox, which can quickly clone and run multiple twin scenario copies based on the current control situation diagram, so as to test the evolution results of different control strategies in the short term in a parallel and isolated manner, and compare and evaluate their performance indicators.

[0013] Preferably, the system further includes a personnel intelligent guidance interface, which is connected to the instruction execution module and is used to receive path planning and speed suggestion instructions from the control entity, convert them into visual, audible or tactile guidance information, and send them to the smart terminal or wearable device carried by the inspection personnel.

[0014] The beneficial effects of this invention are as follows: 1. This invention utilizes digital twin advanced simulation, enabling the system not only to avoid risks but also to proactively plan optimal collaborative strategies, including using mobile devices as dynamic barriers. This proactive safety approach greatly improves the survival rate in extreme situations. Based on the global real-time situation, it plans and adjusts safe paths and speed suggestions for inspection personnel in real time, transforming passive monitoring into proactive navigation and reducing the probability of personnel accidentally entering risk areas from the source.

[0015] 2. Through dynamic risk grading and seamless mode switching mechanism, the system can flexibly adjust the level of automation according to the depth of personnel involvement and the level of risk, which not only ensures the absolute safety of personnel, but also maintains the continuity of production to the greatest extent, making the system behavior more intelligent and flexible. Attached Figure Description

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0017] Figure 1 This is a flowchart of the method in this invention; Figure 2 This is a schematic diagram illustrating the safe collaboration between the control entity cluster and human-machine interface in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the construction principle of the enhanced control situation diagram in an embodiment of the present invention; Figure 4 This is a flowchart of the emergency collaborative protection decision-making process for rockfalls in an embodiment of the present invention. Detailed Implementation

[0018] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0019] like Figure 1-4 As shown, the operation control method of the unmanned intelligent control platform for mines according to the present invention includes the following steps: S1. Setting of entity dynamic configuration and safety rules: Based on production tasks and personnel activity plans, the dynamic configuration includes a cluster of control entities for equipment control entities and human-machine safety collaboration entities, and injects dynamic safety rules associated with personnel status into the human-machine safety collaboration entities; the dynamic safety rules define the mapping relationship between personnel proximity, action posture and equipment operating parameters (such as speed, start and stop), as well as emergency collaborative protection logic. S2. Real-time acquisition of equipment operation data, environmental data, and personnel status data, wherein the personnel status data includes at least personnel coordinates based on a precise positioning system and personnel action categories and orientations based on video behavior analysis; after fusion processing, a control situation diagram is generated that simultaneously covers the dynamic information of equipment, environment, and personnel. S3. Synchronize the control situation diagram to the mine's digital twin. The control entity performs parallel simulation and evaluation of various strategies, including equipment action adjustment, personnel path guidance, and emergency collaborative protection, within the twin to generate the optimal collaborative control instruction set. For the current situation and future predictions (such as personnel movement paths and rockfall trajectories), perform parallel simulation of multiple control strategies, evaluate the impact of these strategies on production efficiency and personnel safety, and generate the optimal collaborative control instruction set. These strategies include: equipment speed reduction and avoidance, dynamic planning and guidance of personnel paths, and, in critical situations, moving equipment to the vicinity of personnel to form a physical barrier. S4. Arbitrate control commands from different entities based on safety priorities, distribute the arbitration commands for execution, and monitor personnel status and equipment response to form a safety closed loop. S5. Calculate the dynamic risk level based on the control situation diagram, and drive the control system to smoothly switch between fully automatic unmanned mode, human-machine collaborative deceleration mode, local protection mode and remote emergency takeover mode; S6. Based on the data and results of historical human-machine collaboration scenarios, optimize and learn the personnel behavior prediction model, safety rules, and collaboration strategies.

[0020] As one embodiment of the present invention, the human-machine safety collaborative entity in S1 is a dedicated software intelligent agent, whose dynamic safety rules are defined at least as: speed limit rules based on the distance partition between personnel and equipment, equipment response rules based on personnel action posture, and emergency protection rules for scheduling nearby equipment to move to form a physical barrier between personnel and risk sources when environmental risks threaten personnel.

[0021] As one embodiment of the present invention, the personnel status data in S2 includes real-time position coordinates obtained through a precise positioning system, and personnel action categories, walking speed and orientation obtained through a visual analysis unit; the control situation diagram is a structured data object, whose data domain includes at least an equipment operating parameter domain, an environmental state domain, a personnel dynamic information domain, and an association domain representing the relative position and state relationship between humans, machines and the environment.

[0022] As one embodiment of the present invention, the emergency collaborative protection strategy simulation in S3 includes: simulating the development trajectory of environmental risk events in a digital twin, identifying threatened personnel, retrieving and calculating the feasibility, time and effect of moving nearby mobile devices to a preset protection location, and incorporating effective equipment scheduling instructions into the optimal collaborative control instruction set.

[0023] As one embodiment of the present invention, the safety instruction priority arbitration in S4 follows a preset level rule that personnel safety instructions are greater than equipment safety instructions, which are greater than production process maintenance instructions, to ensure that control instructions directly related to human and machine safety can be executed immediately and unconditionally.

[0024] As one embodiment of the present invention, the smooth switching in S5 refers to the system maintaining basic control over associated devices and areas during the mode switching process, without causing a loss of control, and automatically generating a control handover list containing the current situation, executed actions, system suggestions, and pending matters, for reporting to remote monitoring personnel.

[0025] As one embodiment of the present invention, the optimization learning in S6 adopts a method based on reinforcement learning or case reasoning, and uses the outcome feedback of collaborative scenarios, such as safety / insecurity and efficiency impact, to adjust the trigger threshold of safety rules, the weight of strategy selection, and the parameters of the personnel behavior prediction model.

[0026] An operation control system for an unmanned intelligent control platform for mines includes: The dynamic configuration module for control entities is used to dynamically instantiate equipment control entities and human-machine safety collaboration entities from the preset model library based on the received production plan and personnel work plan, and to configure a dynamic safety rule library for the latter. The human-machine status fusion module is used to access and fuse multi-source data from equipment sensors, environmental monitoring units, personnel positioning and visual recognition units to generate a control situation diagram. The digital twin collaborative simulation module is used to maintain the mine digital twin, receive the control situation diagram, provide a parallel simulation environment for the control entity, and output the optimal collaborative control instruction set verified by simulation. The instruction execution module is used to receive the optimal collaborative control instruction set, arbitrate and resolve conflicts according to the preset security priority, distribute the final instructions to the execution agency and personnel guidance interface, and monitor the execution feedback. The mode switching module is used to calculate the dynamic risk level based on the real-time control situation diagram and drive the control system to smoothly switch between different operating modes according to predefined logic. The self-learning module is used to collect data throughout the system's operation cycle, optimize personnel behavior prediction models, safety rule parameters, and decision-making strategies through machine learning algorithms, and feed the updates back to relevant modules.

[0027] As one embodiment of the present invention, the digital twin collaborative simulation module includes a strategy simulation sandbox, which can quickly clone and run multiple twin scenario copies based on the current control situation diagram, so as to test the evolution results of different control strategies in the short term in a parallel and isolated manner, and compare and evaluate their performance indicators.

[0028] As one embodiment of the present invention, the system further includes a personnel intelligent guidance interface, which is connected to the instruction execution module and is used to receive path planning and speed suggestion instructions from the control entity, convert them into visual, audible or tactile guidance information, and send them to the smart terminal or wearable device carried by the inspection personnel.

[0029] The principles of this invention will be explained using case studies: When the platform schedules the early morning operation of the crushing system and arranges for personnel to inspect the No. 1 conveyor belt line: Control entity configuration: The control entity dynamic configuration module instantiates the crushing station feeding control entity and the belt conveyor speed coordination entity. At the same time, a human-machine safety coordination entity for the No. 1 belt conveyor area must be instantiated. Safety rule injection: Inject rules into this human-machine safety collaborative entity, for example: Rule 1: If personnel enter the 5-meter warning zone on both sides of the conveyor belt, the conveyor belt speed is limited to 60% of the rated speed; Rule 2: If personnel enter the 2-meter high-risk zone and face the equipment, the conveyor belt brakes smoothly immediately; Rule 3: If the system predicts the risk of falling rocks and personnel are near the falling rock projection area, the nearest mobile equipment (such as a backup flatbed truck) is dispatched to move towards the personnel to the middle position.

[0030] Data Acquisition: The human-machine status fusion module simultaneously receives: 1) the real-time coordinates (x, y, z) of person A from the UWB positioning base station; 2) the walking posture and facing angle of person A identified by the smart camera deployed on the inspection path through AI algorithm; 3) the speed and current sensor data of the No. 1 conveyor belt; 4) the stress and vibration monitoring data of the top plate.

[0031] Situation Construction: This module performs spatiotemporal alignment and correlation on this data. The generated control situation diagram can be represented in terms of data structure as follows: Time: T1 Equipment layer: Equipment ID: 1# conveyor belt, speed: 2.5m / s, status: running; Environmental layer: Area: Crushing station feed inlet, rockfall risk probability: 0.1% Personnel Level: Personnel ID: A, Location: Speed: 0.8 m / s, posture: walking, facing: along the direction of the conveyor belt; Relationship level: Personnel A is 3.2 meters away from conveyor belt #1, which is within the warning zone; The risk of falling rocks from the roof has suddenly increased: Situational Synchronization and Risk Prediction: The digital twin collaborative simulation module synchronizes the situational map at time T1; combining the geological model with real-time monitoring data, the twin predicts that a rockfall may fall at coordinates 3 seconds later. nearby.

[0032] Multi-strategy parallel simulation: Rapid simulation of the human-machine safety collaboration entity in the No. 1 tape machine area in a twin using two strategies: Strategy A (Unmanned Logic): Control the No. 1 conveyor belt to accelerate, so that the part carrying the ore passes through the rockfall point quickly; Strategy B (Human Collaboration Logic): Calculate the location of the rockfall point and personnel A; find that personnel A is within the range of the rockfall splash; the entity immediately queries the nearby mobile devices in the twin and finds an electric flatbed truck in standby mode; simulate the calculation results of the flatbed truck rapidly moving along the preset track to a position between personnel A and the rockfall point, and evaluate its effectiveness as a barrier. Optimized Decision: Simulation shows that strategy B can greatly reduce the risk of personnel injury, and the flatbed truck's movement path does not affect other critical production processes; therefore, the decision generates the optimal collaborative control instruction set: [Instruction 1: Flatbed truck B starts and moves to coordinates...] Command 2: Issue an audible and visual warning to Personnel A's smart bracelet, instructing them to immediately take cover to the left. Command 3: Maintain the original speed of Conveyor 1. Arbitration: The instruction execution module receives the above instruction set; at the same time, it may also receive an instruction from the crushing station feeding control entity requesting an increase in feeding; the arbitrator assigns the human-machine collaborative instruction the highest priority according to the preset rule of personnel safety > equipment safety > production continuity. Collaborative execution: Command 1 is immediately sent to the controller of flatbed truck B; Command 2 is sent to the wristband of personnel A via the industrial wireless network; Command 3 is confirmed; The module synchronously monitors the position feedback of flatbed truck B and the positioning signal of personnel A to confirm that the protective action is being executed. Risk Calculation: The mode switching module calculates the risk level in real time based on the probability of falling rocks, personnel distance, and equipment response status. At this moment, the risk level jumps from low (collaborative mode) to high (local protection mode). Mode switching: The system automatically switches from human-machine collaborative speed reduction mode to local protection mode; in this mode, the production equipment in this area prioritizes safety. The background sends an alarm and mode switching notification to the remote control center, along with a handover list: Flatbed truck B has been automatically started for barrier protection, personnel A has received the warning, and conveyor belt #1 is operating normally. Please remotely monitor and confirm.

[0033] After the event concluded, the self-learning module reviewed the entire process data: Effectiveness evaluation: The degree of conformity between the actual landing point of the falling rocks, the final location of the flatbed truck, the final location of the personnel, and the system's early warning and avoidance suggestions; Model optimization: Based on actual data, fine-tune the air resistance coefficient parameters in the rockfall trajectory prediction model; optimize the decision algorithm for moving the equipment to the barrier position so that the next calculation is faster and more accurate; Rule Evolution: This successful collaborative protection case will be transformed into an empirical rule and stored in the security rule base for reference in similar scenarios.

[0034] The functions, inputs, outputs, and collaborative relationships of each module in this invention; The dynamic configuration module for control entities takes task orders and personnel scheduling plans from the upper-level production management system as input. Its function is to parse tasks and, based on the process flow topology and a pre-set entity model library (such as crushing control entity model, transportation coordination entity model, and regional human-machine safety entity model), dynamically instantiate or unload corresponding software control entities to form a control entity cluster for the current task. When instantiating human-machine safety coordination entities, it loads the corresponding rule set from the safety rule library as its behavioral logic. Its output is the activated control entity cluster, which will serve as the decision-making body for subsequent steps. The human-machine state fusion module takes as input all raw data streams from the mine site, including: I / O signals and bus data from equipment PLCs / sensors (temperature, pressure, speed, position), environmental monitoring sensor data (gas concentration, roof stress), personnel coordinate data from precise positioning systems (such as UWB), and real-time bitstreams from video surveillance systems. Its core function is to perform real-time fusion processing of multi-source heterogeneous data, including timestamp synchronization, spatial coordinate system unification, data validity verification, and confidence weighting, ultimately generating and outputting a unified, structured control situation diagram. This diagram represents the system's authoritative and unified understanding of the current state of the physical world. The digital twin collaborative simulation module takes as input a real-time control situation map from the sensing module. Internally, it maintains a high-fidelity digital twin model consistent with the geometric, physical, and logical relationships of the physical mine. Its functions include: rapidly updating the twin's state after receiving the situation map; providing a parallel and secure simulation environment for the control entity cluster; receiving various control strategies (such as equipment action sequences, personnel guidance plans, and emergency protection scheduling) proposed by each control entity based on the current situation, and performing advanced simulation within the twin; quantitatively evaluating the performance of each strategy under multiple objectives such as safety and efficiency; and outputting the optimal collaborative control instruction set, verified and optimized through simulation. The instruction execution module takes as input the optimal collaborative control instruction set from the decision-making module. This instruction set may contain potentially conflicting instructions from multiple entities. Its core function is to run a safety instruction arbitrator, which sorts and resolves conflicts among concurrent instructions according to preset absolute priority rules (e.g., personal safety protection instructions > equipment protection instructions > process maintenance and optimization instructions). Its output is the arbitrated, conflict-free final control instruction sequence, which is reliably distributed to the corresponding field actuators (e.g., frequency converters, valve controllers, vehicle navigation systems, and personnel intelligent terminals). Simultaneously, this module monitors the feedback signals from the actuators, forming a closed-loop control. Mode Switching Module: Its input is a continuous control situation flow. Its function is to run a risk assessment engine in real time, integrating information such as personnel location, equipment status, environmental threats, and historical accident data to calculate a quantified dynamic risk level. This module has a pre-built control mode state machine that automatically triggers the switching between overall system operating modes (such as fully automatic, collaborative, protective, and remote takeover) based on the real-time risk level. During switching, it outputs mode switching commands to relevant modules and generates a detailed handover list to send to the remote monitoring station, ensuring a smooth, controllable, and transparent switching process. The self-learning module takes as input the historical data chain of the entire system operation cycle, including raw perception data, generated situation maps, decision-making instruction sets, arbitration results, execution feedback, and the final event outcome (whether it is safe or has an impact on efficiency). Its function is to use machine learning algorithms (such as reinforcement learning and deep learning) to analyze this data offline, evaluate the effectiveness of historical decisions, and optimize relevant models accordingly. For example, it optimizes the parameters of the personnel behavior recognition algorithm in the human-machine state fusion module; optimizes the strategy evaluation weights in the digital twin collaborative simulation module; and optimizes the thresholds of the safety rule base in the control entity dynamic configuration module. Its output is the updated model and rule parameters, which are fed back to other modules to achieve continuous iterative evolution of the system's safety collaborative capabilities. The above six modules are connected through a high-speed industrial data bus or message middleware, and work together in publish / subscribe or request / response mode to form a complete human-machine safe collaborative intelligent control system.

[0035] Specific application scenarios include people crossing transport tunnels: Personnel need to traverse a transport tunnel where driverless mining vehicles (AGVs) are running in a loop; Perception and Situation Building: The human-machine state fusion module learns the personnel's intention to enter through the access control system and positioning system at the entrance of the alley, and tracks their position through the camera in the alley; the generated control situation map shows that personnel P is located at the entrance of the alley, AGV-1 is 300 meters away from the entrance, and the speed is 15km / h; Twin simulation and decision-making: In the digital twin collaborative simulation module, the human-machine safety collaborative entity in the tunnel simulates two scenarios in the twin: a) AGV-1 stops suddenly; b) AGV-1 decelerates to 5km / h, while calculating the time required for person P to cross the tunnel at a normal walking speed; the simulation shows that scenario b can ensure safety (AGV-1 arrives after the person crosses) and reduce the energy consumption and equipment wear caused by sudden stops; decision: AGV-1 decelerates and sends a suggestion to person P's wearable device to pass quickly, indicating that the AGV has decelerated. Execution and guidance: The instruction execution module prioritizes the execution of deceleration and guidance instructions; the mode switching module detects that personnel P hesitates slightly, dynamically adjusts the risk level, and broadcasts the report again through the roadway voice system. Learning: The self-learning module records the successful experience of this deceleration event and uses it to optimize the calculation of the best deceleration curve for AGV in similar scenarios in the future.

[0036] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for operating and controlling an unmanned intelligent control platform for mines, characterized in that, Includes the following steps: S1. Setting up dynamic entity configuration and safety rules: Based on production tasks and personnel activity plans, the dynamic configuration includes a cluster of control entities for equipment control entities and human-machine safety collaboration entities, and injects dynamic safety rules associated with personnel status into the human-machine safety collaboration entities. S2. Real-time collection of equipment operation data, environmental data, and personnel status data, which are then processed to generate a control situation diagram that simultaneously covers the dynamic information of equipment, environment, and personnel. S3. Synchronize the control situation diagram to the mine digital twin, and have the control entity perform parallel simulation and evaluation of various strategies, including equipment action adjustment, personnel path guidance and emergency collaborative protection, in the twin to generate the optimal collaborative control instruction set; S4. Arbitrate control commands from different entities based on safety priorities, distribute the arbitration commands for execution, and monitor personnel status and equipment response to form a safety closed loop. S5. Calculate the dynamic risk level based on the control situation diagram, and drive the control system to smoothly switch between fully automatic unmanned mode, human-machine collaborative deceleration mode, local protection mode and remote emergency takeover mode; S6. Based on the data and results of historical human-machine collaboration scenarios, optimize and learn the personnel behavior prediction model, safety rules, and collaboration strategies.

2. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The human-machine safety collaborative entity in S1 is a dedicated software intelligent agent whose dynamic safety rules define at least the following: speed limit rules based on the distance partition between personnel and equipment, equipment response rules based on personnel action posture, and emergency protection rules that dispatch nearby equipment to form a physical barrier between personnel and the risk source when environmental risks threaten personnel.

3. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The personnel status data in S2 includes real-time location coordinates obtained through a precise positioning system, as well as personnel action categories, walking speeds, and orientations obtained through a visual analysis unit; the control situation diagram is a structured data object, whose data domains include at least the equipment operating parameter domain, the environmental state domain, the personnel dynamic information domain, and the association domain representing the relative positions and states of people, machines, and the environment.

4. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The emergency collaborative protection strategy simulation in S3 includes: simulating the development trajectory of environmental risk events in a digital twin, identifying threatened personnel, retrieving and calculating the feasibility, time, and effectiveness of moving nearby mobile devices to a preset protection location, and incorporating effective equipment scheduling instructions into the optimal collaborative control instruction set.

5. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The safety command priority arbitration in S4 follows a preset level rule that personnel safety commands take precedence over equipment safety commands, which in turn take precedence over production process maintenance commands, ensuring that control commands directly related to human and machine safety can be executed immediately and unconditionally.

6. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The smooth switching in S5 refers to the system maintaining basic control over associated devices and areas during mode switching, preventing loss of control, and automatically generating a control handover list that includes the current situation, executed actions, system suggestions, and pending matters for reporting to remote monitoring personnel.

7. The operation control method of the unmanned intelligent control platform for mines according to claim 1, characterized in that, The optimization learning in S6 adopts a method based on reinforcement learning or case reasoning, and uses the outcome feedback of collaborative scenarios, such as safety / insecurity and efficiency impact, to adjust the trigger threshold of safety rules, the weight of strategy selection, and the parameters of the personnel behavior prediction model.

8. An operation control system for an unmanned intelligent control platform in a mine, applicable to the operation control method of the unmanned intelligent control platform in a mine as described in any one of claims 1-7, characterized in that, include: The dynamic configuration module for control entities is used to dynamically instantiate equipment control entities and human-machine safety collaboration entities from the preset model library based on the received production plan and personnel work plan, and to configure a dynamic safety rule library for the latter. The human-machine status fusion module is used to access and fuse multi-source data from equipment sensors, environmental monitoring units, personnel positioning and visual recognition units to generate a control situation diagram. The digital twin collaborative simulation module is used to maintain the mine digital twin, receive the control situation diagram, provide a parallel simulation environment for the control entity, and output the optimal collaborative control instruction set verified by simulation. The instruction execution module is used to receive the optimal collaborative control instruction set, arbitrate and resolve conflicts according to the preset security priority, distribute the final instructions to the execution agency and personnel guidance interface, and monitor the execution feedback. The mode switching module is used to calculate the dynamic risk level based on the real-time control situation diagram and drive the control system to smoothly switch between different operating modes according to predefined logic. The self-learning module is used to collect data throughout the system's operation cycle, optimize personnel behavior prediction models, safety rule parameters, and decision-making strategies through machine learning algorithms, and feed the updates back to relevant modules.

9. The operation control system of the unmanned intelligent control platform for mines according to claim 8, characterized in that, The digital twin collaborative simulation module includes a strategy simulation sandbox, which can quickly clone and run multiple twin scenario copies based on the current control situation diagram, so as to test the evolution results of different control strategies in the short term in a parallel and isolated manner, and compare and evaluate their performance indicators.

10. The operation control system of the unmanned intelligent control platform for mines according to claim 8, characterized in that, The system also includes a personnel intelligent guidance interface, which is connected to the instruction execution module. This interface is used to receive path planning and speed suggestion instructions from the control entity, convert them into visual, audible, or tactile guidance information, and send them to the smart terminal or wearable device carried by the inspection personnel.

Images