Control method and system under combined transportation scene of steelmaking travelling crane and trolley

Abstract

The invention provides a control method and system under a steelmaking travelling crane and trolley combined transportation scene. The control method comprises the steps that S1, a virtual element cosmic space is constructed based on real-time running state data of a steelmaking travelling crane and a trolley and workshop environment data; s2, based on the obtained real-time running state data of the steelmaking travelling crane and the trolley and the workshop environment data, combining a production plan and task requirements, and obtaining an optimal transportation strategy through reinforcement learning and ant colony algorithm collaboration; and S3, the current optimal transportation strategy is evaluated based on the virtual element cosmic space, and when the current optimal transportation strategy meets the preset requirement, the steelmaking crane and the trolley are controlled based on the current optimal combined transportation strategy.

Description

Technical Field

[0001] This invention relates to the field of combined transport control technology, specifically to a control method and system for combined transport scenarios of steelmaking overhead cranes and trolleys, and more specifically to a method and system for metaverse information space mapping and control in combined transport scenarios of steelmaking overhead cranes and trolleys. Background Technology

[0002] In current steelmaking workshops, overhead cranes and trolleys are the core transportation equipment for key materials such as molten iron, ladles, and billets. Their scheduling and control effects directly affect production efficiency, energy consumption, and operational safety.

[0003] In existing technologies, most steelmaking workshops still rely on manual scheduling or partial automation of single equipment: manual scheduling relies on experience-based decision-making, which is prone to path conflicts and long equipment waiting times due to information lag and insufficient overall planning, and is also labor-intensive; while automation of single equipment can achieve precise operation of a single machine, it lacks the coordinated linkage of overhead cranes and trolleys, and cannot adapt to complex scenarios of multiple equipment operating simultaneously.

[0004] Patent document CN112446642A (application number: CN202011456180.6) discloses a method and system for optimizing the scheduling of multiple overhead cranes and trolleys. By matching tasks and overhead cranes / trolleys, a dynamic scheduling model for multiple overhead cranes / trolleys is established. A dual-tangent crossover genetic algorithm based on real-number encoding is used to calculate the optimal crane driving scheme for the multiple overhead cranes / trolleys in the steelmaking-refining-continuous casting production process, effectively reducing the number of ineffective crane yields and improving the effective operating rate of the overhead cranes / trolleys. However, this method does not involve the construction of a metaverse virtual space and immersive interaction, making it unable to achieve high real-time response control under combined paths of ladles, overhead cranes, and trolleys. It is difficult to provide operators with intuitive holographic monitoring and control capabilities, and the depth of collaborative control between overhead cranes and trolleys is limited. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide a control method and system for the combined transportation of steelmaking overhead cranes and trolleys.

[0006] A control method for a combined transport scenario of a steelmaking overhead crane and a trolley, provided by the present invention, includes: Step S1: Construct a virtual metaverse space based on the real-time operating status data of the steelmaking overhead crane and trolley, and the workshop environment data; Step S2: Based on the real-time operating status data of the steelmaking crane and trolley and the workshop environment data, combined with the production plan and task requirements, the optimal transportation strategy is obtained through reinforcement learning and ant colony algorithm. Step S3: Evaluate the current optimal transportation strategy based on the virtual meta-universe space. When the current optimal transportation strategy meets the preset requirements, control the steelmaking crane and trolley based on the current optimal combined transportation strategy.

[0007] Preferably, step S1 includes: Step S1.1: Collect data on the operating status of the overhead crane and trolley, as well as workshop environmental data; Step S1.2: Perform preprocessing, including cleaning, on the collected crane and trolley operating status data and workshop environmental data to obtain preprocessed crane and trolley operating status data and workshop environmental data; Step S1.2: Based on the preprocessed crane and trolley operation status data and workshop environment data, create a virtual metaverse space consistent with the actual scene of the steelmaking workshop using 3D modeling and real-time rendering technologies; The operating status data of the crane and trolley includes: the position, speed and load information of the crane and trolley collected by sensors; The workshop environment data includes: the transportation equipment of overhead cranes and trolleys, the workshop layout, and track information.

[0008] Preferably, step S2 includes: Step S2.1: Based on the real-time operating status data of steelmaking cranes and trolleys and the workshop environment data, combined with the production plan and task requirements, the optimal task allocation scheme and global path planning of multiple steelmaking cranes and trolleys are obtained through the ant colony algorithm. Step S2.2: Construct a global state vector containing real-time operating status data of steelmaking cranes and trolleys, as well as production plans and task requirements, using discrete time steps at the second level; train the agent using the DQN algorithm to select the optimal action in real time from the preset legal action space, combined with the reward function.

[0009] Preferably, step S2.2 includes: The global state vector includes:

[0010] in, Represents the state of a collection of intelligent agent objects. N represents the state of the task object set; N represents the number of agents. Indicates the number of task objects; The reward function includes:

[0011] in, For the number of reward sub-items; For the first Item reward value; This is a conditional indicator function; it is 1 when the triggering condition is met, and 0 otherwise.

[0012] Preferably, step S3 includes: simulating the current optimal transportation strategy based on the virtual metaverse space, collecting multi-dimensional indicators in real time during the simulation, including: the number of safety violations, the total task completion time, the average equipment waiting time, virtual energy consumption, and the task completion rate; calculating the evaluation result based on the collected multi-dimensional indicators; when the evaluation result is greater than the threshold, the current optimal joint transportation strategy meets the preset requirements; after the quantitative indicators meet the standards, the strategy execution is reviewed by experts through the immersive visualization function of the virtual metaverse space to eliminate hidden problems that cannot be covered by the indicators.

[0013] Preferably, step S3 further includes: Step S3.1: Simulate the optimal joint transportation strategy and generate control commands based on the virtual model in the virtual metaverse space; Step S3.2: Send the generated control commands to the steelmaking crane and trolley control interface to realize real-time control of the crane and trolley.

[0014] A control system for a combined transport scenario of a steelmaking overhead crane and a trolley, provided by the present invention, includes: Module M1: Constructs a virtual metaverse space based on real-time operating status data of steelmaking cranes and trolleys and workshop environmental data; Module M2: Based on the real-time operating status data of steelmaking cranes and trolleys and workshop environment data, combined with production plans and task requirements, the optimal transportation strategy is obtained through reinforcement learning and ant colony algorithm collaboration. Module M3: Evaluates the current optimal transportation strategy based on the virtual meta-universe space. When the current optimal transportation strategy meets the preset requirements, it controls the steelmaking crane and trolley based on the current optimal combined transportation strategy.

[0015] Preferably, the module M1 includes: Module M1.1: Collects operating status data of overhead cranes and trolleys, as well as workshop environmental data; Module M1.2: Performs preprocessing, including cleaning, on the collected crane and trolley operating status data and workshop environmental data to obtain preprocessed crane and trolley operating status data and workshop environmental data; Module M1.2: Based on preprocessed crane and trolley operating status data and workshop environment data, a virtual metaverse space consistent with the actual scene of the steelmaking workshop is created using 3D modeling and real-time rendering technologies; The operating status data of the crane and trolley includes: the position, speed and load information of the crane and trolley collected by sensors; The workshop environment data includes: the transportation equipment of overhead cranes and trolleys, the workshop layout, and track information.

[0016] Preferably, the module M2 includes: Module M2.1: Based on the real-time operating status data of steelmaking cranes and trolleys and workshop environment data, combined with production plans and task requirements, the optimal task allocation scheme and global path planning for multiple steelmaking cranes and trolleys are obtained through ant colony algorithm. Module M2.2: Constructs a global state vector containing real-time operating status data of steelmaking cranes and trolleys, as well as production plans and task requirements, using discrete time steps at the second level; trains the agent using the DQN algorithm, and selects the optimal action in real time from the preset legal action space in combination with the reward function; The global state vector includes:

[0017] in, Represents the state of a collection of intelligent agent objects. N represents the state of the task object set; N represents the number of agents. Indicates the number of task objects; The reward function includes:

[0018] in, For the number of reward sub-items; For the first Item reward value; This is a conditional indicator function; it is 1 when the triggering condition is met, and 0 otherwise.

[0019] Preferably, module M3 includes: simulating the current optimal transportation strategy based on a virtual metaverse space; collecting multi-dimensional indicators in real time during the simulation, including: number of safety violations, total task completion time, average equipment waiting time, virtual energy consumption, and task completion rate; comprehensively calculating the evaluation result based on the collected multi-dimensional indicators; when the evaluation result is greater than a threshold, the current optimal joint transportation strategy meets the preset requirements; after the quantitative indicators meet the standards, the strategy execution is reviewed by experts through the immersive visualization function of the virtual metaverse space to eliminate hidden problems that cannot be covered by the indicators. The module M3 also includes: Module M3.1: Simulates and generates control commands for optimal joint transportation strategies based on virtual models in the virtual metaverse space; Module M3.2: Sends the generated control commands to the steelmaking crane and trolley control interface to realize real-time control of the crane and trolley.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves proactive identification and avoidance of security risks through "full-scene simulation pre-play" and real-time data linkage in the virtual metaverse space; 2. This invention integrates intelligent algorithms such as reinforcement learning and ant colony algorithm, and relies on the simulation environment provided by the metaverse space to achieve rapid and optimal solutions to complex problems such as cooperative path planning, collision avoidance and task allocation between the vehicle and the trolley. 3. This invention achieves safe pre-simulation and iterative optimization of control strategies through the control mode of "first simulating and verifying in virtual space, and then issuing commands to physical devices", which fundamentally eliminates safety accidents such as on-site collisions and material spillage caused by control command conflicts or errors, and improves transportation safety. 4. This invention achieves global optimization of the transportation process of heavy objects such as iron ladles, materials and spare parts in steel plants, which directly results in a shorter transportation cycle, reduced waiting time for cranes and trolleys, and lower energy consumption. This significantly improves the stability of the production rhythm of downstream processes such as continuous casting machines, and ultimately drives the efficiency and capacity improvement of the entire steelmaking process. 5. This invention enables intelligent, visual, and collaborative control of steelmaking transportation equipment, effectively improving transportation efficiency, reducing production costs, and enhancing production safety, while providing a scientific basis and optimization methods for production decisions. Attached Figure Description

[0021] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the control system for a combined transport scenario of a steelmaking overhead crane and a trolley.

[0022] Figure 2 This is a schematic diagram of the control system for a combined transport scenario of a steelmaking overhead crane and a trolley. Detailed Implementation

[0023] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0024] Example 1 According to the present invention, a control system for a combined transport scenario of a steelmaking overhead crane and a trolley is provided, such as... Figures 1 to 2 As shown, it includes: Data acquisition and processing module: High-precision sensors are installed on the overhead crane and trolley to collect real-time operating status parameters and workshop environmental data; wherein, the operating status parameters include: position, speed, load and other operating data; the collected data is transmitted to the data processing center through the industrial network for cleaning and preprocessing. The processed data is used on the one hand to construct and update the virtual scene in the metaverse space, and on the other hand to provide basic data support for the intelligent control and decision-making module.

[0025] The Metaverse Space Mapping Module: Based on collected operational status parameters and workshop environment data, this module uses 3D modeling technology and real-time rendering algorithms to create a virtual metaverse space that highly replicates the actual steelmaking workshop environment. Within this virtual space, precise models of transportation equipment such as overhead cranes and trolleys, as well as elements like factory layout and tracks, are established, enabling real-time association and interaction between the virtual model and the physical objects. When the operational status of the equipment in reality changes, the model in the virtual space can be updated and reflected synchronously.

[0026] The intelligent control and decision-making module, based on collected operational status parameters and workshop environment data, and combined with production plans and task requirements, formulates optimal crane and trolley joint transportation strategies through reinforcement learning and ant colony algorithms to achieve intelligent allocation, path planning, and collaborative scheduling of transportation tasks. It evaluates the optimal crane and trolley joint transportation strategy using a virtual metaverse space to determine if it meets preset requirements. If the current optimal transportation strategy meets the preset requirements, it controls the steelmaking crane and trolley based on the current optimal joint transportation strategy. It simulates the optimal crane and trolley joint transportation strategy using a virtual model in the virtual metaverse space to generate control commands. The generated control commands are sent to the equipment control interface to achieve real-time control of the crane and trolley. Simultaneously, the optimized decision scheme is fed back to the metaverse space mapping module to update the transportation strategy display in the virtual scene. The acquired data includes virtual scene data transmitted by the metaverse space mapping module, equipment operation data provided by the data acquisition and processing module, and task scheduling information from the production management system.

[0027] The process, based on collected operational status parameters and workshop environment data, and combined with production plans and task requirements, utilizes reinforcement learning, ant colony algorithms, and other methods to formulate optimal combined transportation strategies for overhead cranes and trolleys. This enables intelligent allocation, path planning, and collaborative scheduling of transportation tasks, including: Based on the real-time operating status data of steelmaking cranes and trolleys and workshop environment data, combined with production plans and task requirements, the optimal task allocation scheme and global path planning for multiple steelmaking cranes and trolleys are obtained through ant colony algorithm. Specifically, including: Data preprocessing: Cleaning and normalizing real-time operating status data of cranes and trolleys and workshop environmental data, and converting production plans and task requirements into ant colony algorithm input format; Ant colony algorithm initialization: Set parameters such as the number of ants and pheromone evaporation rate, and construct a workshop layout diagram including tracks, work areas, and obstacles.

[0028] Task allocation and path planning: Ants select tasks and paths based on task allocation and pheromone concentration, iteratively update pheromones, and output the optimal task allocation scheme and global path planning results.

[0029] Using discrete time steps at the second level, a global state vector is constructed that includes real-time operating status data of steelmaking cranes and trolleys, as well as production plans and task requirements. The agent is trained using the DQN algorithm and selects the optimal action in real time from a preset legal action space in combination with a reward function.

[0030] Specifically, the global state vector is constructed by collecting real-time operating status data of the crane and trolley every second and integrating production plan and task requirement information to form a global state vector.

[0031] The global state consists of a tuple comprising the states of the vehicle and trolley intelligent agent sets and the task object sets, containing complete spatiotemporal information of all dynamic entities within the range. Assume the environment has... Taiwan-type train, first The status of the trolley's intelligent agent includes its current location, current loading status, and current transportation task target location information; The task object, the first If the state of a task object includes starting position and destination position information, then the global state is represented as follows:

[0032] in, Represents the state of a collection of intelligent agent objects. This indicates the state of the task object collection.

[0033] DQN algorithm initialization: Initialize the DQN network structure and set training parameters such as learning rate, discount factor, and experience replay buffer size.

[0034] Agent training and action selection: The agent selects actions based on the global state vector, obtains reward signals and new state vectors after execution, stores experience and updates the weights of the DQN network, and finally achieves real-time optimal action selection.

[0035] Each trolley intelligent agent is an independent decision-making unit. The atomic actions that can be executed within a discrete time step constitute the individual action space, which is defined as follows: Action space of a single intelligent agent: The action of a single trolley intelligent agent consists of a set of 5 mutually exclusive basic operations (movement and hoisting operations cannot be performed simultaneously at the same time step):

[0036] Action definitions: hold (keep still), left (move to the left), right (move to the right), hoist (perform the lifting action), release (perform the lowering action).

[0037] Joint action space: The Cartesian product of the action spaces of all agents, where each joint action is represented as a vector. :

[0038] A time-based design framework is adopted, and dynamic continuous simulation and decision-making are coupled by discretizing the time step (1 second). The environment advances the simulation process with second-level precision: the agent executes a decision action every 1 second, and the system updates the global state across the span (trolley position, trolley state, task queue, etc.) according to physical rules and process constraints, while completing the reward calculation.

[0039] State transition function:

[0040] in, State at time t; The state is at time t+1; State transition time interval (set to 1 second); The reward function consists of three dimensions: safety constraints, behavioral norms, and task incentives. It guides the agent to complete tasks safely and efficiently through positive incentives and negative penalties. The total reward is the sum of the individual components. Total reward function:

[0041] in, The number of reward sub-items. For the first Item reward value. : Conditional indicator function (1 if the trigger condition is met, 0 otherwise).

[0042] Early completion reward:

[0043] in, : The maximum number of steps allowed per training round. : The actual number of steps in the current training round.

[0044] Reward function decomposition table

[0045] Virtual Reality Interaction Module: Utilizing virtual reality devices, such as head-mounted displays and motion-sensing interactive devices, this module provides operators with an immersive virtual operation experience. Operators can intuitively observe the operating status of the crane and trolley in the virtual metaverse space, monitor the transportation process, and send commands to the intelligent control and decision-making module through interactive operations, enabling real-time intervention and optimization of the transportation equipment. The data acquired by this module includes input signals from the virtual reality devices, such as operator actions and voice commands, as well as virtual scene rendering data provided by the metaverse space mapping module. This embodiment provides an immersive first-person perspective through VR / AR devices, enabling operators to remotely "enter" the workshop. They can perform operations such as grasping, moving, and zooming in on virtual equipment using natural gestures and voice, and issue control commands. This greatly reduces the psychological burden and skill threshold of remote operation, allowing experts to remotely "immerse themselves" in equipment diagnosis, maintenance guidance, and technical training, breaking the limitations of geographical space.

[0046] Historical Data Playback and Dynamic Reconstruction Module: This module aims to enable the playback and dynamic reconstruction of historical data from combined train and trolley transportation. By retrospectively analyzing and reproducing historical data, it provides operators and technicians with an intuitive reconstruction of historical transportation scenarios, facilitating the analysis of past problems, summarizing lessons learned, and optimizing transportation strategies and operational processes. Regarding dynamic reconstruction, the reconstructed scenario can be dynamically adjusted based on changes in historical data, ensuring the accuracy and authenticity of the presented historical scene. This embodiment, through the historical data playback and dynamic reconstruction module, achieves full, high-fidelity reproduction of the transportation process within any historical time period. Operators can "rewind" at any time to review and accurately pinpoint past failure points and efficiency bottlenecks.

[0047] The Future Planning Simulation and Prediction Module is used to simulate and predict future plans for the combined transportation of overhead cranes and trolleys. By modeling and analyzing production plans, task scheduling information, and related constraints, and combining metaverse space mapping technology, it predicts various aspects of the future transportation process in a virtual environment, including equipment operating status, potential congestion points, and task completion times. This provides strong support for production decisions, allows for the development of reasonable countermeasures in advance, optimizes transportation plans, and improves production efficiency and resource utilization. In this embodiment, the Future Planning Simulation and Prediction Module enables advanced simulation of the transportation process under a new production plan. The system can accurately predict equipment load, potential conflict points, and task completion times within the next half hour to several hours, supporting administrators in making forward-looking decisions, such as activating backup trolleys in advance and adjusting furnace schedules, transforming passive response into proactive intervention, and significantly improving the anti-interference capability and fulfillment rate of production plans.

[0048] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0049] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A control method for a combined transport scenario of steelmaking overhead crane and trolley, characterized in that, include: Step S1: Construct a virtual metaverse space based on the real-time operating status data of the steelmaking overhead crane and trolley, and the workshop environment data; Step S2: Based on the real-time operating status data of the steelmaking crane and trolley and the workshop environment data, combined with the production plan and task requirements, the optimal transportation strategy is obtained through reinforcement learning and ant colony algorithm. Step S3: Evaluate the current optimal transportation strategy based on the virtual meta-universe space. When the current optimal transportation strategy meets the preset requirements, control the steelmaking crane and trolley based on the current optimal combined transportation strategy.

2. The control method for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 1, characterized in that, Step S1 includes: Step S1.1: Collect data on the operating status of the overhead crane and trolley, as well as workshop environmental data; Step S1.2: Perform preprocessing, including cleaning, on the collected crane and trolley operating status data and workshop environmental data to obtain preprocessed crane and trolley operating status data and workshop environmental data; Step S1.2: Based on the preprocessed crane and trolley operation status data and workshop environment data, create a virtual metaverse space consistent with the actual scene of the steelmaking workshop using 3D modeling and real-time rendering technologies; The operating status data of the crane and trolley includes: the position, speed and load information of the crane and trolley collected by sensors; The workshop environment data includes: the transportation equipment of overhead cranes and trolleys, the workshop layout, and track information.

3. The control method for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 1, characterized in that, Step S2 includes: Step S2.1: Based on the real-time operating status data of steelmaking cranes and trolleys and the workshop environment data, combined with the production plan and task requirements, the optimal task allocation scheme and global path planning of multiple steelmaking cranes and trolleys are obtained through the ant colony algorithm. Step S2.2: Construct a global state vector containing real-time operating status data of steelmaking cranes and trolleys, as well as production plans and task requirements, using discrete time steps at the second level; train the agent using the DQN algorithm to select the optimal action in real time from the preset legal action space, combined with the reward function.

4. The control method for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 3, characterized in that, Step S2.2 includes: The global state vector includes: in, Represents the state of a collection of intelligent agent objects. N represents the state of the task object set; N represents the number of agents. Indicates the number of task objects; Represents the state of a single agent; Individual task status; The reward function includes: in, For the number of reward sub-items; For the first Item reward value; This is a conditional indicator function; it is 1 when the triggering condition is met, and 0 otherwise.

5. The control method for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 1, characterized in that, Step S3 includes: simulating the current optimal transportation strategy based on the virtual metaverse space, collecting multi-dimensional indicators in real time during the simulation, including: number of safety violations, total task completion time, average equipment waiting time, virtual energy consumption, and task completion rate; calculating and evaluating the results based on the collected multi-dimensional indicators; when the evaluation results are greater than the threshold, the current optimal joint transportation strategy meets the preset requirements; after the quantitative indicators meet the standards, the strategy execution is reviewed by experts through the immersive visualization function of the virtual metaverse space to eliminate hidden problems that cannot be covered by the indicators.

6. The control method for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 1, characterized in that, Step S3 further includes: Step S3.1: Simulate the optimal joint transportation strategy and generate control commands based on the virtual model in the virtual metaverse space; Step S3.2: Send the generated control commands to the steelmaking crane and trolley control interface to realize real-time control of the crane and trolley.

7. A control system for a combined transport scenario of a steelmaking overhead crane and a trolley, characterized in that, include: Module M1: Constructs a virtual metaverse space based on real-time operating status data of steelmaking cranes and trolleys and workshop environmental data; Module M2: Based on the real-time operating status data of steelmaking cranes and trolleys and workshop environment data, combined with production plans and task requirements, the optimal transportation strategy is obtained through reinforcement learning and ant colony algorithm collaboration. Module M3: Evaluates the current optimal transportation strategy based on the virtual meta-universe space. When the current optimal transportation strategy meets the preset requirements, it controls the steelmaking crane and trolley based on the current optimal combined transportation strategy.

8. The control system for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 7, characterized in that, The module M1 includes: Module M1.1: Collects operating status data of overhead cranes and trolleys, as well as workshop environmental data; Module M1.2: Performs preprocessing, including cleaning, on the collected crane and trolley operating status data and workshop environmental data to obtain preprocessed crane and trolley operating status data and workshop environmental data; Module M1.2: Based on preprocessed crane and trolley operating status data and workshop environment data, a virtual metaverse space consistent with the actual scene of the steelmaking workshop is created using 3D modeling and real-time rendering technologies; The operating status data of the crane and trolley includes: the position, speed and load information of the crane and trolley collected by sensors; The workshop environment data includes: the transportation equipment of overhead cranes and trolleys, the workshop layout, and track information.

9. The control system for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 7, characterized in that, The module M2 includes: Module M2.1: Based on the real-time operating status data of steelmaking cranes and trolleys and workshop environment data, combined with production plans and task requirements, the optimal task allocation scheme and global path planning for multiple steelmaking cranes and trolleys are obtained through ant colony algorithm. Module M2.2: Constructs a global state vector containing real-time operating status data of steelmaking cranes and trolleys, as well as production plans and task requirements, using discrete time steps at the second level; trains the agent using the DQN algorithm, and selects the optimal action in real time from the preset legal action space in combination with the reward function; The global state vector includes: in, Represents the state of a collection of intelligent agent objects. N represents the state of the task object set; N represents the number of agents. Indicates the number of task objects; Represents the state of a single agent; Individual task status; The reward function includes: in, For the number of reward sub-items; For the first Item reward value; This is a conditional indicator function; it is 1 when the triggering condition is met, and 0 otherwise.

10. The control system for the combined transportation scenario of steelmaking overhead crane and trolley as described in claim 7, characterized in that, The module M3 includes: simulating the current optimal transportation strategy based on a virtual metaverse space; collecting multi-dimensional indicators in real time during the simulation, including: number of safety violations, total task completion time, average equipment waiting time, virtual energy consumption, and task completion rate; and comprehensively calculating the evaluation result based on the collected multi-dimensional indicators; when the evaluation result is greater than a threshold, the current optimal joint transportation strategy meets the preset requirements; after the quantitative indicators meet the standards, the strategy execution is reviewed by experts through the immersive visualization function of the virtual metaverse space to eliminate hidden problems that cannot be covered by the indicators. The module M3 also includes: Module M3.1: Simulates and generates control commands for optimal joint transportation strategies based on virtual models in the virtual metaverse space; Module M3.2: Sends the generated control commands to the steelmaking crane and trolley control interface to realize real-time control of the crane and trolley.

Images