Maintenance task scheduling optimization system and method based on three-dimensional visualization

Abstract

The invention discloses a maintenance task scheduling optimization system and method based on three-dimensional visualization. The system comprises a three-dimensional environment construction module used for constructing a digital twin environment accurately mapped with a physical entity; the task spatialization module is used for analyzing the maintenance task into a four-dimensional space-time occupant; the three-dimensional scheduling module is used for carrying out collision detection on the four-dimensional space-time occupant and carrying out iterative adjustment according to a preset strategy to eliminate conflicts; and the visual simulation module is used for carrying out dynamic rehearsal on the conflict-free optimal scheduling scheme. According to the method, unified modeling and intelligent scheduling are carried out on the maintenance tasks in time and space dimensions, potential collision and interference risks between parallel tasks can be automatically identified and solved, the safety, the cooperation efficiency and the plan predictability of maintenance work are remarkably improved, and the maintenance efficiency is improved. And powerful visual decision support is provided for maintenance management of a complex industrial scene.

Description

Technical Field

[0001] This invention relates to the field of computer-aided maintenance, and in particular to a maintenance task scheduling and optimization system and method based on three-dimensional visualization. Background Technology

[0002] Existing scheduling for large and complex equipment maintenance tasks primarily relies on project management software based on classic operations research theories such as the critical path method. While these technologies are mature in handling the temporal sequence and logical dependencies of tasks, their core flaw lies in the severe decoupling of the time dimension of scheduling from the physical space dimension of execution. In traditional scheduling models, the work site is abstracted as an infinite or conflict-free resource, completely ignoring the actual occupancy and dynamic changes of personnel, equipment, and materials in the three-dimensional physical space during maintenance task execution. This "spatial blind" planning approach often results in scheduling schemes that are theoretically logically consistent but physically unenforceable. For example, problems such as interference between multiple work teams in confined spaces, collisions between the turning paths of large hoisting equipment and fixed facilities, and material transport channels being occupied by temporary operations cannot be foreseen during the planning stage. Although some advanced solutions have introduced 3D visualization technology, its application is mostly limited to post-event animation demonstrations or results presentations of the formulated scheduling schemes. The 3D model does not participate as a core element in the closed-loop calculation of scheduling conflict detection and optimization. Therefore, existing technologies generally lack in-depth insight into and proactive avoidance of spatiotemporal coupling conflicts, resulting in frequent on-site adjustments to maintenance plans. This not only seriously affects maintenance efficiency and prolongs downtime, but also creates serious safety hazards. Summary of the Invention

[0003] To address the aforementioned problems in the existing technology, the present invention aims to provide a maintenance task scheduling optimization system based on three-dimensional visualization, comprising: The 3D environment construction module is configured to integrate the static geometric information of the object under maintenance, the equipment BOM information, and dynamic operating data to construct a 3D digital environment that is precisely mapped to the physical entity.

[0004] The task spatialization module, connected to the three-dimensional environment construction module, is configured to parse maintenance task instructions into spatial execution elements and generate a four-dimensional spatiotemporal occupancy volume for each maintenance task in the three-dimensional digital environment, representing the physical space and time period occupied during its execution.

[0005] The three-dimensional scheduling module, connected to the task spatialization module, is configured to receive the initial task sequence, perform collision and interference detection on all four-dimensional spatiotemporal occupies in the three-dimensional digital environment, and when a conflict is detected, iteratively adjust the temporal arrangement or spatial path of the conflicting tasks according to the preset optimization goal to eliminate the conflict and output a conflict-free scheduling scheme.

[0006] The visualization simulation module is configured to receive the conflict-free scheduling scheme, perform dynamic animation pre-show of the maintenance process in the three-dimensional digital environment, and respond to user interactive queries and scheme fine-tuning commands.

[0007] Furthermore, the 3D environment construction module is also configured to connect to an IoT data platform, which aggregates on-site sensor data, including at least: Equipment status data: originates from sensors integrated with the maintenance equipment. The three-dimensional scheduling module dynamically adjusts the priority of associated maintenance tasks based on the equipment status data.

[0008] Spatial positioning data: originates from positioning tags deployed on personnel or mobile equipment. The three-dimensional environment construction module uses the spatial positioning data to calibrate the position of dynamic or flowing occupies in the four-dimensional spatiotemporal occupancy in real time.

[0009] Furthermore, when generating the four-dimensional spatiotemporal occupancy volume, the task spatialization module divides the four-dimensional spatiotemporal occupancy volume into static occupancy volume, dynamic occupancy volume, and circulating occupancy volume. The static occupancy volume corresponds to the fixed space occupied by the maintenance equipment body; the dynamic occupancy volume corresponds to the dynamically changing space formed by the operating range of maintenance personnel, the movement trajectory of tools, and the reach of large equipment; and the circulating occupancy volume corresponds to the channel space occupied by the preset path of spare parts from the warehouse to the maintenance point.

[0010] Furthermore, the three-dimensional scheduling module follows the following priority strategy during iterative adjustments: First priority: Without violating the task logic constraints, time-shift the low-priority tasks that conflict.

[0011] Second priority: If time translation fails to eliminate the conflict, then a collision-free path is replanned for the flow of the conflicting task in the three-dimensional digital environment.

[0012] Third priority: If the path replanning adjustment is ineffective, mark the conflict as a strong conflict point and output alternative resource solutions.

[0013] A maintenance task scheduling optimization method based on 3D visualization includes the following steps: Step S1: Construct a high-fidelity 3D digital environment.

[0014] Step S2: Receive the list of maintenance tasks to be scheduled, and convert each maintenance task instruction into a set of spatiotemporal parameters associated with the three-dimensional digital environment. The conversion includes generating a four-dimensional spatiotemporal occupancy volume for each maintenance task, which includes the three-dimensional spatial range and time period required for its execution.

[0015] Step S3: Generate an initial scheduling sequence based on the preset priority and pre- and post-task logical dependencies of the tasks.

[0016] Step S4: Load the initial scheduling sequence into the three-dimensional digital environment, and scan all four-dimensional spatiotemporal occupants in chronological order to detect whether there are two or more four-dimensional spatiotemporal occupants that geometrically overlap or intrude into the safe distance on any time slice.

[0017] Step S5: If a conflict is detected in step S4, the maintenance task that caused the conflict is scheduled and adjusted according to the preset conflict resolution rules, and steps S4 and S5 are repeated until no conflict is detected in step S4, thus obtaining an optimized scheduling scheme.

[0018] Step S6: Output the optimized scheduling scheme and perform dynamic simulation in the three-dimensional digital environment to present the execution process of the scheduling scheme in a visual manner.

[0019] Furthermore, step S1 also includes establishing a data mapping channel that is synchronized with the physical site in real time. The data mapping channel performs the following operations: Through IoT gateways, real-time access and analysis of equipment operating data and personnel or equipment location data uploaded by field sensors are achieved.

[0020] The parsed data is bound to model instances in the 3D digital environment, and the attributes, colors, or positions of the model instances are automatically changed based on real-time data changes.

[0021] Furthermore, in step S2, the operation of generating the four-dimensional spatiotemporal occupancy volume includes: Step S201: Identify the equipment associated with the maintenance task in the three-dimensional digital environment and generate a static spatial envelope based on its geometric shape.

[0022] Step S202: Based on the personnel operation safety radius and tool usage space defined in the maintenance procedure, a dynamic operating space is generated by expanding the periphery of the static space envelope.

[0023] Step S203: Combine the estimated start and end times of the maintenance task execution, stretch the static and dynamic spaces in the time dimension to form the four-dimensional spatiotemporal occupancy volume.

[0024] Furthermore, the spatiotemporal conflict scanning in step S4 includes: Hard collision detection: Detects whether equipment entities, tool entities, or personnel entities with different maintenance tasks physically overlap in space at the same time.

[0025] Soft conflict detection: Detects whether the dynamic operating space or flow path of one maintenance task intrudes into the safety warning area or restricted area set by another maintenance task.

[0026] Furthermore, the conflict resolution rule in step S5 is a hierarchical adjustment strategy based on conflict type, the strategy including: For soft conflicts, prioritize adjusting the execution time of low-priority maintenance tasks.

[0027] For hard conflicts, first try to adjust the execution time. If this is ineffective, then replan the equipment or material transportation path of the conflict maintenance task in the three-dimensional digital environment. If the path cannot be replanned, then decompose the maintenance task into multiple sub-tasks and regenerate a smaller four-dimensional spatiotemporal occupancy volume for each sub-task.

[0028] Furthermore, the visualization in step S6 includes executing at least two of the following interactive verification modes: Global Preview Mode: Based on the timeline of the optimized scheduling plan, the mode dynamically displays the parallel activity trajectories and space occupancy changes of all maintenance tasks, personnel, equipment, and materials in a sped-up playback format, and highlights maintenance tasks on the critical path.

[0029] Focus Perspective Mode: In response to the user's pause and selection commands during the simulation, after selecting any maintenance task, personnel or equipment, the system enters the perspective view, semi-transparent other unrelated models, and highlights the complete four-dimensional spatiotemporal occupancy of the selected object and its activity space range throughout the entire task cycle, while listing the tasks that are spatiotemporally related to it.

[0030] Compared to existing technologies, the advantages of this invention are as follows: By transforming abstract maintenance tasks into concrete four-dimensional spatiotemporal occupancy entities, this invention enables the scheduling optimization process to move beyond the limitations of time and logic, allowing for precise collision and interference detection of the physical space occupancy of task execution within a three-dimensional digital environment. This spatiotemporal integrated computing approach fundamentally solves the problem of planning and execution disconnect caused by "spatial blindness" in existing technologies. It proactively eliminates various physical execution conflicts during the planning stage, significantly improving the physical executability and on-site safety of scheduling schemes.

[0031] This invention transforms 3D visualization technology from a "post-event demonstration" tool in existing technologies into a core component of scheduling optimization computation. The 3D digital environment is not only a simulation platform but also a computational space for conflict detection and path planning. Through a "conflict-driven iterative optimization" method, an automated closed loop of "spatial conflict scanning - scheduling scheme adjustment" is constructed. This makes the scheduling scheme generation process intelligent and automated, significantly reducing reliance on human experience and ensuring a high degree of coordination and optimization of the final scheme across the logical, temporal, and physical dimensions. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the system module structure of the present invention.

[0033] Figure 2 This is an exemplary flowchart of the task scheduling method of the present invention.

[0034] Figure 3 This is an exemplary flowchart of the steps for generating a spatiotemporal occupancy volume according to the present invention. Detailed Implementation

[0035] This application discloses a maintenance task scheduling optimization system based on 3D visualization. It aims to solve the spatiotemporal conflict problems caused by limited space, parallel tasks, and overlapping resources during the maintenance of large industrial equipment by constructing a digital twin environment that precisely maps to physical entities and introducing the concept of a four-dimensional spatiotemporal occupancy volume. This system can achieve automated conflict detection, intelligent scheduling optimization, and visual simulation pre-playing of maintenance tasks, significantly improving the safety, efficiency, and predictability of complex maintenance projects. The invention is further described below with reference to specific embodiments.

[0036] like Figure 1 As shown in this embodiment, a maintenance task scheduling and optimization system based on three-dimensional visualization is provided, including: The 3D environment construction module is configured to integrate the static geometric information, equipment BOM information, and dynamic operational data of the object under maintenance to build a 3D digital environment that is precisely mapped to the physical entity. In one embodiment, the 3D environment construction module serves as the digital foundation of the entire system. It constructs a high-fidelity 3D digital environment by integrating multiple data sources from the object under maintenance. For example, for the maintenance of a gas turbine, the module imports the CAD or BIM model of the turbine and its auxiliary equipment to obtain its precise static geometric information. Simultaneously, it integrates the equipment's BOM information to understand the hierarchy and assembly relationships of various components. Furthermore, it is configured to connect to an IoT data platform that aggregates field sensor data. This sensor data includes at least: equipment status data from sensors integrated with the maintenance equipment, such as temperature, pressure, and vibration sensors; and spatial positioning data from positioning tags deployed on personnel or mobile equipment, such as UWB positioning tags. Through the fusion of these multi-source data, the system constructs a richly informative digital twin environment that is synchronized in real-time with the physical site.

[0037] The task spatialization module, connected to the 3D environment construction module, is configured to parse maintenance task instructions into spatial execution elements and generate a four-dimensional spatiotemporal occupancy volume for each maintenance task in the 3D digital environment, representing the physical space and time period occupied during its execution. In one embodiment, the task spatialization module transforms abstract maintenance tasks into geometric entities that can be collision-detected in 3D space. For example, for a maintenance task of "replacing blade No. 3," this module will parse it into multiple spatial execution elements. It will distinguish the four-dimensional spatiotemporal occupancy volume into static occupancy volumes, dynamic occupancy volumes, and circulating occupancy volumes. Static occupancy volumes correspond to the fixed space occupied by the maintenance equipment itself, such as the geometric model of blade No. 3 itself. Dynamic occupancy volumes correspond to the dynamically changing space formed by the operating range of maintenance personnel, the movement trajectory of tools, and the reach of large equipment, such as an operating safety zone with a radius of 1.5 meters around blade No. 3, and the trajectory envelope of the disassembly tools. Circulating occupancy volumes correspond to the channel space occupied by the preset path of spare parts from the warehouse to the maintenance point, such as the preset channel for transporting new blades from the warehouse to the machine position. Ultimately, these spatial elements, combined with the mission's time cycle, form a unique "occupying body" in four-dimensional spacetime.

[0038] The 3D scheduling module, connected to the task spatialization module, is configured to receive an initial task sequence, perform collision and interference detection on all four-dimensional spatiotemporal occupants in the 3D digital environment, and, upon detecting a conflict, iteratively adjust the temporal arrangement or spatial path of the conflicting tasks according to a preset optimization objective to eliminate the conflict, outputting a conflict-free scheduling scheme. In one embodiment, the 3D scheduling module is the system's intelligent "commander." It receives an initial sequence containing all maintenance tasks and performs collision and interference detection on the four-dimensional spatiotemporal occupants corresponding to all tasks in the 3D digital environment. If the occupants of two tasks overlap in spatiotemporal space, for example, if two different teams plan to use the same narrow passage at the same time, the system will detect a conflict. At this time, according to a preset optimization objective, such as "shortest total project duration" or "shortest downtime of critical equipment," the system will iteratively adjust according to a preset priority strategy. The first priority is to perform time shifting on the lower-priority tasks that conflict, without violating task logic constraints. The second priority is that if time shifting fails to eliminate the conflict, a collision-free path will be replanned for the flow occupants of the conflicting tasks in the 3D digital environment. The third priority is to mark conflicts as strong conflicts if path replanning adjustments are ineffective, and output alternative resource solutions. Through repeated iterative adjustments, until all conflicts are eliminated, a conflict-free and optimized scheduling scheme is finally output.

[0039] The visualization simulation module is configured to receive conflict-free scheduling plans and dynamically animate the maintenance process in a 3D digital environment, while also responding to user interactive queries and plan fine-tuning commands. In one embodiment, the visualization simulation module serves as a "sandbox" platform for scheduling plans. It dynamically recreates the entire maintenance process in 3D animation, allowing managers to visually see the location and status of all personnel, equipment, and materials at any given time, thereby verifying the feasibility of the scheduling plan. Users can also interactively query, for example, by clicking on a specific maintenance worker to view their complete task list and activity trajectory for the day. If improvements are found, users can make fine-tuning adjustments, such as manually adjusting the start time of a non-critical task; the system will immediately re-perform conflict detection and provide feedback on the results.

[0040] The 3D environment building module is also configured to connect to an IoT data platform, which aggregates on-site sensor data. This sensor data includes at least: Equipment status data: originates from sensors integrated with the maintenance equipment. The 3D scheduling module dynamically adjusts the priority of associated maintenance tasks based on the equipment status data. Spatial positioning data: originates from positioning tags deployed on personnel or mobile equipment. The 3D environment construction module uses spatial positioning data to calibrate the position of dynamic or flowing occupies in the four-dimensional spatiotemporal occupancy volume in real time.

[0041] When generating a four-dimensional spatiotemporal occupancy volume, the task spatialization module divides the four-dimensional spatiotemporal occupancy volume into static occupancy volume, dynamic occupancy volume, and circulating occupancy volume. The static occupancy volume corresponds to the fixed space occupied by the maintenance equipment itself; the dynamic occupancy volume corresponds to the dynamically changing space formed by the operating range of maintenance personnel, the movement trajectory of tools, and the reach of large equipment; and the circulating occupancy volume corresponds to the channel space occupied by the preset path of spare parts from the warehouse to the maintenance point.

[0042] The three-dimensional scheduling module follows the following priority strategy during iterative adjustments: First priority: Without violating task logic constraints, perform time shifting on low-priority tasks that conflict; Second priority: If time shifting fails to eliminate the conflict, replan a collision-free path for the flow and occupancy of the conflicting tasks in the 3D digital environment; Third priority: If path replanning is ineffective, mark the conflict as a strong conflict point and output alternative resource solutions.

[0043] like Figure 2 The figure shown is a maintenance task scheduling optimization method based on three-dimensional visualization in this embodiment, which includes the following steps: Step S1 involves constructing a high-fidelity 3D digital environment. In one embodiment, Step S1 is completed through a 3D environment construction module, integrating the static geometric information of the object under maintenance, equipment BOM information, and dynamic operating data. Furthermore, this step includes establishing a data mapping channel that is synchronized with the physical site in real time. This channel, via an IoT gateway, accesses and parses equipment operating data and personnel or equipment positioning data uploaded by field sensors in real time. For example, when the temperature sensor reading of a piece of equipment on site exceeds the limit, the corresponding model instance in the 3D digital environment will automatically turn red and highlight. The parsed data is bound to the model instance in the 3D digital environment, and based on changes in real-time data, the attributes, color, or position of the model instance are automatically changed, thereby achieving real-time synchronization between the digital twin environment and the physical site.

[0044] Step S2: Receive the list of maintenance tasks to be scheduled and convert each maintenance task instruction into a set of spatiotemporal parameters associated with the 3D digital environment. This conversion includes generating a four-dimensional spatiotemporal occupancy volume for each maintenance task, containing the required 3D spatial range and time period for its execution. In one embodiment, step S2 is executed by the task spatialization module. The operation of generating the four-dimensional spatiotemporal occupancy volume includes: Step S201: Determine the equipment associated with the maintenance task in the 3D digital environment and generate a static spatial envelope based on its geometry. For example, generate a static occupancy volume that precisely matches the shape of the pump body that needs to be replaced. Step S202: Based on the personnel operation safety radius and tool usage space defined in the maintenance procedures, expand the static spatial envelope to generate a dynamic operation space. For example, expand a cylindrical area with a radius of 1 meter around the pump body as the dynamic operation occupancy volume for personnel. Step S203: Combine the estimated start and end times of the maintenance task execution, stretch the static and dynamic spaces in the time dimension to form a four-dimensional spatiotemporal occupancy volume. For example, if the task of replacing the pump body is scheduled to take place between 9:00 AM and 11:00 AM, then the aforementioned static and dynamic occupants will exist within this two-hour time period.

[0045] Step S3: Based on the preset priority and pre- and post-task logical dependencies of the tasks, an initial scheduling sequence is generated. In one embodiment, for example, the system will automatically prioritize the "equipment power outage" task before all electrical maintenance tasks and the "scaffolding erection" task before tasks requiring high-altitude operations, based on the task dependencies in the work order, thereby generating a preliminary, logically feasible initial scheduling sequence.

[0046] Step S4: The initial scheduling sequence is loaded into the 3D digital environment, and all four-dimensional spatiotemporal occupants are scanned sequentially along the timeline to detect whether two or more four-dimensional spatiotemporal occupants geometrically overlap or intrude into the safety distance on any time slice. In one embodiment, the 3D scheduling module will advance forward in a very small time step, starting from the planned start time, much like playing a timeline. On each time slice represented by the time step, the system will perform geometric collision detection on all spatiotemporal occupants existing at that moment. Spatiotemporal conflict scanning includes: hard conflict detection, which detects whether equipment entities, tool entities, or personnel entities of different maintenance tasks physically overlap in space at the same time, for example, the booms of two cranes occupy the same airspace at the same time. Soft conflict detection, which detects whether the dynamic operating space or flow path of one maintenance task intrudes into the safety warning area or restricted area set by another maintenance task, for example, the spark splash safety area of ​​a welding task overlaps with the operating area of ​​another ongoing oil circuit maintenance task below.

[0047] Step S5: If a conflict is detected in step S4, the maintenance task that caused the conflict is scheduled and adjusted according to the preset conflict resolution rules, and steps S4 and S5 are repeated until no conflict is detected in step S4, thus obtaining an optimized scheduling scheme.

[0048] Step S6 outputs an optimized scheduling scheme and performs dynamic simulation in a 3D digital environment, visually presenting the execution process of the scheduling scheme. In one embodiment, step S6 is the verification and delivery of the final scheme. The optimized scheduling scheme output by the system is not only a timetable, but also a "digital rehearsal" that can be dynamically simulated in a 3D environment. The visualization presentation includes executing at least two of the following interactive verification modes: global pre-show mode, which dynamically displays the parallel activity trajectories and space occupancy changes of all maintenance tasks, personnel, equipment, and materials in a sped-up playback manner according to the timeline of the optimized scheduling scheme, and highlights maintenance tasks on the critical path, allowing managers to have an intuitive understanding of the overall picture and key nodes of the entire maintenance process. The focus perspective mode responds to the user's pause and selection commands during the simulation. After selecting any maintenance task, personnel, or equipment, the system enters the perspective view, semi-transparently displays other unrelated models, and highlights the complete four-dimensional spatiotemporal occupancy of the selected object and its activity space range throughout the entire task cycle. At the same time, it lists other tasks that are spatiotemporally related to it, helping users to deeply analyze the spatiotemporal resource occupancy and task dependencies of a specific object.

[0049] Step S1 also includes establishing a data mapping channel that is synchronized with the physical site in real time. The data mapping channel performs the following operations: Through IoT gateways, real-time access and analysis of equipment operating data and personnel or equipment positioning data uploaded by field sensors are achieved. The analyzed data is then bound to model instances in the 3D digital environment, and the attributes, colors, or positions of the model instances are automatically changed based on real-time data changes.

[0050] In step S2, the operation of generating the four-dimensional spacetime occupancy volume includes: Step S201: Identify the equipment associated with the maintenance task in the three-dimensional digital environment and generate a static spatial envelope based on its geometric shape.

[0051] Step S202: Based on the personnel operation safety radius and tool usage space defined in the maintenance procedure, a dynamic operating space is generated by expanding the outer perimeter of the static space envelope.

[0052] Step S203: Combine the estimated start and end times of the maintenance task execution, stretch the static and dynamic spaces in the time dimension to form a four-dimensional spatiotemporal occupancy volume.

[0053] The spatiotemporal conflict scan in step S4 includes: Hard collision detection: Detects whether equipment entities, tool entities, or personnel entities with different maintenance tasks physically overlap in space at the same time.

[0054] Soft conflict detection: Detects whether the dynamic operating space or flow path of one maintenance task intrudes into the safety warning area or restricted area set by another maintenance task.

[0055] In this embodiment, the conflict resolution rule in step S5 is a hierarchical adjustment strategy based on conflict type, and the strategy includes: For soft conflicts, prioritize adjusting the execution time of low-priority maintenance tasks; For hard conflicts, first try adjusting the execution time. If this is ineffective, replan the equipment or material transportation path of the conflict maintenance task in the three-dimensional digital environment. If the path cannot be replanned, decompose the maintenance task into multiple sub-tasks and regenerate a smaller four-dimensional spatiotemporal occupancy volume for each sub-task.

[0056] The visualization presentation in step S6 of this embodiment includes executing at least two of the following interactive verification modes: Global Preview Mode: Based on the timeline of the optimized scheduling plan, the mode dynamically displays the parallel activity trajectories and space occupancy changes of all maintenance tasks, personnel, equipment, and materials in a sped-up playback format, and highlights maintenance tasks on the critical path.

[0057] Focus Perspective Mode: In response to the user's pause and selection commands during the simulation, after selecting any maintenance task, personnel or equipment, the system enters the perspective view, semi-transparent other unrelated models, and highlights the complete four-dimensional spatiotemporal occupancy of the selected object and its activity space range throughout the entire task cycle, while listing the tasks that are spatiotemporally related to it.

[0058] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A maintenance task scheduling and optimization system based on three-dimensional visualization, characterized in that, include: The 3D environment construction module is configured to integrate the static geometric information of the maintenance object, the equipment BOM information, and the dynamic operation data to construct a 3D digital environment that is accurately mapped to the physical entity. The task spatialization module, connected to the three-dimensional environment construction module, is configured to parse maintenance task instructions into spatial execution elements and generate a four-dimensional spatiotemporal occupancy volume in the three-dimensional digital environment for each maintenance task, representing the physical space and time period occupied during its execution. The three-dimensional scheduling module, connected to the task spatialization module, is configured to receive the initial task sequence, perform collision and interference detection on all four-dimensional spatiotemporal occupies in the three-dimensional digital environment, and when a conflict is detected, iteratively adjust the temporal arrangement or spatial path of the conflicting tasks according to the preset optimization goal to eliminate the conflict and output a conflict-free scheduling scheme. The visualization simulation module is configured to receive the conflict-free scheduling scheme, perform dynamic animation pre-show of the maintenance process in the three-dimensional digital environment, and respond to user interactive queries and scheme fine-tuning commands.

2. The maintenance task scheduling and optimization system based on three-dimensional visualization according to claim 1, characterized in that: The 3D environment construction module is also configured to connect to an Internet of Things (IoT) data platform, which aggregates on-site sensor data, including at least: Equipment status data: originates from sensors integrated with the maintenance equipment. The three-dimensional scheduling module dynamically adjusts the priority of associated maintenance tasks based on the equipment status data. Spatial positioning data: originates from positioning tags deployed on personnel or mobile equipment. The three-dimensional environment construction module uses the spatial positioning data to calibrate the position of dynamic or flowing occupies in the four-dimensional spatiotemporal occupancy in real time.

3. The maintenance task scheduling and optimization system based on three-dimensional visualization according to claim 1, characterized in that: When generating the four-dimensional spatiotemporal occupancy volume, the task spatialization module divides the four-dimensional spatiotemporal occupancy volume into static occupancy volume, dynamic occupancy volume, and circulating occupancy volume. The static occupancy volume corresponds to the fixed space occupied by the maintenance equipment body; the dynamic occupancy volume corresponds to the dynamically changing space formed by the operating range of maintenance personnel, the movement trajectory of tools, and the reach of large equipment; and the circulating occupancy volume corresponds to the channel space occupied by the preset path of spare parts from the warehouse to the maintenance point.

4. The maintenance task scheduling and optimization system based on three-dimensional visualization according to claim 1, characterized in that: The three-dimensional scheduling module follows the following priority strategy during iterative adjustments: First priority: Time shifting of conflicting low-priority tasks without violating task logic constraints; Second priority: If time translation fails to eliminate the conflict, then a collision-free path is replanned for the flow of the conflicting task in the three-dimensional digital environment. Third priority: If the path replanning adjustment is ineffective, mark the conflict as a strong conflict point and output alternative resource solutions.

5. A maintenance task scheduling optimization method based on three-dimensional visualization, characterized in that, Includes the following steps: Step S1: Construct a high-fidelity 3D digital environment; Step S2: Receive the list of maintenance tasks to be scheduled, and convert each maintenance task instruction into a set of spatiotemporal parameters associated with the three-dimensional digital environment. The conversion includes generating a four-dimensional spatiotemporal occupancy volume for each maintenance task, which includes the three-dimensional spatial range and time period required for its execution. Step S3: Generate an initial scheduling sequence based on the preset priority and pre- and post-task logical dependencies of the tasks; Step S4: Load the initial scheduling sequence into the three-dimensional digital environment, and scan all four-dimensional spatiotemporal occupants in chronological order to detect whether there are two or more four-dimensional spatiotemporal occupants that geometrically overlap or intrude into the safe distance on any time slice. Step S5: If a conflict is detected in step S4, the maintenance task that caused the conflict is scheduled and adjusted according to the preset conflict resolution rules, and steps S4 and S5 are repeated until no conflict is detected in step S4, thus obtaining an optimized scheduling scheme. Step S6: Output the optimized scheduling scheme and perform dynamic simulation in the three-dimensional digital environment to present the execution process of the scheduling scheme in a visual manner.

6. The maintenance task scheduling optimization method based on three-dimensional visualization according to claim 5, characterized in that: Step S1 further includes establishing a data mapping channel that is synchronized in real time with the physical site. The data mapping channel performs the following operations: Through the Internet of Things (IoT) gateway, real-time access and analysis of equipment operating data and personnel or equipment location data uploaded by field sensors are achieved. The parsed data is bound to model instances in the 3D digital environment, and the attributes, colors, or positions of the model instances are automatically changed based on real-time data changes.

7. The maintenance task scheduling optimization method based on three-dimensional visualization according to claim 5, characterized in that: In step S2, the operation of generating the four-dimensional spatiotemporal occupancy volume includes: Step S201: Identify the equipment associated with the maintenance task in the three-dimensional digital environment and generate a static spatial envelope based on its geometric shape; Step S202: Based on the personnel operation safety radius and tool usage space defined in the maintenance procedure, a dynamic operation space is generated by expanding the periphery of the static space envelope. Step S203: Combine the estimated start and end times of the maintenance task execution, stretch the static and dynamic spaces in the time dimension to form the four-dimensional spatiotemporal occupancy volume.

8. The maintenance task scheduling optimization method based on three-dimensional visualization according to claim 5, characterized in that: The spatiotemporal conflict scan in step S4 includes: Hard collision detection: Detects whether equipment entities, tool entities, or personnel entities with different maintenance tasks physically overlap in space at the same time; Soft conflict detection: Detects whether the dynamic operating space or flow path of one maintenance task intrudes into the safety warning area or restricted area set by another maintenance task.

9. The maintenance task scheduling optimization method based on three-dimensional visualization according to claim 5, characterized in that: The conflict resolution rule in step S5 is a hierarchical adjustment strategy based on conflict type, which includes: For soft conflicts, prioritize adjusting the execution time of low-priority maintenance tasks; For hard conflicts, first try to adjust the execution time. If this is ineffective, then replan the equipment or material transportation path of the conflict maintenance task in the three-dimensional digital environment. If the path cannot be replanned, then decompose the maintenance task into multiple sub-tasks and regenerate a smaller four-dimensional spatiotemporal occupancy volume for each sub-task.

10. The maintenance task scheduling optimization method based on three-dimensional visualization according to claim 5, characterized in that: The visualization in step S6 includes executing at least two of the following interactive verification modes: Global Preview Mode: Based on the timeline of the optimized scheduling plan, the parallel activity trajectories and space occupancy changes of all maintenance tasks, personnel, equipment, and materials are dynamically displayed globally in an accelerated playback manner, and maintenance tasks on the critical path are highlighted. Focus Perspective Mode: In response to the user's pause and selection commands during the simulation, after selecting any maintenance task, personnel or equipment, the system enters the perspective view, semi-transparent other unrelated models, and highlights the complete four-dimensional spatiotemporal occupancy of the selected object and its activity space range throughout the entire task cycle, while listing the tasks that are spatiotemporally related to it.

Images