Multi-view inspection task intelligent scheduling method based on BIM data integration

Abstract

The invention discloses a multi-view inspection task intelligent scheduling method based on BIM data integration, and relates to the related technical field of power plant inspection, and the method comprises the steps: carrying out the three-dimensional reconstruction of a target power plant, and determining a power plant BIM model; receiving a multi-view inspection task under timestamp constraint, performing space-time dimension decoupling, and determining a task set; a scheduling factor matrix in a power plant operation and maintenance scene is mined, a coding decision maker is introduced, coding sequence conversion based on scheduling factors is carried out, task side-execution side bidirectional preemptive pairing is executed, hierarchical verification simulation is executed, concurrent inspection instructions are generated, and inspection execution end codes are marked on the concurrent inspection instructions. The technical problems that in the prior art, power plant inspection tasks depend on artificial experience, the efficiency is low, multi-view inspection task arrangement is difficult to process effectively, and the inspection scheduling efficiency and quality are poor are solved, intelligent scheduling of the multi-view inspection tasks is achieved, and the technical effect of improving the inspection efficiency and quality is achieved.

Description

Technical Field

[0001] This application relates to the technical field of power plant inspection, and particularly to an intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration. Background Art

[0002] In modern power plant operations, it is crucial to ensure the stable operation of equipment and promptly detect and handle potential faults. Traditional inspection methods often rely on manual experience, resulting in low efficiency and prone to omissions. With the development of digital technology, Building Information Modeling (BIM) technology has gradually been applied to the power plant field. BIM can perform three-dimensional reconstruction of power plant facilities and provide detailed spatial and attribute information. However, the inspection tasks in power plants are complex and diverse, the inspection requirements from different perspectives are increasing, and there are time constraints. How to efficiently integrate BIM data and achieve intelligent scheduling of multi-perspective inspection tasks has become an urgent problem to be solved. Currently, there is a lack of effective methods to reasonably arrange multi-perspective inspection tasks under timestamp constraints, and it is difficult to make full use of BIM models to improve the inspection efficiency and quality in complex power plant operation and maintenance scenarios.

[0003] In the current related technologies, there are technical problems such as low efficiency of power plant inspection tasks relying on manual experience and difficulty in effectively handling the arrangement of multi-perspective inspection tasks, resulting in poor inspection scheduling efficiency and quality. Summary of the Invention

[0004] This application provides an intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration, which solves the technical problems in the prior art that power plant inspection tasks rely on manual experience with low efficiency and difficulty in effectively handling the arrangement of multi-perspective inspection tasks, resulting in poor inspection scheduling efficiency and quality, realizes intelligent scheduling of multi-perspective inspection tasks, and achieves the technical effect of improving inspection efficiency and quality.

[0005] This application provides an intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration, including: performing three-dimensional reconstruction on the target power plant to determine the power plant BIM model, which includes a first BIM layer and a second lightweight layer; receiving multi-perspective inspection tasks under timestamp constraints, performing spatio-temporal dimension decoupling to determine a task set, with the smallest single-thread task as the decoupling standard; mining the scheduling factor matrix in the power plant operation and maintenance scenario, through introducing an encoding decision maker, performing encoding sequence conversion based on scheduling factors on the task set and the inspection terminal, and executing two-way preemption pairing between the task side and the execution side, then flowing to the power plant BIM model to perform hierarchical verification simulation to generate concurrent inspection instructions, where each concurrent inspection instruction is marked with an inspection execution terminal code.

[0006] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: establishing an interactive communication among the inspection execution terminal - the power plant BIM model - the coding decision maker, where the inspection execution terminal consists of a power plant inspection system, including a manual inspection part and an automated inspection part.

[0007] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: setting a coding specification; through multi-dimensional real-time data transmission of the target power plant, traversing the scheduling factor matrix for matching based on execution-side factors and coding conversion based on the coding specification to determine a first coding set; for the task set, traversing the scheduling factor matrix for matching based on task-side factors and coding conversion based on the coding specification to determine a second coding set; performing two-way preemption pairing on the first coding set and the second coding set to determine a task scheduling strategy.

[0008] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: each inspection execution terminal corresponds to a first coding sequence, and each task in the task set corresponds to a second coding sequence.

[0009] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: for the task side - execution side, performing a forward preemption decision on the first coding set to the second coding set to determine a first one-way strategy; performing a reverse preemption decision on the second coding set to the first coding set to determine a second one-way strategy; determining the task scheduling strategy according to the first one-way strategy and the second one-way strategy.

[0010] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: performing two-way preemption pairing on the first one-way strategy and the second one-way strategy. If the pairing is successful, determining a first task scheduling strategy; if the pairing fails, performing two-way preemption pairing based on secondary priorities, and iterating until the pairing ends to determine a second task scheduling strategy; adding the first task scheduling strategy and the second task scheduling strategy to the task scheduling strategy.

[0011] In a possible implementation, the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration further performs the following processing: if there is a preemption collision, performing a preemption game; where performing the preemption game further includes: if it is a sequential collision, performing a sequential game; if it is a same-order collision, performing a priority avoidance game.

[0012] In a possible implementation, the multi-perspective inspection task intelligent scheduling method based on BIM data integration also performs the following processing: in combination with the power plant BIM model, the task scheduling strategy is executed based on the second lightweight layer scheduling simulation and tuning, and based on the first BIM layer preview simulation, to generate the concurrent inspection instructions; by identifying the inspection execution end code, the concurrent inspection instructions are interactively managed.

[0013] In a possible implementation, the multi-perspective inspection task intelligent scheduling method based on BIM data integration also performs the following processing: based on the second lightweight layer, perform two-dimensional scheduling simulation and trajectory collision optimization to determine the pre-scheduling strategy; based on the first BIM layer, perform a preview simulation of the pre-scheduling strategy to determine the driving information based on the strategy inspection cycle; and generate the concurrent inspection instructions based on the driving information.

[0014] In a possible implementation, the multi-perspective inspection task intelligent scheduling method based on BIM data integration also performs the following processing: dividing the power plant inspection tasks into daily inspection category-engineering inspection category-follow-up inspection category; determining rigid inspection conditions and flexible inspection conditions based on daily inspection category-engineering inspection category-follow-up inspection category, wherein the flexible inspection conditions are marked with scheduling slack; and using the rigid inspection conditions and the flexible inspection conditions to perform scheduling decision constraints on the task set.

[0015] In a possible implementation, the multi-view inspection task intelligent scheduling method based on BIM data integration also performs the following processing: by performing inspection tracking and information collection, the information is transmitted back to the first BIM layer of the power plant BIM model; the first BIM layer is updated to perform layer updates and provide feedback inspection guidance Through the intelligent scheduling method of multi-perspective inspection tasks based on BIM data integration proposed in this application, the target power plant is reconstructed in three dimensions to determine the BIM model of the power plant; the multi-perspective inspection tasks under timestamp constraints are received, the time and space dimensions are decoupled, and the task set is determined; the scheduling factor matrix under the power plant operation and maintenance scenario is mined, and the coding sequence conversion based on the scheduling factor is performed by introducing a coding decision maker, and the task side-execution side two-way preemptive pairing is performed, and a hierarchical verification simulation is performed to generate concurrent inspection instructions, wherein each concurrent inspection instruction is marked with an inspection execution end code. The technical problems existing in the prior art that the power plant inspection tasks rely on manual experience with low efficiency and it is difficult to effectively handle the multi-perspective inspection task arrangement, resulting in poor inspection scheduling efficiency and quality are solved, and the intelligent scheduling of multi-perspective inspection tasks is realized, achieving the technical effect of improving inspection efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of this application. It should be understood that the operations described above or below do not necessarily need to be executed precisely in sequence. On the contrary, as needed, various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 It is a schematic flowchart of the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration provided by the embodiments of this application.

[0018] Figure 2 It is a schematic flowchart of the coding sequence conversion in the intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration provided by the embodiments of this application. Detailed implementation manners

[0019] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented in accordance with the content of the description. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically gives the detailed implementation manners of this application.

[0020] In order to make the purpose, technical solutions, and advantages of this application clearer, the following will further describe this application in detail with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0021] In the following descriptions, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and they can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, product, or server including a series of steps does not necessarily have to be limited to those steps clearly listed, but may include other steps not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0022] The embodiments of this application provide an intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration, asFigure 1 As shown in, the method includes: Step S100, perform three-dimensional reconstruction on the target power plant to determine the power plant BIM model, where the BIM model includes a first BIM layer and a second lightweight layer.

[0023] Preferably, performing three-dimensional reconstruction on the target power plant specifically includes collecting the original design drawings of the power plant, including detailed drawings of architecture, structure, electrical, HVAC, water supply and drainage, etc., to understand the layout and structure of the power plant, which contains a large amount of geometric information such as dimensions, shapes, positions, etc., as well as attribute information such as the specifications and models of equipment and components; then obtaining relevant records during the construction process, such as construction logs, change records, acceptance reports, etc.; using measuring instruments to conduct on-site measurements of the power plant to obtain data such as the dimensions and positions of key parts, which are used to supplement and correct the drawing information. Then, using BIM modeling software, such as Autodesk Revit, Bentley Architecture, etc., according to the collected design drawings and measurement data, create the building structure model of the power plant and add various equipment models one by one according to the actual positions and connection relationships of the equipment, such as generators, transformers, boilers, fans, pumps, etc.; then collect the detailed parameters of various equipment in the power plant, such as the external dimensions, interface positions, operating parameters, etc. of the equipment, which are used to accurately construct the equipment model and conduct the connection and layout design of the equipment with pipelines, lines, etc. in the BIM model, and then generate the power plant BIM model; then perform lightweight processing on the complete power plant BIM model, that is, by simplifying the geometric structure of the model, compressing texture data, optimizing the model algorithm, etc., to reduce the data volume of the model without affecting the main features and accuracy of the model, and improve the loading speed and operation efficiency of the model; in addition, using the collision check function of BIM software, conduct a comprehensive collision detection on the building structure, equipment, pipelines, lines, etc. in the model to check whether there are spatial conflicts between different components, such as whether there is a collision between the pipeline and the beam, or between the equipment and the wall, to ensure the stability and reliability of the model in practical applications.

[0024] Preferably, the power plant BIM model includes two layers, namely the first BIM layer and the second lightweight layer. Specifically, the first BIM layer is usually a complete and detailed building information model layer, which contains the accurate three-dimensional geometric information of all building structures, equipment, pipelines, lines and other various facilities in the power plant, as well as the detailed attribute information of these components, such as material, specification, model, manufacturer, etc. The first BIM layer is constructed based on various data sources such as original design drawings, construction data, and on-site actual measurements, aiming to present the actual physical form and detailed information of the power plant as accurately as possible; the second lightweight layer is a layer obtained by simplifying and optimizing the model on the basis of the first BIM layer. Since the data volume of the complete BIM model is often very large, in practical applications, especially in some scenarios with high requirements for data transmission speed and display performance, such as viewing on mobile devices or network transmission, directly using the original BIM model may cause problems such as slow loading and running jams. Therefore, the first BIM layer model is lightweighted, removing some detail information that is not very important for the current application scenario, reducing the data volume of the model, while retaining the main structure and key features of the model, so as to improve the loading speed and running efficiency of the model on the premise of ensuring a certain model accuracy, and facilitate quick display and interaction on different devices and platforms.

[0025] Furthermore, step S100 further includes step S110, where through inspection tracking and information collection, it is transmitted back to the first BIM layer of the power plant BIM model; step S120, performing layer update on the first BIM layer to provide feedback on inspection guidance.

[0026] Preferably, during the power plant inspection tour, by means of positioning technologies (such as GPS, indoor positioning systems, etc.) and mobile devices, the positions and inspection paths of the inspection personnel are recorded in real time to ensure that the inspection personnel conduct inspections according to the specified routes and times. At the same time, it is also convenient to quickly locate the positions of the inspection personnel in case of problems. The inspection personnel use mobile terminals or other professional devices to collect various information related to the inspection objects, including the operating status of the equipment (such as parameters like temperature, pressure, vibration, etc.), the appearance of the equipment (whether there is wear, corrosion, leakage, etc.), on-site environmental information (such as humidity, noise, etc.), as well as descriptions, photos or videos of the abnormalities found. The collected inspection tracking data and information are transmitted back to the first BIM layer of the power plant BIM model through network communication technologies (such as Wi-Fi, 4G / 5G, etc.). The real-time information during the actual inspection process is associated and integrated with the basic information in the BIM model, and the relevant data in the first BIM layer are updated. For example, if it is found during the inspection that the actual operating parameters of a certain equipment are different from the preset parameters in the model, or if new wear conditions are found on the equipment, an update is made so that the BIM model can accurately reflect the current actual state of the power plant equipment and facilities. Finally, based on the updated information of the first BIM layer, the problems found during the inspection and the overall operating condition of the equipment are analyzed, and targeted feedback inspection guidance information is generated and sent to the inspection personnel to help them better conduct subsequent inspection work, improve the inspection efficiency and quality, timely discover and handle potential problems, and ensure the safe and stable operation of the power plant equipment.

[0027] Step S200: Receive multi-perspective inspection tasks under timestamp constraints, perform spatio-temporal dimension decoupling, and determine the task set, where the decoupling standard is the minimum single-thread task.

[0028] Preferably, receive multi-perspective inspection tasks under timestamp constraints. Here, the timestamp constraint means that the inspection task has time limitations and requirements, such as specifying the start time, end time of the task, or requiring certain inspection contents to be completed within a specific time period. The multi-perspective inspection task means that the inspection work needs to be carried out from multiple different angles or aspects, which may include inspections of different areas and different equipment in the power plant, as well as evaluations from different professional perspectives (such as electrical, mechanical, instrumentation, etc.). Then, perform spatio-temporal dimension decoupling on the multi-perspective inspection task. Specifically, in the spatial dimension, the power plant is a place with a complex spatial structure, and different equipment and areas have different spatial positions. In the inspection task, it is necessary to clarify the specific position and scope of each task in the power plant space, such as a specific section of a certain pipeline or a specific area of a certain workshop. In the time dimension, combined with the above-mentioned timestamp constraint, it is necessary to determine the time arrangement of each inspection task, whether it is to be executed immediately, or at a specific time point or within a specific time period.

[0029] Preferably, taking the minimum single-threaded task as the decoupling criterion, the two interrelated dimensions of time and space are separated for analysis and processing to more clearly understand and manage the inspection tasks, avoid the complexity brought by the interweaving of time and space factors, and thus more effectively plan and schedule tasks. Specifically, it is divided according to the functional attributes of the inspection tasks. For example, in the power plant inspection, the tasks can be divided into modules such as equipment status inspection, line connection detection, and environmental parameter monitoring. Each module is further subdivided into specific tasks. For example, the equipment status inspection can be further divided into generator temperature detection, transformer voltage monitoring, etc.; or the inspection areas can be divided according to the physical space layout of the power plant, such as different workshops, floors or functional areas. An independent set of inspection subtasks is formulated for each area. These subtasks have clear boundaries in space and the inspection sequence and time window can be arranged according to factors such as the importance of the area and the operating conditions of the equipment. For example, first inspect the area where the key equipment is located, and then inspect the auxiliary area. Through the analysis after decoupling the time and space dimensions, the entire multi-perspective inspection task is decomposed into multiple specific and relatively independent small tasks, constituting a task set. For example, the inspection of a certain large-scale equipment is decomposed into inspection tasks for its different components at specific time points, and each inspection task becomes an element in the task set. Among them, the minimum single-threaded task refers to the smallest task unit that can be independently executed and cannot be further divided in the entire inspection task system. Taking it as the decoupling criterion means decomposing and dividing the multi-perspective inspection task according to this smallest and independent task unit, which can more finely control and manage the inspection tasks. Each minimum single-threaded task can be individually scheduled, allocated resources, and monitored for the execution status, which helps to improve the efficiency and accuracy of task execution and avoid interference between tasks.

[0030] Step S300, mining the scheduling factor matrix in the power plant operation and maintenance scenario, by introducing a coding decision maker, performing a coding sequence conversion based on scheduling factors for the task set and the inspection terminal, and executing two-way preemption pairing between the task side and the execution side, and transferring to the power plant BIM model for hierarchical verification simulation to generate concurrent inspection instructions, where each concurrent inspection instruction is marked with an inspection execution terminal code.

[0031] Preferably, during the operation and maintenance of a power plant, there are multiple factors affecting the scheduling of inspection tasks, which may include task factors (the urgency, importance, estimated execution time, required resources, etc. of the tasks), inspection terminal factors (the skill level, workload, current location of the inspection personnel, the status and performance of the inspection equipment, etc.), and environmental factors (the danger level of different areas in the power plant, the operating status of the equipment, the real-time weather conditions, etc.); then these factors are quantified and sorted to construct a scheduling factor matrix. For example, the rows of the matrix can represent different inspection tasks, the columns can represent each scheduling factor, and the elements in the matrix are the quantified values of the corresponding tasks under each factor. Then, a coding decision-maker is introduced to perform a coding sequence conversion based on the scheduling factors for the task set and the inspection terminal, that is, to code the task set and the inspection terminal so that their features can be presented in a digital and standardized form. For example, tasks can be coded into different digital combinations according to factors such as urgency and importance; the inspection terminal can be coded according to personnel skills, equipment status, etc. Among them, the coding decision-maker is a tool with specific rules for converting the relevant information of the task set and the inspection terminal into a specific coding sequence according to the scheduling factor matrix.

[0032] Preferably, then perform two-way preemption pairing between the task side and the execution side, that is, match the tasks and the inspection terminals with each other, and not only consider the requirements of the tasks for the inspection terminals, but also consider the adaptability of the inspection terminals to the tasks. Preemption means that during the matching process, when a more suitable inspection terminal or task appears, it can replace the previous pairing. Specifically, according to the coding sequence, combined with the requirements of the tasks (such as skill requirements, time requirements, etc.) and the capabilities of the inspection terminals (such as personnel skills, equipment conditions, etc.), perform optimal pairing, including determining the coding sequence based on the scheduling factors, and according to the trained built-in rules, each execution terminal preempts the tasks in the task set. At the same time, when there is a preemption collision between the execution terminals, they play a game with each other to determine a set of results; each task terminal preempts the execution entity, and at the same time, when there is a preemption collision between the tasks, they play a game with each other to determine another set of results; match the two sets, that is, if the mutual pairing and double selection are successful, it belongs to the best scheduling state. For those that are not successfully paired, that is, the remaining small amount, they are allocated based on conventional means. For example, an urgent task that requires specific skills will be preferentially assigned to an inspection terminal with matching skills and a relatively small current workload; and an inspection terminal will also preferentially select a task that matches its capabilities and location.

[0033] Preferably, after successful pairing, it is transferred to the power plant BIM model for hierarchical verification and simulation. That is, the paired tasks and inspection terminal information are imported into the power plant BIM model, and verification and simulation are carried out according to different levels (such as spatial level, task type level, etc.), including checking whether the execution path of the task is reasonable, whether it will conflict with other equipment or tasks, and simulating and predicting the execution effect of the task and possible problems. Specifically, check whether the execution paths of different inspection tasks in the same spatial layer will conflict with each other, whether it will cause personnel or equipment to appear at the same location at the same time. For example, two different inspection routes cannot be carried out simultaneously in the same narrow passage to avoid collisions and safety accidents; for tasks with a sequential or dependent relationship, formulate rules to ensure that the subsequent task can only be carried out after the previous task is completed. For example, before carrying out equipment maintenance tasks, it is necessary to first complete the pre-task such as shutting down the equipment and safety isolation; according to the resource requirements of the inspection task (such as manpower, tools, equipment, etc.) and the resource allocation of the inspection terminal, check whether the resource allocation is reasonable, whether there is a shortage or waste of resources. For example, a certain task requires a specific detection instrument, and whether the allocated inspection terminal has the instrument and is in an available state.

[0034] Preferably, according to the time requirements of the task and the operation plan of the power plant, check whether the arrangement of the inspection task meets the time limit. For example, the inspection of some equipment must be carried out within a specific operation time period, or a certain emergency task needs to be completed within the specified time; use the visualization function of the BIM model to simulate the execution of the inspection task according to the set hierarchical and verification rules. During the simulation process, display information such as the movement path of the inspection personnel or equipment in the power plant, the order and time nodes of task execution in a graphical manner, and finally generate corresponding concurrent inspection instructions, indicating that multiple inspection tasks can be carried out simultaneously, and clarify the specific content, execution time, execution path, etc. of each task; among them, each concurrent inspection instruction is added with the corresponding inspection execution terminal code to ensure that during the execution process, it can be clearly known which inspection terminal is responsible for executing each task, which is convenient for management and tracking. For example, when a problem occurs with a certain task, the corresponding inspection personnel or equipment can be quickly found, thereby improving the efficiency and quality of power plant inspection.

[0035] Furthermore, step S300 also includes establishing an interactive communication between the inspection execution terminal - power plant BIM model - coding decision maker, where the inspection execution terminal consists of a power plant inspection system, including a manual inspection part and an automated inspection part.

[0036] Preferably, an interactive communication is established among the inspection execution terminal - the power plant BIM model - the coding decision maker. Specifically, the inspection execution terminal (manual inspection personnel through handheld devices, and automated inspection devices through the network) sends requests to the power plant BIM model to obtain detailed information related to the current inspection task, such as the 3D model of the equipment, operation manuals, historical maintenance records, etc., in order to better complete the inspection task; the power plant BIM model feeds back real-time equipment status information, spatial location information, etc. to the inspection execution terminal to help inspection personnel quickly locate the equipment position, understand the current operating status of the equipment, and timely discover potential problems. For example, when the automated inspection device detects abnormal operating parameters of a certain equipment, the power plant BIM model can immediately send the detailed information and location information of the equipment to the nearest manual inspection personnel so that they can go to the site for further inspection.

[0037] Preferably, the inspection execution terminal sends its own status information to the coding decision maker, including the current position, workload, skill level of the manual inspection personnel, and the operating status, power level, etc. of the automated inspection device; the coding decision maker generates and sends inspection task instructions to the inspection execution terminal according to the received status information of the inspection execution terminal and the overall operation of the power plant, specifying requirements such as the content, execution time, execution location of the task. For example, the coding decision maker decides to dispatch the nearest manual inspection personnel or automated inspection device with the corresponding skills to the faulty equipment for inspection according to the equipment fault alarm information and the status of the inspection execution terminal, and sends the task instructions to the corresponding inspection execution terminal.

[0038] Preferably, the power plant BIM model provides the power plant's static information (such as equipment layout, building structure, etc.) and real-time dynamic information (such as equipment operating status, space occupancy, etc.) to provide data support for the coding decision maker to perform task allocation and decision-making; the coding decision maker encodes and plans the task according to the information provided by the power plant BIM model and the requirements of the inspection task, and feeds back the encoded task information to the power plant BIM model for simulation and display in the model. For example, the coding decision maker plans an optimal inspection route according to the position and status information of the equipment in the power plant BIM model and sends the information of this route to the power plant BIM model for visual display in the model, which is convenient for inspection personnel to view.

[0039] Preferably, the inspection execution end consists of a power plant inspection system, including a manual inspection part and an automated inspection part. Among them, the manual inspection part is composed of the inspection personnel of the power plant. They conduct on-site inspections of the equipment and facilities of the power plant according to the specified inspection routes and time intervals, using portable detection equipment (such as thermometers, vibration meters, etc.), and obtain the operation status information of the equipment through methods such as visual observation and instrument measurement, such as whether there is abnormal noise in the equipment, whether the temperature is too high, whether the instrument readings are normal, etc.; the automated inspection part uses automated equipment such as sensors, monitoring cameras, and robots installed in various key parts of the power plant to complete the inspection tasks. The automated equipment can collect the operation data of the equipment in real time (such as pressure, flow rate, current, voltage, etc.) and transmit the data to the monitoring center through the network. For example, intelligent robots can move autonomously in the power plant according to the preset routes, take pictures and detect the equipment, and upload the image and data information.

[0040] Further, as Figure 2 shown, step S300 further includes step S310 of setting the coding specification; step S320 of traversing the scheduling factor matrix through multi-dimensional real-time data transmission of the target power plant, performing matching based on the execution-side factors and coding conversion based on the coding specification to determine the first coding set; step S330 of traversing the scheduling factor matrix for the task set, performing matching based on the task-side factors and coding conversion based on the coding specification to determine the second coding set; step S340 of performing two-way preemption pairing on the first coding set and the second coding set to determine the task scheduling strategy.

[0041] Preferably, the coding specifications are uniformly set to convert various information related to power plant operation and maintenance into a specific coding form, which stipulates how to code different types of data, including the format, length, value range of the data, and the specific meaning represented by the coding, etc. For example, it may be stipulated that a specific number of digits represent the equipment type, and another number of digits represent the operating status of the equipment, etc., so that information from different sources and of different types has consistency and standardization after coding; then various sensors, monitoring devices, etc. in the target power plant are used to collect various data during the operation of the power plant in real time, such as the temperature, pressure, current, voltage of the equipment, etc., as well as the relevant data of the inspection execution end (including the working status and location information of manual inspection personnel, the operating parameters of automated inspection equipment, etc.) and transmit them back; then the corresponding execution-side factors are found in the scheduling factor matrix. For example, according to the qualification certificate information of manual inspection personnel, their skill level factors are matched, and according to the power and task completion progress of automated inspection equipment, their equipment availability status and workload factors are matched; then according to the pre-set coding specifications, these factors are converted into a specific coding form. For example, if the coding specification stipulates that the skill level of "advanced" is represented by the number "01", then when the skill level of a certain manual inspection personnel is matched to be advanced, it is converted into "01". After processing all the transmitted data, a series of codes are obtained and form a first coding set, which mainly reflects the relevant information of the inspection execution end, including the status and capabilities of personnel and equipment.

[0042] Preferably, for each task in the task set, the relevant task-side factors also need to be found in the scheduling factor matrix, which may include the priority of the task, the required skill type, the time requirement of the task, etc. For example, for an urgent equipment failure repair task, its priority is high, it may require personnel with specific skills to complete, and there are strict time limits. By matching these factors in the scheduling factor matrix, the specific requirements of each task can be clarified; similar to the generation of the first coding set, according to the coding specifications, the matched task-side factors are converted into a coding form. For example, it is stipulated that the task priority of "high" is represented by "1", "medium" is represented by "2", and "low" is represented by "3". Then when the priority of a task is high, it is converted into "1". The codes of all tasks are combined to form a second coding set, which mainly describes the characteristics and requirements of the tasks, providing a basis for the matching and scheduling of tasks and the execution end.

[0043] Preferably, the first coding set and the second coding set are compared and matched to find the most suitable inspection execution end to execute each task. At the same time, the priority of the task and the actual situation of the execution end should also be considered. Specifically, during the pairing process, it may occur that multiple execution ends can execute a certain task, or a certain execution end can execute multiple tasks simultaneously. In this case, a "preemption" decision needs to be made according to certain rules. For example, for high-priority tasks, an execution end with a high skill level and a relatively low current workload is preferentially selected; for low-priority tasks, if there is no other more suitable execution end, it can also be assigned to an execution end with a relatively high workload but still capable of completing the task. The two-way matching and preemption process is to find the best matching method between tasks and execution ends. After two-way preemption pairing, it is finally determined which inspection execution end will execute each task, as well as the execution order and time arrangement of the tasks, etc., constituting a task scheduling strategy, that is, comprehensively formulated based on the actual operation situation of the power plant, the requirements of the tasks, and the capabilities and states of the execution ends, etc., to achieve the efficient and orderly progress of the power plant operation and maintenance inspection work, ensure the normal operation of the equipment, timely discover and solve potential problems, and improve the overall operation efficiency and safety of the power plant.

[0044] Further, step S300 also includes that each inspection execution end corresponds to a first coding sequence, and each task in the task set corresponds to a second coding sequence.

[0045] Preferably, each inspection execution end has its unique attributes and status information. For example, in the manual inspection part, the skill level, work experience, current location, working hours, etc. of the inspection personnel; in the automated inspection part, the model, function, power, operating status, etc. of the automated equipment. These information related to the inspection execution end are encoded and converted, and all the relevant information of an inspection execution end is encoded and arranged in a certain order to form the first coding sequence corresponding to this inspection execution end. This sequence is the "digital identity" of this inspection execution end, containing the key information that can distinguish this execution end from other execution ends, as well as various parameters required for subsequent task assignment and scheduling. Each task in the task set also has its specific attributes and requirements, such as the type of the task (whether it is a daily inspection, equipment maintenance or fault repair, etc.), the priority of the task, the type of skills required for the task, the estimated duration of the task, the location of the task, etc. Similarly, according to the coding specifications, these attributes and requirements of the task are converted into coding forms and arranged in a specific order to form the second coding sequence corresponding to each task. This sequence represents the characteristics and requirements of the task and is the digital representation of the task in the entire scheduling system, facilitating the search for a suitable inspection execution end for task assignment. Through this coding sequence method, the complex information of the inspection execution end and the task can be quantified and standardized, thereby realizing efficient power plant operation and maintenance task management.

[0046] Further, step S340 further includes step S341, for the task side - execution side, making a forward preemption decision on the first encoding set to the second encoding set to determine a first one - way strategy; step S342, performing a reverse preemption decision on the second encoding set to the first encoding set to determine a second one - way strategy; step S343, determining the task scheduling strategy according to the first one - way strategy and the second one - way strategy.

[0047] Preferably, the forward preemption decision from the task side to the execution side is based on the requirements and priorities of the tasks, and selects the most suitable execution end among all inspection execution ends to complete the tasks. Specifically, for each task coding sequence in the second encoding set, according to the task information it contains (such as task type, priority, required skills, etc.), find the inspection execution end coding sequence with the highest matching degree in the first encoding set. For example, first consider whether the skills of the execution end exactly match the required skills of the task, and then consider the current workload, location, etc. of the execution end. If a task requires a certain specific skill and a certain inspection execution end has this skill and a relatively low current workload, it is more likely to be selected. Through the forward preemption decision, a preferred inspection execution end is determined for each task. Organizing the matching relationships between these tasks and the corresponding execution ends into a strategy, the first one - way strategy is obtained. It mainly starts from the perspective of the tasks and determines which execution end each task should be executed by ideally, that is, the execution entity of the task is preempted among tasks, namely inspection devices, robots, personnel, etc., to ensure that the tasks can be completed with high quality.

[0048] Preferably, the reverse preemption decision from the execution side to the task side starts from the perspective of the inspection execution end and considers which tasks each execution end can undertake. For each inspection execution end coding sequence in the first encoding set, according to the information such as the capabilities and status of the execution end it represents, find the tasks that the execution end can be competent for in the second encoding set. For example, an automated inspection device has specific detection functions and working ranges, then find the tasks within its functions and ranges from the task set, and at the same time consider factors such as the priority and time requirements of the tasks to determine which tasks the execution end should execute first; after the reverse preemption decision, it is clear which tasks each inspection execution end can undertake and the execution order of the tasks. Organizing this information into a strategy, the second one - way strategy is obtained, which more considers the actual situation of the execution end to ensure that each execution end can reasonably undertake tasks within its capabilities and avoid situations where the execution end is over - burdened or unable to complete tasks.

[0049] Preferably, the first one-way policy and the second one-way policy are combined, that is, the requirements of the task and the actual situation of the execution end are comprehensively considered. Specifically, a trade-off and coordination are carried out between the first one-way policy and the second one-way policy. For example, the first one-way policy may assign a certain high-priority task to an execution end with a perfect skill match but a heavy current workload, while in the second one-way policy, there may be other tasks more suitable for the current state of this execution end. At this time, adjustments need to be made according to specific rules and actual situations. The execution priority of this high-priority task may be appropriately reduced, or resources may be added to the execution end to reduce its load, so as to find a balanced solution that can meet the task requirements and make full use of the resources of the execution end. The finally determined task scheduling policy will clarify which inspection and execution end will execute each task at what time and in what order, ensuring that the entire power plant operation and maintenance inspection work can be carried out efficiently and orderly.

[0050] Further, step S343 further includes step A: performing two-way preemption pairing on the first one-way policy and the second one-way policy. If the pairing is successful, determining the first task scheduling policy; step B: if the pairing fails, performing two-way preemption pairing based on secondary priorities, and iterating until the pairing ends to determine the second task scheduling policy; step C: adding the first task scheduling policy and the second task scheduling policy to the task scheduling policy.

[0051] Preferably, the first one-way policy selects an execution end for the task from the task side, and the second one-way policy selects a task for the execution end from the execution end side. The two-way preemption pairing is to combine these two policies and perform two-way matching adjustments, that is, comprehensively considering the requirements of the task (such as task priority, skill requirements, etc.) and the capabilities and states of the execution end (such as workload, skill level, etc.). For example, for the execution end selected for a certain task in the first one-way policy, check in the second one-way policy whether this execution end also selects this task as a suitable choice. If both sides think it is a suitable match, that is, the pairing is successful, the first task scheduling policy is formed, which clarifies which execution end will execute each task and information such as the execution order.

[0052] Preferably, if in the first two-way preemption pairing, there are some tasks and execution ends that cannot achieve a suitable match, that is, the pairing fails. This may be because the high-priority requirements of the tasks do not match the actual capabilities or status of the execution ends, or there are conflicts in the matching priorities of both sides. Then, the two-way preemption pairing is carried out again according to the preset secondary priority rules, that is, some strict requirements of the tasks for the execution ends are reduced (such as slightly relaxing the skill level requirements), or the selection criteria of the execution ends for the tasks are adjusted (such as allowing the execution ends to undertake some tasks with slightly lower priorities). During the re-pairing process, various factors of the tasks and the execution ends are also comprehensively considered to find new matching combinations. If there are still unpaired tasks and execution ends after the two-way preemption pairing based on the secondary priority, the pairing will continue according to the next-level priority rules, and so on in a loop iteration, continuously adjusting the matching strategy until all tasks can be successfully paired with suitable execution ends, or the preset maximum iteration times are reached; the finally formed matching scheme of tasks and execution ends is determined as the second task scheduling strategy, which takes into account the matching relationship between tasks and execution ends under various priority situations, ensuring that even in the case of unsuccessful initial matching, the execution of tasks can be arranged as reasonably as possible. Finally, both the first task scheduling strategy and the second task scheduling strategy are added to the task scheduling strategy. By combining the two, the task scheduling strategy can more flexibly and comprehensively handle various situations of tasks and execution ends that may occur in the power plant operation and maintenance inspection, improving the reliability and efficiency of the entire inspection work.

[0053] Further, step A further includes that if there is a preemption collision, a preemption game is executed. Among them, the execution of the preemption game further includes: a: if it is a sequential collision, a sequential game is executed; b: if it is a same-order collision, a priority avoidance game is executed.

[0054] Preferably, during task scheduling, preemption collisions occur when there are resource competitions or task assignment conflicts among multiple tasks or execution terminals. For example, when two tasks both require the same inspection and execution terminal with special skills to perform tasks at the same time, or when an execution terminal is simultaneously assigned multiple tasks that cannot be completed simultaneously, a conflict occurs, and these conflicts need to be resolved through preemption games to determine the final task assignment plan. Specifically, sequential collision refers to the situation in task scheduling where the preemption requests of multiple tasks or execution terminals arrive sequentially in a certain order, but there are conflicts among these requests. For example, task A requests execution terminal X first, and then task B also requests execution terminal X, and execution terminal X cannot process these two tasks simultaneously, thus forming a sequential collision. When a sequential collision occurs, a sequential game needs to be executed to resolve it. Among them, the sequential game is a dynamic game process that considers the order of requests of each task or execution terminal and their mutual relationships. In a sequential game, the later actor can observe the choices of the earlier actor and then make its optimal decision based on this information. For example, in the above example, when task B requests execution terminal X, it knows that task A has already requested execution terminal X first. Then task B needs to decide whether to wait for execution terminal X to complete task A before executing, or try to find other alternative execution terminals, or negotiate with task A to adjust the execution order, etc., based on factors such as the priority of task A, its own urgency, and the available time of execution terminal X.

[0055] Preferably, same-order collision refers to the situation where multiple tasks or execution terminals submit preemption requests at the same time and have the same priority or status to a certain extent, resulting in the inability to directly determine which request should be satisfied first. For example, when two tasks have the same priority and request the same execution terminal at the same time, a same-order collision occurs at this time. For same-order collisions, a priority avoidance game needs to be executed to resolve them. Among them, the priority avoidance game is a negotiation mechanism based on priority rules. In this case, although multiple requests have the same initial priority, their priorities can be further distinguished according to some additional factors, or some requests can be negotiated to temporarily avoid to satisfy more important or urgent tasks. For example, the priorities of tasks can be temporarily adjusted according to factors such as the urgency of the tasks and the degree of impact on the power plant operation, so that relatively less urgent or important tasks avoid, and the execution terminal first executes more critical tasks; or through negotiation, the relevant parties of the two tasks jointly determine a temporary execution order to avoid conflicts and ensure that the tasks can be reasonably completed. Through the priority avoidance game, a relatively reasonable solution can be found in the case of same-order collisions, ensuring the fairness and efficiency of task scheduling.

[0056] Further, step S300 further includes step S350, which combines the power plant BIM model to perform scheduling simulation and optimization based on the second lightweight layer for the task scheduling strategy, and generates the concurrent inspection instructions through preview simulation based on the first BIM layer; step S360, which performs interactive communication management on the concurrent inspection instructions by identifying the inspection execution end code.

[0057] Preferably, after the task scheduling strategy is determined, the second lightweight layer is used for scheduling simulation, that is, according to the task scheduling strategy, the execution process of the inspection task is simulated in the lightweight BIM model. During the simulation, factors such as the execution order of tasks, the movement path of the inspection execution end, and the time interval between tasks are considered. Through simulation, problems that may exist in the actual execution of the task scheduling strategy can be found, such as whether the inspection path is reasonable, whether there will be conflicts between tasks or excessive waiting time, etc. Then, based on the problems found in the simulation, the task scheduling strategy is optimized, adjusting the task allocation, execution order, or inspection path, etc., to improve the efficiency and smoothness of task execution. After the scheduling simulation and optimization are completed based on the second lightweight layer, a preview simulation is performed based on the first BIM layer to more accurately display the actual situation of task execution, considering more detailed factors such as the specific installation location of equipment and the influence of the surrounding environment. Through the preview simulation on the first BIM layer, the feasibility and effectiveness of the task scheduling strategy can be further verified to ensure that no problems will occur in actual execution. For example, during the preview simulation, it can be checked whether the inspection execution end will encounter obstacles when moving between actual equipment and building structures, or whether the execution of tasks will affect surrounding equipment or personnel. If problems are found during the preview simulation, the task scheduling strategy can be fine-tuned again.

[0058] Preferably, concurrent inspection instructions are generated according to the finally determined task scheduling strategy, which clarifies information such as the specific content, execution time, execution path, and responsible inspection execution end of each inspection task. And since they are concurrent inspection instructions, the coordination and cooperation issues when multiple tasks are executed simultaneously are also considered; then interactive communication management is performed by identifying the inspection execution end code. Specifically, the inspection execution end code responsible for executing the instruction is marked in the generated concurrent inspection instructions. When the inspection execution end receives the instruction, it confirms whether the instruction is assigned to itself by identifying its own code, thereby realizing the interactive communication management of the concurrent inspection instructions. Specifically, it includes the inspection execution end feeding back the execution status of the task (such as started, in execution, completed, etc.), adjusting subsequent tasks or issuing new instructions according to the execution status; the inspection execution ends can also communicate through the code to coordinate cooperation during the task execution process. For example, when multiple inspection execution ends jointly execute a large task, information sharing and collaborative operations are carried out to ensure that the entire inspection task can be completed smoothly and efficiently.

[0059] Furthermore, step S350 further includes step S351, performing two-dimensional scheduling simulation and trajectory collision optimization based on the second lightweight layer to determine a pre-scheduling strategy; step S352, performing a rehearsal simulation on the pre-scheduling strategy based on the first BIM layer to determine driving information based on the inspection cycle of the strategy; step S353, generating the concurrent inspection instruction according to the driving information.

[0060] Preferably, the two-dimensional scheduling simulation refers to simulating the execution of inspection tasks on a two-dimensional plane (usually based on the plant's layout), according to the time arrangement and execution sequence of tasks in the task scheduling strategy. For example, each inspection task is represented by a line or mark on the plane to show its execution path and time progress, and different tasks are distinguished by different colors or symbols; during the simulation, factors such as the start time, estimated completion time, and moving direction on the plane of the task are considered. For example, an inspection task for a certain device starts at point A and moves towards point B at a certain time rhythm, and its dynamic process on the two-dimensional plane is accurately presented in the simulation; it is possible that the execution trajectories of different inspection tasks occupy the same spatial position at the same time point during the simulation, that is, trajectory collision. For example, the inspection routes of two inspectors conflict at a certain passage, which may lead to collision or obstruction, then trajectory collision optimization is carried out, which may include re-planning the execution sequence of tasks, such as appropriately delaying the start time of one of the tasks, or changing the inspection path of some tasks; through continuous optimization, a relatively reasonable pre-scheduling strategy is determined to effectively avoid the collision of inspection task trajectories and ensure the smoothness of task execution.

[0061] Preferably, based on the pre-scheduling strategy determined in the second lightweight layer, a more realistic and comprehensive rehearsal simulation is carried out in the first BIM layer, including simulating the whole process of inspectors or automated inspection equipment executing tasks according to the pre-scheduling strategy in three-dimensional space, considering the actual space limitations, such as the narrow gaps between devices and the complex pipeline layout, etc., on the inspection path. For example, when an inspection robot passes through an area full of pipelines, it needs to make accurate path planning according to the actual space; accurately calculate the actual time required for each inspection task, considering the pauses and waiting times during task execution (such as waiting for the equipment to cool down before detection) and the actual moving speed, etc., and then determine the inspection cycle of the current pre-scheduling strategy; at the same time, other relevant driving information can also be obtained, such as the time interval requirements between different tasks and the time node limitations of key tasks.

[0062] Preferably, based on the driving information of the policy-based inspection cycle obtained from the first BIM layer pre-rehearsal simulation, specific concurrent inspection instructions are generated. Among them, the concurrent inspection instructions will clarify the specific execution time, execution path, responsible inspection execution end, and the coordination relationship between tasks of each inspection task. For example, the instruction will stipulate that inspector A starts at 9 am and inspects the equipment in a certain area according to a specific path, and inspector B starts to execute the task in another area at 9:15 am, and the two tasks need to cooperate with each other in some links (such as sharing certain detection data or collaborating during specific equipment operations); in addition, according to factors such as the inspection cycle and the urgency of the task, the priority of the task is reasonably arranged. For urgent tasks that need to be completed in a short time, resources will be preferentially allocated and execution time will be arranged to ensure the safe and stable operation of the power plant equipment and the efficient completion of the inspection work.

[0063] Further, step S300 further includes step S370 of dividing the power plant inspection tasks into daily inspection category - project inspection category - follow-up inspection category; step S380 of determining the rigid inspection conditions and flexible inspection conditions according to the daily inspection category - project inspection category - follow-up inspection category, wherein the flexible inspection conditions are marked with scheduling slack; step S390 of using the rigid inspection conditions and the flexible inspection conditions to perform scheduling decision constraints on the task set.

[0064] Preferably, the daily inspection category is the routine inspection task of the power plant, and its main purpose is to conduct daily status monitoring and basic inspections on the equipment and facilities of the power plant to ensure its normal operation. For example, check the temperature, pressure, vibration and other parameters of the generator every day, and check whether there are damages, leaks, etc. on the appearance of the equipment; the project inspection category is usually related to the new construction project, renovation project or maintenance project of the power plant, and conducts inspections on the progress, quality, safety, etc. of the project. For example, in the new equipment installation project of the power plant, check whether the installation position of the equipment is accurate, the connection is firm, and the construction technology meets the standards, etc.; the follow-up inspection category is to conduct subsequent follow-up inspections on the previously discovered equipment problems, failures or abnormalities to confirm whether the problems have been solved, whether the treatment measures are effective, and whether new problems have emerged. For example, after the equipment is repaired, conduct multiple follow-up inspections on the repaired parts to check whether the performance of the equipment has returned to normal and whether there are potential hidden dangers.

[0065] Preferably, considering the characteristics of different types of inspection tasks, the importance of equipment, safety requirements, etc., rigid inspection conditions and flexible inspection conditions are determined. Specifically, rigid inspection conditions are inspection requirements and constraints that must be strictly met. For example, for some high-voltage electrical equipment, insulation detection must be carried out at specified time intervals, and the time interval is a rigid inspection condition; flexible inspection conditions refer to inspection requirements that can be adjusted within a certain range, usually related to the priority of the task and the availability of resources. Among them, flexible inspection conditions will be marked with a scheduling slack. The scheduling slack indicates the extent to which this condition can be adjusted without affecting the overall inspection effect. For example, for the daily inspection of some non-critical equipment, its inspection time can be flexibly arranged within a certain time period, and the range of the time period is the scheduling slack. If the scheduling slack is ±2 hours, it means that this inspection task can be carried out within 2 hours before and after the original scheduled time without causing a significant impact on the operation of the power plant.

[0066] Preferably, using rigid inspection conditions and flexible inspection conditions, scheduling decision constraints are imposed on the task set. Specifically, for rigid inspection conditions, they must be strictly adhered to to ensure that the task arrangement does not violate these key constraints. For example, if the rigid inspection condition for a certain equipment is to conduct a comprehensive inspection every 7 days, when scheduling tasks, it must be ensured that the inspection task for this equipment is accurately arranged within the 7-day cycle. For flexible inspection conditions, appropriate adjustments can be made according to the actual situation. During the scheduling process, the scheduling slack can be used to optimize the task arrangement, improve the utilization rate of resources and the execution efficiency of tasks. For example, when the resources of inspection personnel are tight during a certain time period, some flexible inspection tasks with a large scheduling slack can be postponed or advanced appropriately to balance the workload; ensure that it will not affect the rigid inspection conditions and operate within the allowable range of the scheduling slack to achieve the efficient and orderly progress of the power plant inspection work.

[0067] The above specific implementation manners do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be executed in a different order from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

Claims

1. An intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration, characterized in that, The method comprises: Perform three-dimensional reconstruction of the target power plant and determine the BIM model of the power plant, which includes the first BIM layer and the second lightweight layer; Receive multi-view inspection tasks under timestamp constraints, decouple them in time and space dimensions, and determine the task set, where the minimum single-threaded task is used as the decoupling standard; The scheduling factor matrix under the power plant operation and maintenance scenario is mined, and by introducing a coding decision maker, the coding sequence conversion based on the scheduling factors is performed on the task set and the inspection end, and a two-way preemptive pairing between the task side and the execution side is performed. The task is transferred to the power plant BIM model to perform a hierarchical verification simulation and generate concurrent inspection instructions, wherein each concurrent inspection instruction is marked with an inspection execution end code.

2. The intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration according to claim 1, characterized in that Establish interactive communication among the inspection execution end, the power plant BIM model and the coding decision maker, wherein the inspection execution end is composed of the power plant inspection system, including a manual inspection part and an automated inspection part.

3. The intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration according to claim 1, wherein The task set and the inspection execution end are converted into a coding sequence based on scheduling factors, and a task side-execution side bidirectional preemption pairing is performed, including: Set encoding specifications; By transmitting multi-dimensional real-time data back to the target power plant, traversing the scheduling factor matrix to perform matching based on execution-side factors and code conversion based on coding specifications, a first code set is determined; For the task set, traverse the scheduling factor matrix to perform matching based on task side factors and encoding conversion based on encoding specifications to determine a second encoding set; Perform bidirectional preemptive pairing on the first code set and the second code set to determine a task scheduling strategy.

4. The intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration according to claim 3, wherein, Each inspection execution end corresponds to a first coding sequence, and each task in the task set corresponds to a second coding sequence.

5. The intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration according to claim 3, wherein Performing bidirectional preemptive pairing on the first code set and the second code set includes: For the task side-execution side, a forward preemption decision is made from the first code set to the second code set to determine a first unidirectional strategy; Performing a reverse preemption decision on the second code set to the first code set to determine a second unidirectional strategy; The task scheduling strategy is determined according to the first unidirectional strategy and the second unidirectional strategy.

6. The intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration according to claim 5, wherein, Determining the task scheduling strategy according to the first unidirectional strategy and the second unidirectional strategy includes: Performing bidirectional preemptive pairing on the first unidirectional strategy and the second unidirectional strategy, and if the pairing is successful, determining a first task scheduling strategy; If the pairing fails, a two-way preemptive pairing based on the secondary priority is performed, and the cycle is iterated until the pairing is completed, and the second task scheduling strategy is determined; The first task scheduling policy and the second task scheduling policy are added into the task scheduling policy.

7. The intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration according to claim 6, wherein, If there is a preemption collision, execute the preemption game; Among them, executing the preemption game also includes: If it is a sequential collision, execute the sequential game; If it is a collision of the same order, a priority avoidance game is performed.

8. The intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration according to claim 3, wherein Transfer to the power plant BIM model to perform layered verification simulation, including: In combination with the power plant BIM model, the task scheduling strategy is subjected to scheduling simulation and tuning based on the second lightweight layer and preview simulation based on the first BIM layer to generate the concurrent inspection instructions; By identifying the inspection execution end code, interactive communication management is performed on the concurrent inspection instructions.

9. The intelligent scheduling method for multi-perspective inspection tasks based on BIM data integration according to claim 8, wherein Generating the concurrent inspection instruction includes: Based on the second lightweight layer, performing two-dimensional scheduling simulation and trajectory collision optimization to determine a pre-scheduling strategy; Based on the first BIM layer, performing a preview simulation on the pre-scheduling strategy to determine driving information based on the inspection cycle of the strategy; Generating the concurrent inspection instruction according to the driving information.

10. The intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration according to claim 1, wherein, Dividing the power plant inspection tasks into daily inspection category - engineering inspection category - follow-up inspection category; Determining rigid inspection conditions and flexible inspection conditions according to the daily inspection category - engineering inspection category - follow-up inspection category, wherein the flexible inspection conditions are marked with scheduling slack; Using the rigid inspection conditions and the flexible inspection conditions to perform scheduling decision constraints on the task set.

11. The intelligent scheduling method for multi - perspective inspection tasks based on BIM data integration according to claim 1, characterized in that, The method further includes: Through inspection tracking and information collection, transmitting back to the first BIM layer of the power plant BIM model; Performing layer update on the first BIM layer to provide feedback for inspection guidance.

Images