Aviation equipment cluster detection scheduling method under high-frequency dynamic situation

Abstract

The invention discloses an aeronautical equipment cluster detection scheduling method under a high-frequency dynamic situation, which relates to the technical field of detection equipment scheduling, and comprises the following steps: constructing an aeronautical equipment cluster detection scheduling model, analyzing an aeronautical equipment cluster detection task, detecting resource association and scheduling process requirements, and constructing a model of the detection task, the detection resource and the scheduling process; real-time monitoring and advanced prediction of a detection task are realized; the evaluation of the real-time state of the detection resources in the aviation equipment cluster is realized by integrating a plurality of indexes; comprehensively evaluating task urgency and task importance and dividing task priorities; and generating a scheduling scheme. According to the aeronautical equipment cluster detection scheduling method for the high-frequency dynamic situation, the aeronautical equipment cluster detection task, resource association and scheduling requirements can be analyzed to construct a model, the task state is sensed and predicted in real time, the resource state is comprehensively evaluated, task priority division is achieved, and the scheduling efficiency is improved. And an efficient scheduling scheme is generated according to the task state, the resource state and the task priority.

Description

Technical Field

[0001] The present invention belongs to the technical field of manufacturing service scheduling in a service-oriented manufacturing system, and particularly relates to an aviation equipment cluster detection scheduling method under a high-frequency dynamic situation. Background Art

[0002] With the rapid development of the aviation industry, the size of both civil and military fleets continues to expand. To ensure flight safety and improve operational efficiency, airlines are increasingly demanding the frequency and quality of aircraft inspections. A large number of aircraft require regular, diverse inspections. Furthermore, in wartime scenarios, mission arrivals are highly random, and the demand for mission inspections is high. This increases the number and quality of aircraft inspections, leading to a high frequency and dynamic nature of aviation equipment cluster inspections. Traditional aviation equipment cluster inspections rely heavily on manual labor, resulting in low efficiency and accuracy. With the advancement of automation technology, an increasing number of intelligent inspection devices are being used for aviation equipment cluster inspections. These high frequency and dynamic scenarios pose new challenges to the scheduling of inspection resources.

[0003] In high-frequency, dynamic scenarios, tasks execute at a fast pace and in large numbers. Scheduling tasks after they arrive will result in delays, impacting execution efficiency. To enhance system flexibility and adaptability, advance resource planning and rational allocation are crucial. Furthermore, due to the high frequency of inspection tasks, scheduling too many tasks simultaneously places excessive pressure on the algorithm, often making it difficult to generate an optimal solution. Therefore, hierarchical processing of inspection tasks is crucial and can significantly improve scheduling efficiency. Summary of the Invention

[0004] To address the above technical issues, the present invention provides a method for scheduling aviation equipment cluster inspections in high-frequency dynamic scenarios, comprising five steps: constructing an aviation equipment cluster inspection scheduling model, task status perception based on multi-timescale data fusion, resource status assessment based on a multi-dimensional indicator system, a priority division mechanism based on task multi-feature fusion, and scheduling solution generation. The present invention is able to establish an aviation equipment cluster inspection scheduling model suitable for high-frequency dynamic scenarios, and based on the model, perform task status monitoring and task arrival trend prediction, assess inspection resource status, prioritize inspection tasks, and finally generate a scheduling solution based on task status, resource status, and task priority, thereby improving the system's processing capabilities in dealing with high-frequency dynamic problems and maximizing production efficiency.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for scheduling aviation equipment cluster detection in a high-frequency dynamic scenario includes the following steps:

[0007] Step 1: Analyze the difficulties faced by the detection of aviation equipment clusters in high-frequency dynamic scenarios, extract the association between tasks and resources in the detection scenario of aviation equipment clusters, and construct an aviation equipment cluster detection scheduling model from the scenario model, detection task model, detection resource model, and scheduling process model, laying a foundation for subsequent index establishment and scheduling solution;

[0008] Step 2: Based on the aviation equipment cluster detection scheduling model, collect the operation parameters of detection tasks in real time through sensors, and perform data preprocessing through filtering and data cleaning technologies to obtain accurate task status monitoring information; and based on the currently monitored task arrival data and historical task data, perform task data fusion on the time scales of annual, quarterly, and daily, and predict task arrivals based on the fused data, so as to realize task status perception;

[0009] Step 3: Based on the aviation equipment cluster detection scheduling model, construct indicators for resource availability, resource busyness, and resource execution ability, evaluate the current resource status from multiple dimensions, and comprehensively evaluate the overall resource status by combining the three indicators to quantitatively evaluate the system resources and provide a basis for the subsequent execution of scheduling optimization;

[0010] Step 4: Based on the aviation equipment cluster detection scheduling model, construct indicators for task urgency and task importance. The task urgency and task importance indicators integrate task execution time, task benefits, task levels, and task relevance, comprehensively evaluate the task urgency and task importance, divide priorities, and tasks with higher priorities preferentially obtain high-quality detection resources, while low-priority tasks are flexibly arranged when resources are abundant, so as to maximize the task execution benefits in high-frequency dynamic scenarios;

[0011] Step 5: Based on the perceived task status, resource status, and task priority division results, give priority to high-priority detection tasks, use intelligent optimization algorithms to generate high-priority aviation equipment cluster detection scheduling plans. When resources are insufficient, low-priority tasks are in a waiting state. If resources are sufficient, they are provided for low-priority tasks to execute, and intelligent optimization algorithms are used to generate low-priority aviation equipment cluster detection scheduling plans, so as to achieve optimal matching and efficient scheduling of tasks and resources.

[0012] The beneficial effects of the present invention compared with the prior art are as follows:

[0013] (1) The present invention fuses time-scale data and uses a prediction model to predict the task arrival trend. The prediction results are updated in real time, and the model parameters are adjusted according to new tasks and actual situations to improve the prediction accuracy and ensure that the dynamic changes in the aviation equipment cluster detection scenario can be reflected in a timely manner.

[0014] (2) The present invention comprehensively considers the characteristics of task urgency and importance to divide task priorities, and can flexibly adjust weights according to the real-time requirements of the aviation equipment cluster detection scenario, enabling high-priority tasks to preferentially allocate resources to avoid delays and revenue losses, and low-priority tasks to flexibly utilize idle resources, thus maximizing resource efficiency.

[0015] In summary, through the construction of an aviation equipment cluster detection and scheduling model, task status perception based on multi-time-scale data fusion, resource status evaluation based on a multi-dimensional index system, a priority division mechanism based on multi-feature fusion of tasks, and the generation of a scheduling plan, the present invention realizes the hierarchical scheduling of detection tasks in a high-frequency dynamic scenario through predictive task status perception and real-time resource evaluation, so as to achieve the optimal matching and efficient scheduling of tasks and resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a method for detecting and scheduling an aviation equipment cluster in a high-frequency dynamic scenario of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. The present invention will be described in further detail below with reference to the accompanying drawings.

[0018] The present invention discloses a method for detecting and scheduling an aviation equipment cluster in a high-frequency dynamic scenario, which includes five steps: constructing an aviation equipment cluster detection and scheduling model, task status perception based on multi-time-scale data fusion, resource status evaluation based on a multi-dimensional index system, a priority division mechanism based on multi-feature fusion of tasks, and the generation of a scheduling plan. The present invention can establish an aviation equipment cluster detection and scheduling model suitable for high-frequency dynamic scenarios, and based on the model, monitor task status, predict task arrival trends, evaluate the status of detection resources, divide the priorities of detection tasks, and finally generate a scheduling plan based on task status, resource status and task priorities, so as to improve the system's processing ability in dealing with high-frequency dynamic problems and maximize production efficiency.

[0019] As Figure 1 shown, a method for detecting and scheduling an aviation equipment cluster in a high-frequency dynamic scenario according to an embodiment of the present invention includes the following steps:

[0020] Step 1: Construct an aviation equipment cluster detection scheduling model, analyze the associations and scheduling process requirements of aviation equipment cluster detection tasks and detection resources, and construct models for detection tasks, detection resources, and the scheduling process, including:

[0021] Step 1.1 During system initialization, obtain system scenario configuration information through connection with aviation equipment cluster resources and collaboration with relevant monitoring systems. Obtain the total number of aircraft in the aviation equipment cluster 、the aircraft set is 、the number of engines of aircraft (currently, the common engine configurations of aircraft are single-engine, twin-engine, and four-engine, and r represents the r-th aircraft), knowledge scope information ( represents fully reactive information, and information is only known after a task arrives; represents partially reactive information, and information for a period of time in the future at the current moment is known; represents complete information, and all task information is known), scheduling mechanism ( represents periodic rescheduling; represents event-triggered rescheduling; represents hybrid rescheduling; represents one-time scheduling), partial information range 、scheduling period 、number of trigger events 、predicted scheduling range 、scheduling task mode ( represents scheduling tasks that are known but not yet executed each time; represents scheduling tasks that are known but not yet scheduled each time). In addition, set some flag bits to judge system status selection for subsequent scheduling, such as the event-triggered rescheduling flag bit 、prediction selection flag bit 、one-time scheduling flag bit .

[0022] Step 1.2 Based on the aviation equipment cluster detection scenario model, construct a detection task model.

[0023] The total number of detection tasks in the aviation equipment cluster detection scheduling scenario is ,the detection task set is ,there are types of detection tasks, the number of tasks of type is ,the task type of detection task is ,each type of task requires different detection resources, The number of N types of detection resources required for the detection task is .

[0024] For each detection task , configure its attributes, such as the detection task 's arrival time is , the detection task 's execution duration is , the detection task 's deadline is , At time , the remaining execution time of the detection task . In addition, set some flag bits to judge the task status for subsequent scheduling, such as Flag At time whether the detection task has arrived, Flag At time whether the detection task will be completed on time, Flag At time whether the detection task is being executed, Flag At time whether the detection task has been completed, At time when the information of the detection task is known but not executed, schedule it. If , and , , then the task is to be scheduled, ; At time when the information of the detection task is known but not scheduled, schedule it. If , and , then the task is to be scheduled, .

[0025] Step 1.3: Based on the detection scenario model of the aviation equipment cluster, construct a detection resource model.

[0026] The total number of detection resources in the detection scheduling scenario of the aviation equipment cluster is , the detection resource set , there are types of detection resources, Type pieces of detection resources The type is detection resources The detection quality of , The set of available quantities of N detection resources at time , The detection resources at time battery level , and the set of power consumption speeds of each detection resource is , The total number of available detection resources at time , and the minimum available battery level of the detection resources is . In addition, set some flag bits to judge the resource status for subsequent scheduling. Flag The detection resources at time Whether it is currently executing a detection task , Flag The detection resources at time Whether it is idle, Flag The detection resources at time Whether it is faulty. At time, if , and , then the detection resources are available.

[0027] Step 1.4 Construct the scheduling process parameters, including the optimization object, objective function, and constraint conditions.

[0028] The optimization object is usually set to the start execution time of the detection task , and generate a detection scheduling table by solving the optimal start time for each detection task.

[0029] The objective function usually includes the total waiting duration of the detection tasks, the total delay duration of the detection tasks, the total quality of the detection tasks, the total cost of the detection tasks, etc.

[0030] The constraint conditions usually include that the start time of the detection task is later than the arrival time (as shown in Equation (1)), the number of occupied detection resources is less than or equal to the total number of detection resources (as shown in Equation (2)), the battery level of the detection resources is less than or equal to It cannot be called when [condition] (as shown in Equation (3)), the remaining power of the detection resource should meet the requirements for task execution completion (as shown in Equation (4)), the waiting time cannot exceed the maximum waiting time limit (as shown in Equation (5)), the delay time cannot exceed the maximum delay time limit (as shown in Equation (6)), the task quality cannot be less than the minimum task quality limit (as shown in Equation (7)), the task cost cannot exceed the maximum task cost limit (as shown in Equation (8)), the task load cannot exceed the maximum task load limit (as shown in Equation (9)), etc. By selecting one or more of the above objectives as the objective function and based on the existing constraints, the optimization object that maximizes the objective function can be solved. Among them, is the maximum waiting time, is the maximum delay time, is the minimum task quality, is the maximum task cost, is the maximum load.

[0031] (1)

[0032] (2)

[0033] (3)

[0034] (4)

[0035] (5)

[0036] (6)

[0037] (7)

[0038] (8)

[0039] (9)

[0040] Step 2: Sense the task status based on multi-time-scale data fusion: Detect the task and detection resource status through the sensor perception system, fuse task data at different time scales to predict the task arrival trend in the future for a period of time, so as to achieve real-time monitoring and advanced prediction of the detection task, including:

[0041] Step 2.1. Monitor the task status.

[0042] Deploy various types of sensors in key parts of the aviation equipment cluster (such as parts that seriously affect the health status of the aircraft, like engines and wings) and related operation scenarios (such as aircraft maintenance scenarios during downtime) to grasp the real-time operation status of the detection tasks. Position sensors track the precise geographical location, flight altitude, and heading information of each aircraft in the aviation equipment cluster. These data can be used to judge the progress of task execution. Condition sensors detect the operating details of the equipment, such as key indicators like engine speed, oil temperature, and oil pressure, and can promptly detect potential equipment failure risks. Environmental sensors continuously monitor temperature, humidity, air pressure, and electromagnetic interference intensity. Changes in the external environment may interfere with the detection accuracy. Mastering environmental information in advance facilitates timely adjustment of detection strategies. Various sensors transmit the collected data to the data processing center at a set frequency to build an initial data pool for task status monitoring. In a complex operating environment, the data collected by sensors is extremely likely to be mixed with noise and interference signals, resulting in data deviation or even errors. Filtering techniques need to be used to remove interference. The common Kalman filtering algorithm, based on the state transition model and measurement model of the system, continuously corrects the measured values, removes high-frequency noise and random interference, and realizes data restoration. For outliers caused by occasional sensor failures or data missing problems caused by temporary signal loss, data cleaning is used. Data cleaning identifies and removes outliers that significantly deviate from the normal range, and then fills in the missing data through methods such as interpolation and mean filling. Finally, accurate task status monitoring information is output.

[0043] Step 2.2. Use annual data to reflect the long-term trend of the aviation equipment cluster detection tasks, use quarterly data to highlight the seasonal patterns of task arrivals, and use daily data to reflect the occurrence patterns of unforeseen tasks such as the issuance of emergency detection instructions. Integrate the data at these three time scales using methods such as weighted average and feature splicing. Based on the integrated data, select an appropriate prediction model to predict the task arrival trend.

[0044] The prediction model can be selected from the following two or other prediction models. The time series analysis model, such as the ARIMA (Autoregressive Integrated Moving Average) model, by analyzing the autocorrelation and lag of historical data, explores the internal laws of the data and predicts the time points and quantities of task arrivals in the short term in the future. The LSTM (Long Short-Term Memory) model in the field of machine learning, with its unique memory cell structure, is good at processing long sequence data and capturing non-linear relationships in time series, and accurately estimates the task trend. The prediction results need to be updated in real time. As new tasks arrive continuously and the actual situation changes, collect feedback information in real time, dynamically adjust the model parameters and weights, and continuously optimize the prediction accuracy, so that the task arrival prediction can closely fit the high-frequency dynamic aviation equipment cluster detection scenario.

[0045] Step 3: Evaluate the resource status constructed based on the multi-dimensional index system: Establish a multi-dimensional resource status evaluation index system, which includes multiple aspects such as resource availability, busyness, and execution ability. The real-time status of detection resources in the aviation equipment cluster is evaluated by integrating multiple indicators, including:

[0046] Step 3.1. Resource availability aims to intuitively reflect the remaining resources that can be allocated and immediately put into the detection task, and it measures whether the resources can flexibly respond to new or urgent detection tasks. The core of constructing the resource availability index lies in constructing the resource idle rate, resource allocation delay, and total available resources and comprehensively considering the influence of the three. Resource idle rate As shown in Equation (10), it represents the proportion of system resources that are idle within a certain period of time. The higher the idle rate, the stronger the availability of the resources, and the more quickly they can be allocated to new tasks. Resource allocation delay As shown in Equation (11), it represents the time delay between the task arrival time and the actual resource allocation time. The shorter the delay, the higher the availability of the resources, where represents the th task's resource allocation time. Total available resources As shown in Equation (12), it represents the number of resources in the system that are not occupied and have no faults, and measures the sufficiency of the resources. Resource availability Is the weighted sum of the resource idle rate, resource allocation delay, and total available resources, as shown in Equation (13), where , , Are the weight coefficients.

[0047] (10)

[0048] (11)

[0049] (12)

[0050] (13)

[0051] Step 3.2. Resource busyness reflects the workload of the resources during operation and embodies the current usage intensity of the system resources. This indicator mainly focuses on the load situation of the resources, resource utilization rate, task queue length, and task execution timeout rate. Resource utilization rate As shown in Equation (14), it represents the ratio of resources occupied within a certain period of time. The higher the resource utilization rate, the higher the resource busyness and the heavier the system load. Task queue length As shown in Equation (15), it represents the number of tasks to be processed. The longer the queue length, the higher the current resource busyness, and there may be a phenomenon of task backlog. Task execution timeout rate As shown in Equation (16), it reflects the proportion of tasks that fail to be completed within the scheduled time. A higher timeout rate may mean that the resources are too busy, affecting the efficiency and quality of task execution. Resource busyness Is the weighted sum of resource utilization rate, task queue length, and task execution timeout rate, as shown in Equation (17), where 、 、 Are weight coefficients.

[0052] (14)

[0053] (15)

[0054] (16)

[0055] (17)

[0056] Step 3.3. The resource execution ability measures the efficiency and quality of the system resources to complete tasks and reflects the actual performance of the resources when executing tasks. This indicator mainly involves task completion rate, task execution time, and resource failure rate. Task completion rate As shown in Equation (18), it represents the ratio of resources successfully executing and completing tasks. A high task completion rate indicates strong resource execution ability and can efficiently process tasks. Task average execution time As shown in Equation (19), it refers to the time required for resources to complete tasks. The shorter the execution time, the higher the execution ability of the resources and the better the system operation efficiency. Resource failure rate As shown in Equation (20), it represents the proportion of resources that fail or become unavailable during execution. A low failure rate indicates strong reliability and execution ability of the resources. Resource execution ability Is the weighted sum of task completion rate, task execution time, and resource failure rate, as shown in Equation (21), where 、 、 Are weight coefficients.

[0057] (18)

[0058] (19)

[0059] (20)

[0060] (21)

[0061] Step 4: Construct a priority division mechanism based on multi-feature fusion of tasks: Construct task urgency and task importance indicators. These task urgency and task importance indicators fuse feature parameters such as task execution time, task benefit, task level, and task relevance, and comprehensively evaluate task urgency and task importance to divide task priorities, including:

[0062] Step 4.1. Task urgency reflects the degree of urgency of the task in the time dimension, measuring whether the task can be completed on time to avoid adverse effects caused by delays. It mainly considers the remaining executable time of the task and the penalty cost brought by the delay. The deadline for task completion is , then the remaining executable time of the task . Obviously, The smaller it is, the closer the task is to the deadline and the higher the urgency. Due to the differences in the nature of different tasks, the consequences of delays are very different. Therefore, a delay penalty coefficient is introduced to quantify the delay loss. For emergency repair and detection tasks related to flight safety, delays may lead to serious consequences such as the grounding of the aviation equipment cluster and safety accidents, and the value of is relatively high; while for general equipment daily inspections, is relatively small. The task urgency indicator

[0063] is shown in Equation (22):

[0064] Step 4.2. Task importance focuses on measuring task benefits, the degree of association with subsequent tasks, and task levels, reflecting the contribution of the task to the achievement of the overall goal. Task benefit is a key parameter that intuitively reflects the value of the task. High-benefit tasks often carry more expectations and should be given higher priorities. The task levels are divided according to factors such as task nature, technical difficulty involved, and safety risks . Tasks with high execution difficulty require more resources and should be considered later in resource allocation. Task relevance reflects the closeness of the task to subsequent tasks. If the result of a detection task directly affects the combat deployment and flight plan adjustment of the subsequent aviation equipment cluster, its relevance is high; conversely, tasks with weak relevance have less impact on the overall operation even if they are delayed. Considering the above factors comprehensively, the task importance indicator is shown in Equation (23):

[0065] (23);

[0066] Among them, , , are weight coefficients.

[0067] Step 4.3. Based on the task urgency index and task importance index, collaboratively determine the task priority. The task priority is as shown in Equation (24):

[0068] (24)

[0069] where and are the weight coefficients of the urgency index and importance index respectively. In the high-frequency dynamic detection scenario of aviation equipment clusters, if it is currently in an emergency repair period, can be appropriately increased to ensure that emergency tasks can quickly obtain resources; if long-term strategic development is emphasized, the weight can be increased to prioritize high-correlation tasks. All tasks are sorted according to the magnitude of . High-priority tasks are preferentially allocated high-quality detection resources and executed quickly; low-priority tasks are flexibly arranged when resources are idle and abundant, realizing the refined allocation of resources and maximizing the benefits of task execution.

[0070] Step 5: Generation of scheduling plan: Based on the task status, resource status, and task priority division results obtained in the above steps, an intelligent optimization algorithm is used to generate an aviation equipment cluster detection scheduling strategy adapted to high-frequency dynamic scenarios to achieve the optimal matching and efficient scheduling of tasks and resources, including:

[0071] The setting of the objective function is usually divided into time-optimal orientation, maximizing resource utilization rate, and maximizing task benefits. The objective function of time-optimal orientation is often set to minimize the total task completion time , minimize the total task waiting time , and minimize the total task delay time , as shown in Equations (25), (26), and (27). In scenarios such as emergency rescue and military operations, it is crucial to complete the detection quickly. Shortening the total duration can timely obtain key information and gain time for subsequent decision-making. Maximizing resource utilization rate focuses on making various detection resources operate at full load as much as possible and reducing idle waste. The resource utilization rate is as shown in Equation (14). Maximizing task benefits combines task importance and benefit situations and is applicable to the detection of commercial aviation equipment clusters. High-benefit detection tasks are preferentially completed to maximize profits. The benefit maximization objective is as shown in Equation (28). In actual application, the above factors are often comprehensively weighed to adapt to different scenario requirements. The comprehensive objective function is as shown in Equation (29), where , , , , is the weight coefficient. Constraint configuration usually includes resource capacity constraints, task execution order constraints, and time constraints. The quantity of each type of resource is limited. For equipment - type resources, the amount of resources allocated to a task cannot exceed the total amount of resources. Some tasks have a sequential logical relationship, that is, if the pre - task is not completed, the subsequent task cannot be started, and the execution order needs to be restricted by constraint conditions. Tasks can only be scheduled and executed after they arrive, and some tasks need to be executed within a specific time, so time constraint conditions are required to limit this.

[0072] (25)

[0073] (26)

[0074] (27)

[0075] (28)

[0076] (29)

[0077] Step 5.2 Predictive task scheduling considering priorities. In terms of resource allocation, the requirements of high - priority tasks are satisfied first, and rapid allocation is carried out according to task urgency and resource availability. Based on task arrival trend prediction, resource status assessment, and task priority division, an intelligent optimization algorithm is used to first solve the scheduling scheme for high - priority tasks. Once resource conflicts occur for high - priority tasks, the algorithm is used for dynamic adjustment to allocate key resources. When resources are sufficient, the scheduling of low - priority tasks is started. The algorithm design takes into account both flexibility and efficiency, and also uses intelligent algorithms for optimization, but the parameter adjustment focuses on the fragmented utilization of resources. A dynamic monitoring mechanism is established to track the resource release situation in real - time, improve the overall resource utilization efficiency, and achieve a deep and dynamic matching between tasks and resources.

[0078] The aviation equipment cluster detection and scheduling method for high - frequency dynamic scenarios disclosed by the present invention can analyze the aviation equipment cluster detection tasks, resource associations, and scheduling requirements to build a model, perceive and predict the task status in real - time, comprehensively evaluate the resource status, realize task priority division, and generate an efficient scheduling scheme based on task status, resource status, and task priority.

[0079] The content not described in detail in the specification of the present invention belongs to the prior art well - known to those skilled in the art.

[0080] The above - mentioned are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for detecting and scheduling an aviation equipment cluster in a high-frequency dynamic scenario, characterized in that The steps are as follows: Step 1: Analyze the difficulties faced by the detection of aviation equipment clusters in high-frequency dynamic scenarios, extract the associations between tasks and resources in the aviation equipment cluster detection scenario, and construct an aviation equipment cluster detection scheduling model from the scenario model, detection task model, detection resource model, and scheduling process model, laying a foundation for subsequent indicator establishment and scheduling solution; Step 2: Based on the aviation equipment cluster detection scheduling model, collect the operation parameters of detection tasks in real time through sensors, perform data preprocessing through filtering and data cleaning technologies to obtain accurate task status monitoring information; and based on the currently monitored task arrival data and historical task data, perform task data fusion on the time scales of annual, quarterly, and daily, and predict task arrivals based on the fused data to achieve task status perception; Step 3: Based on the aviation equipment cluster detection scheduling model, construct indicators for resource availability, resource busyness, and resource execution ability, evaluate the current resource status from multiple dimensions, and comprehensively evaluate the overall resource status based on the three indicators to quantitatively evaluate the system resources and provide a basis for the execution of subsequent scheduling optimization; Step 4: Based on the aviation equipment cluster detection scheduling model, construct indicators for task urgency and task importance. The task urgency and task importance indicators integrate task execution time, task revenue, task level, and task correlation degree, comprehensively evaluate the task urgency and task importance, divide priorities for tasks, and tasks with high priorities obtain high-quality detection resources first, while low-priority tasks are flexibly arranged when resources are abundant to maximize the task execution revenue in high-frequency dynamic scenarios; Step 5: Based on the perceived task status, resource status, and task priority division results, give priority to high-priority detection tasks, use intelligent optimization algorithms to generate high-priority aviation equipment cluster detection scheduling plans. When resources are insufficient, low-priority tasks are in a waiting state. If resources are sufficient, they are provided for low-priority tasks to execute, and intelligent optimization algorithms are used to generate low-priority aviation equipment cluster detection scheduling plans to achieve optimal matching and efficient scheduling of tasks and resources.

2. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 1, wherein The said Step 1 includes: Step 1.

1. Construct an aviation equipment cluster detection scenario model, which includes the total number of aircraft, the number of engines of the aircraft, knowledge scope information, scheduling mechanism, predicted scheduling scope, and scheduling task mode.

3. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 2, wherein The said Step 1 also includes: Step 1.

2. Construct a detection task model, which contains the total number of scheduling scenario tasks, the number of detection task types, the number of each type of task, the task type of each detection task, the arrival time of the detection task, the execution duration of the detection task, the deadline of the detection task, the number of detection resources required for each type of detection task, the remaining executable time of the detection task, and the detection task status.

4. The method for detecting and scheduling an aviation equipment cluster facing high-frequency dynamic scenarios according to claim 3, wherein The said Step 1 also includes: Step 1.

3. Construct a detection resource model, which contains the total number of detection resources, the number of detection resource types, the number of each type of detection resource, each detection resource type, the detection quality of the detection resource, the power of the detection resource, the occupied state of the detection resource, the fault state of the detection resource, and the power consumption speed of the detection resource; Step 1.

4. Describe the matching and allocation rules between tasks and resources during the scheduling process. Consider the status of task resources, establish an objective function, which includes the minimum total waiting time, the minimum total delay time, the maximum inspection task quality, and the minimum inspection task cost. Set constraint conditions according to the actual situation, including resource quantity constraints, arrival time constraints, and resource power constraints. Set the optimization object to perform optimization solving on the optimization object based on the objective function and constraint conditions subsequently, thereby constructing a scheduling process model.

5. The method for detecting and scheduling an aircraft equipment cluster for high-frequency dynamic scenarios according to claim 1, wherein The said Step 2 includes: Step 2.

1. During the detection process of the aviation equipment cluster, deploy multiple sensors to monitor the task status. Step 2.

2. Through weighted average and feature splicing, fuse the annual data reflecting the long-term trend of the aviation equipment cluster tasks, the quarterly data showing seasonal patterns, and the daily data revealing the occurrence patterns of emergency tasks, and then use a prediction model to predict the task arrival trend.

6. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 1, wherein The said Step 3 includes: Step 3.

1. Comprehensively consider the resource idle rate, resource allocation delay, and total available resources to construct an availability index of resources. The resource idle rate represents the proportion of system resources that are idle within a certain time period. The resource allocation delay represents the time delay between the task arrival time and the actual resource allocation time. The total available resources represent the number of resources in the system that have not been occupied and are free of faults.

7. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 6, wherein The said Step 3 also includes: Step 3.

2. Comprehensively consider the resource utilization rate, task queue length, and task execution timeout rate to construct an index of resource busyness. The resource utilization rate represents the ratio of resources occupied within a certain period of time. The task queue length represents the number of tasks to be processed. The task execution timeout rate reflects the proportion of tasks that fail to be completed within the scheduled time. Step 3.

3. Comprehensively consider the task completion rate, task execution time, and resource failure rate to construct an index of resource execution ability. The task completion rate represents the ratio of resources that successfully execute and complete tasks. The average task execution time refers to the time required for resources to complete tasks. The resource failure rate represents the proportion of resources that fail or become unavailable during the execution process.

8. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 1, wherein The said Step 4 includes: Step 4.

1. Construct a task urgency index. By comprehensively considering time urgency and delay severity, ensure that emergency tasks have a higher priority in the priority ranking. When multiple tasks compete for resources, the task with a higher urgency index obtains resources first, avoiding major losses caused by time delays.

9. The method for detecting and scheduling an aviation equipment cluster for high-frequency dynamic scenarios according to claim 8, wherein The said Step 4 also includes: Step 4.

2. Construct a task importance index to reflect the contribution of the task to the achievement of the overall goal. Step 4.

3. Based on the task urgency index and task importance index, finally determine the task priority.

10. The method for detecting and scheduling an aviation equipment cluster facing high-frequency dynamic scenarios according to claim 1, wherein The said Step 5 includes: Step 5.

1. Configure the objective function and constraint conditions. The setting of the objective function is usually divided into time-optimal orientation, maximizing resource utilization rate, and maximizing task revenue. The constraint conditions configure resource capacity constraints, task execution order constraints, and time constraints. Step 5.

2. In terms of resource allocation, prioritize meeting the requirements of high-priority tasks and rapidly allocate resources according to task urgency and resource availability; based on task arrival trend prediction, resource status assessment, and task priority division, adopt an intelligent optimization algorithm to preferentially solve the high-priority task scheduling scheme, and start the low-priority task scheduling when resources are sufficient; establish a dynamic monitoring mechanism to track the resource release situation in real time, improve the overall resource utilization efficiency, and achieve deep and dynamic matching between tasks and resources.

Images