An aircraft powerplant maintenance method, system, apparatus, and storage medium

Abstract

The application discloses an aviation power device maintenance method, system, equipment and storage medium, and relates to the technical field of aviation. The method comprises the following steps: constructing a maintenance plan model based on device maintenance data; decomposing the maintenance plan model into interrelated maintenance time sequence sub-models, device allocation sub-models and maintenance resource scheduling sub-models according to maintenance decision types; solving based on the sub-models and the correlation between the sub-models to generate a device maintenance plan and execute the device maintenance plan. The method reduces the solving complexity of a high-dimensional complex model, enables a feasible solution of the maintenance plan of a large-scale aviation power device fleet to be obtained within an acceptable time, ensures the coordination of three decision dimensions, namely, maintenance time arrangement, device allocation and maintenance resource scheduling, and avoids conflicts in the scheme. The method realizes the automatic generation of the aviation power device maintenance plan, reduces the dependence on individual experience, and improves the repeatability and standardization of plan making.

Description

Technical Field

[0001] This invention relates to the field of aviation technology, and more specifically to a method, system, equipment, and storage medium for maintaining aircraft power units. Background Technology

[0002] Currently, the development of maintenance plans for aircraft engines generally relies on manual methods. The routine process is as follows: planning personnel collect core business data such as engine status data, operating data, and contract terms, and then engineers analyze and judge the data based on their personal experience to finally develop a maintenance plan that can meet the basic operational needs of the fleet.

[0003] However, this manual-driven approach has at least the following obvious drawbacks: high safety risks, as manual scheduling is prone to violations of compliance deadlines due to subjective negligence, or may lead to insufficient fleet capacity, thereby affecting aviation operational safety; poor economic efficiency, as manual solutions usually only meet feasible requirements and are difficult to achieve the best cost for the entire repair process, resulting in redundant operating costs; and low efficiency, as each plan takes a long time to formulate and cannot quickly respond to dynamic changes in aviation operations. Summary of the Invention

[0004] The main objective of this invention is to provide a method, system, equipment, and storage medium for the maintenance of aircraft power units. By constructing a maintenance planning model aimed at minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization, it achieves a balanced optimization of maintenance plans across the dimensions of economy, safety, and efficiency. This avoids resource waste caused by insufficient backup equipment or excessive redundancy due to solely pursuing the lowest cost. By decomposing the maintenance planning model into three interconnected sub-models—maintenance timing, equipment allocation, and maintenance resource scheduling—the complexity of solving high-dimensional complex models is reduced, enabling feasible solutions for the maintenance planning of large-scale aircraft power unit fleets to be obtained within an acceptable timeframe. By solving and executing the maintenance plan based on each sub-model and their interrelationships, it ensures coordination among the three decision dimensions of maintenance scheduling, equipment allocation, and maintenance resource scheduling, avoiding internal conflicts in the plans. It also achieves automated generation of aircraft power unit maintenance plans, reducing reliance on individual experience and improving the repeatability and standardization of plan formulation.

[0005] To achieve the above objectives, the embodiments of this application provide the following technical solutions: According to a first aspect of the embodiments of this application, a method for maintaining an aircraft power plant is provided, comprising: Acquire equipment maintenance data for aircraft power plants; Based on the device maintenance data, a maintenance plan model is constructed, with the optimization objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. The maintenance planning model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types; The solution is obtained based on each sub-model and the relationships between them, and a device maintenance plan is generated and executed.

[0006] In some feasible implementations, the device maintenance data includes device update data; the device update data includes the latest status data and current operational requirements data of the aircraft power unit; after generating and executing the device maintenance plan, it further includes: The entire maintenance plan cycle is divided into several consecutive rolling optimization phases, each of which includes a decision-making period and a forecasting period. During the decision-making process of the current rolling optimization phase, device update data is continuously collected, and the device maintenance data is updated based on the device update data. Based on the updated equipment maintenance data, the maintenance timing sub-model, equipment allocation sub-model, and maintenance resource scheduling sub-model are solved to generate and execute the equipment maintenance plan for the next rolling optimization decision period.

[0007] In some feasible implementations, during the decision-making phase of the current rolling optimization stage, device update data is continuously collected, and the device maintenance data is updated based on the device update data, including: Before generating the device maintenance plan for the next rolling optimization phase decision period, obtain the actual device status data and actual resource usage data at the end of the current rolling optimization phase decision period; The actual device status data and actual resource occupancy data are updated in the device maintenance data and used as input parameters for solving the maintenance timing sub-model, device allocation sub-model, and maintenance resource scheduling sub-model.

[0008] In some feasible implementations, the device maintenance data includes fixed device data, which includes maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data; the process of constructing a maintenance plan model based on the device maintenance data includes: Based on the preset maintenance business requirements, the objectives of minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization are determined. Constraints are determined based on the maintenance cost accounting data, available equipment quantity data, and maintenance resource occupancy data. These constraints include cost constraints, redundancy constraints, and resource constraints. The comprehensive optimization objective is obtained by using the weighting coefficient method based on the objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. Based on the device maintenance data, the comprehensive optimization objective, and the constraints, the maintenance plan model is constructed.

[0009] In some feasible implementations, the device maintenance data further includes device fluctuation data, which includes historical maintenance cycle fluctuation data and historical flight volume fluctuation data; the determination of constraints further includes: Based on the historical maintenance cycle fluctuation data and the historical flight volume fluctuation data, the maintenance cycle fluctuation coefficient and the flight volume fluctuation coefficient are determined respectively. The flight volume fluctuation coefficient is introduced into the redundancy constraint, and the maintenance cycle fluctuation coefficient is introduced into the resource constraint to determine the constraints required for the maintenance plan model.

[0010] In some feasible implementations, the maintenance planning model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types, including: According to the maintenance decision type, maintenance timing decision variables and constraints, equipment allocation decision variables and constraints, and maintenance resource decision variables and constraints are extracted from the maintenance plan model. Based on the maintenance time series decision variables and constraints, the maintenance time series sub-model is constructed, which is used to determine the maintenance time of each aero-engine unit. Based on the device allocation decision variables and constraints, a device allocation sub-model is constructed. The device allocation sub-model is used to determine the external resource allocation required to fill maintenance gaps. Based on the maintenance resource decision variables and constraints, and the output of the maintenance timing sub-model, the maintenance resource scheduling sub-model is constructed. The maintenance resource scheduling sub-model is used to determine the scheduling and resource allocation of maintenance operations. The maintenance time period and resource requirement information output by the maintenance timing sub-model are used as input parameters for the device allocation sub-model and the maintenance resource scheduling sub-model.

[0011] In some feasible implementations, the step of solving based on each sub-model and the relationships between them to generate a device maintenance plan includes: The maintenance time sequence sub-model is solved based on a heuristic algorithm to obtain multiple candidate maintenance time sequence schemes. Each candidate maintenance time sequence scheme includes a planned maintenance time period and corresponding resource requirement information. For each candidate maintenance timing scheme, the corresponding planned maintenance time period and resource requirement information are respectively input into the maintenance resource scheduling sub-model and the device allocation sub-model; If both the maintenance resource scheduling sub-model and the device allocation sub-model can obtain feasible solutions, then the candidate maintenance timing scheme is retained. Starting with each retained candidate maintenance sequence scheme, check whether the generated candidate solution satisfies the parameter linkage relationship between the maintenance sequence sub-model, the equipment allocation sub-model, and the maintenance resource scheduling sub-model. If not, terminate the search for the branch where the candidate solution is located. Search the current branch node based on the branch and bound method, and call the heuristic algorithm to perform local optimization on the current branch node to update the feasible solution and the boundary of the objective function. From all candidate solutions obtained after searching and optimization that satisfy all parameter linkage relationships, the solution that makes the overall optimization objective of the maintenance plan model optimal is selected, and combined with the corresponding maintenance resource scheduling scheme and equipment allocation scheme to generate the equipment maintenance plan.

[0012] According to a second aspect of the embodiments of this application, an aircraft power plant maintenance system is provided, the system comprising: The data acquisition module is used to acquire equipment maintenance data of the aircraft power unit; The model building module is used to build a maintenance plan model based on the device maintenance data. The maintenance plan model has the optimization objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. The model decomposition module is used to decompose the maintenance plan model into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to the maintenance decision type. The plan generation module is used to solve the problem based on each sub-model and the relationships between them, generate a device maintenance plan, and execute it.

[0013] According to a third aspect of the present application, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect above.

[0014] According to a fourth aspect of the embodiments of this application, a computer-readable storage medium is provided having computer-readable instructions stored thereon, which can be executed by a processor to implement the method described in the first aspect above.

[0015] In summary, the embodiments of this application provide a method, system, equipment, and storage medium for maintaining an aircraft power plant. The method involves acquiring maintenance data of the aircraft power plant; constructing a maintenance plan model based on the maintenance data, with the optimization objectives of minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization; decomposing the maintenance plan model into interrelated maintenance sequence sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types; and solving the problems based on each sub-model and the relationships between them to generate and execute an equipment maintenance plan. By constructing a maintenance planning model aimed at minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization, a balanced optimization of maintenance plans is achieved across the three dimensions of economy, safety, and efficiency. This avoids resource waste caused by insufficient backup equipment or excessive redundancy due to solely pursuing the lowest cost. By decomposing the maintenance planning model into three interrelated sub-models—maintenance timing, equipment allocation, and maintenance resource scheduling—the solution complexity of high-dimensional complex models is reduced, enabling feasible solutions to be obtained within an acceptable timeframe for the formulation of maintenance plans for large-scale aero-engine fleets. By solving and executing maintenance plans based on each sub-model and their interrelationships, the coordination among the three decision dimensions of maintenance scheduling, equipment allocation, and maintenance resource scheduling is ensured, avoiding internal conflicts in the plans. The automated generation of aero-engine maintenance plans is achieved, reducing reliance on individual experience and improving the repeatability and standardization of plan formulation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0017] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0018] Figure 1 This is a schematic diagram of the aircraft power unit maintenance method provided in the embodiments of this application; Figure 2 A schematic diagram illustrating the overall process of developing an aircraft power unit repair plan for an embodiment of this application; Figure 3 This is a schematic diagram of the sub-model-level collaborative solution process provided in the embodiments of this application; Figure 4 This is a schematic diagram of aircraft power plant maintenance provided in an embodiment of this application; Figure 5 This paper shows a structural diagram of an electronic device provided in an embodiment of this application; Figure 6 A diagram of a computer-readable storage medium provided in an embodiment of this application is shown.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0022] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0023] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0025] Figure 1 This application illustrates a method for maintaining an aircraft power unit, comprising: Step S101: Obtain equipment maintenance data for the aircraft power unit; Step S102: Based on the device maintenance data, construct a maintenance plan model, wherein the maintenance plan model has the optimization objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. Step S103: Decompose the maintenance plan model into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to the maintenance decision type; Step S104: Solve based on each sub-model and the relationships between them, generate a device maintenance plan and execute it.

[0026] This application provides a method for maintaining an aircraft power unit. By acquiring equipment maintenance data of the aircraft power unit, a data foundation is provided for subsequent model construction. The equipment maintenance data includes fixed equipment data, updated equipment data, and fluctuating equipment data. The fixed equipment data includes maintenance cost accounting data, available equipment quantity data, and maintenance resource occupancy data. The updated equipment data includes the latest status data and current operating demand data of the aircraft power unit. The fluctuating equipment data includes historical maintenance cycle fluctuation data and historical flight volume fluctuation data.

[0027] Based on the device maintenance data, a maintenance plan model is constructed. This model aims to minimize maintenance costs, rationalize device redundancy, and maximize maintenance resource utilization. The three optimization objectives are integrated into a comprehensive optimization objective using a weighted coefficient method. Corresponding cost constraints, redundancy constraints, and resource constraints are set. At the same time, flight volume fluctuation coefficients and maintenance cycle fluctuation coefficients are introduced into the redundancy and resource constraints to construct robust constraint conditions that adapt to uncertain scenarios, forming a complete constraint condition system. This ensures that the economic efficiency of the solution is guaranteed while taking into account safety and efficiency, avoiding the waste of resources caused by insufficient backup devices or excessive redundancy due to the pursuit of the lowest cost.

[0028] The maintenance planning model is decomposed into three interrelated sub-models based on maintenance decision type: maintenance timing sub-model, equipment allocation sub-model, and maintenance resource scheduling sub-model. The maintenance timing sub-model uses maintenance time arrangement as the decision variable, the equipment allocation sub-model uses the quantity of external resource allocation as the decision variable, and the maintenance resource scheduling sub-model uses maintenance task allocation as the decision variable. The maintenance time period and resource demand information output by the maintenance timing sub-model are used as input parameters for the equipment allocation sub-model and the maintenance resource scheduling sub-model. By using the model decomposition strategy, the solution complexity of high-dimensional complex problems is reduced, enabling the maintenance planning of a large-scale aero-engine fleet to obtain a feasible solution within an acceptable time.

[0029] Based on the collaborative solution of each sub-model and its relationships, the maintenance timing sub-model is first solved using a heuristic algorithm to obtain candidate maintenance timing schemes. The candidate schemes are then input into the device allocation sub-model and the maintenance resource scheduling sub-model for feasibility testing. Candidate schemes that can obtain feasible solutions are retained. Starting from the retained candidate schemes, iterative testing is performed to check whether they meet the linkage constraints between sub-models. For solutions that meet the linkage constraints, the branch and bound method and heuristic algorithm are used for collaborative optimization. The solution that optimizes the overall optimization objective is selected and combined with the corresponding maintenance resource scheduling scheme and device allocation scheme to generate a device maintenance plan. This ensures that the three decision dimensions of maintenance time arrangement, device allocation, and maintenance resource scheduling are coordinated with each other and avoids internal conflicts between schemes.

[0030] In one possible implementation, the device maintenance data includes device update data; the device update data includes the latest status data and current operational requirements data of the aero-engine; after generating and executing the device maintenance plan in step S104, the method further includes: dividing the entire maintenance plan cycle into several consecutive rolling optimization phases, each rolling optimization phase including a decision period and a prediction period; continuously collecting device update data during the execution of the decision period of the current rolling optimization phase, and updating the device maintenance data based on the device update data; and solving the maintenance sequence sub-model, device allocation sub-model, and maintenance resource scheduling sub-model based on the updated device maintenance data to generate and execute the device maintenance plan for the decision period of the next rolling optimization phase.

[0031] In one possible implementation, during the decision-making period of the current rolling optimization phase, device update data is continuously collected, and the device maintenance data is updated based on the device update data. This includes: before generating the device maintenance plan for the next rolling optimization phase decision-making period, obtaining the actual device status data and actual resource occupancy data at the end of the current rolling optimization phase decision-making period; updating the device maintenance data with the actual device status data and actual resource occupancy data as input parameters for solving the maintenance sequence sub-model, device allocation sub-model, and maintenance resource scheduling sub-model.

[0032] In one possible implementation, the decision period of the rolling optimization phase can be dynamically adjusted according to the fleet operating intensity and maintenance demand fluctuations. For example, when the fleet operating intensity is high and maintenance demand fluctuates greatly, the decision period can be shortened to improve the adaptability of the plan; when the fleet operating intensity is low and maintenance demand is stable, the decision period can be extended to reduce the cost of plan formulation.

[0033] The device maintenance data includes device update data, which includes the latest status data and current operational requirements data of the aircraft power unit. For example, by collecting information such as the number of engine operating cycles, on-wing position status, and changes in flight workload in real time, the dynamic operating status of the aircraft power unit can be reflected. After generating and executing the device maintenance plan in step S104, the entire maintenance plan cycle is divided into several consecutive rolling optimization stages. Each rolling optimization stage includes a decision period and a prediction period. The decision period refers to the planning period for determining execution, and the prediction period refers to the planning period following the decision period to provide reference information. The planning length of the prediction period is not less than the planning length of the decision period.

[0034] During the decision-making phase of the current rolling optimization stage, continuously collect device update data and update the device maintenance data based on this data, achieving dynamic refresh of the model input data. Based on the updated device maintenance data, re-solve the maintenance sequence sub-model, device allocation sub-model, and maintenance resource scheduling sub-model to generate and execute the device maintenance plan for the next rolling optimization decision-making phase. Parameters such as the remaining resources and engine status from the previous phase are used as inputs to the model for the next phase, ensuring continuity and conflict-free maintenance plans, device allocation plans, and maintenance resource scheduling plans between phases. Through this rolling optimization mechanism, when flight structure, capacity configuration, or maintenance resources are adjusted, there is no need to rearrange the original plan; only calibration and solution are required for the next rolling phase, achieving dynamic iterative updates of the maintenance plan to adapt to dynamic changes in business scenarios.

[0035] In one possible implementation, the device maintenance data includes fixed device data, which includes maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data. In step S102, the process of constructing a maintenance plan model based on the device maintenance data includes: determining a maintenance cost minimization target, a device redundancy rate rationalization target, and a maintenance resource utilization maximization target based on preset maintenance business requirements; determining constraints based on the maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data, including cost constraints, redundancy constraints, and resource constraints; obtaining a comprehensive optimization target using a weighted coefficient method based on the maintenance cost minimization target, device redundancy rate rationalization target, and maintenance resource utilization maximization target; and constructing the maintenance plan model based on the device maintenance data, the comprehensive optimization target, and the constraints.

[0036] In one possible implementation, the device maintenance data further includes device fluctuation data, which includes historical maintenance cycle fluctuation data and historical flight volume fluctuation data; the determination of constraints further includes: determining maintenance cycle fluctuation coefficient and flight volume fluctuation coefficient based on the historical maintenance cycle fluctuation data and historical flight volume fluctuation data, respectively; introducing the flight volume fluctuation coefficient into the redundancy constraints and the maintenance cycle fluctuation coefficient into the resource constraints, so as to determine the constraints required for the maintenance plan model.

[0037] The maintenance plan model also includes compliance constraints, which are set based on aviation power plant maintenance compliance standards and fleet operation compliance requirements. These constraints are used to ensure that the generated equipment maintenance plan complies with industry compliance norms and avoids compliance risks.

[0038] The device maintenance data also includes fixed device data, which includes maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data. The maintenance cost accounting data represents the cost information of each stage of the entire repair process; the available device quantity data represents the configuration status of backup power units and on-wing units; and the maintenance resource occupancy data represents the capacity and workload of the repair shop. In step S102, the process of constructing a maintenance plan model based on the device maintenance data first determines the maintenance cost minimization target, the device redundancy rate rationalization target, and the maintenance resource utilization maximization target based on preset maintenance business requirements. The maintenance cost minimization target represents the lowest overall cost of the entire repair process; the device redundancy rate rationalization target represents that the ratio of the number of backup units to the required number is within a reasonable range to avoid excessive redundancy leading to cost waste or insufficient redundancy causing operational risks; and the maintenance resource utilization maximization target represents the optimal utilization of the repair shop's capacity to avoid resource idleness or overload.

[0039] Furthermore, constraints are determined based on the maintenance cost accounting data, available equipment quantity data, and maintenance resource occupancy data. These constraints include cost constraints, redundancy constraints, and resource constraints. Cost constraints limit the upper limit or budget range of maintenance costs; redundancy constraints limit the lower limit of the number of standby devices to ensure operational safety; and resource constraints limit the available capacity of maintenance resources to avoid overload operation. Further, a comprehensive optimization objective is obtained using a weighted coefficient method based on the objectives of minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization. This weighted coefficient method assigns weight coefficients to each objective, and the sum of all weight coefficients is 1, transforming multiple objectives into a solvable comprehensive optimization objective. The weight coefficients can be dynamically adjusted according to business needs to adapt to different scenarios prioritizing cost, safety, or efficiency. Based on the equipment maintenance data, the comprehensive optimization objective, and the constraints, the maintenance plan model is constructed.

[0040] In one possible implementation, the device maintenance data further includes device fluctuation data, which comprises historical maintenance cycle fluctuation data and historical flight volume fluctuation data. Statistical analysis of the historical data generates characteristic values ​​such as fluctuation ranges and probability distributions. Determining constraints also includes determining the maintenance cycle fluctuation coefficient and flight volume fluctuation coefficient respectively based on the historical maintenance cycle fluctuation data and historical flight volume fluctuation data using statistical analysis methods. The maintenance cycle fluctuation coefficient characterizes the deviation of actual maintenance time from standard maintenance time, and the flight volume fluctuation coefficient characterizes the deviation of actual flight workload from planned workload. The flight volume fluctuation coefficient is introduced into the redundancy constraints, and the maintenance cycle fluctuation coefficient is introduced into the resource constraints to determine the constraints required for the maintenance plan model. Specifically, in the redundancy constraints, the lower limit of the available device quantity is multiplied by the flight volume fluctuation coefficient to cope with sudden increases in workload; in the resource constraints, the maintenance time is multiplied by the maintenance cycle fluctuation coefficient to reserve buffer time for extended maintenance cycles. This constructs robust constraints containing uncertainty parameters, ensuring that the maintenance plan still meets compliance and safety requirements within the range of uncertain factors.

[0041] In one possible implementation, in step S103, the maintenance plan model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to the maintenance decision type. This includes: parsing maintenance timing-related decision variables and constraints, equipment allocation-related decision variables and constraints, and maintenance resource-related decision variables and constraints from the maintenance plan model according to the maintenance decision type; constructing the maintenance timing sub-model based on the maintenance timing-related decision variables and constraints, the maintenance timing sub-model being used to determine the maintenance time for each aero-engine unit; constructing the equipment allocation sub-model based on the equipment allocation-related decision variables and constraints, the equipment allocation sub-model being used to determine the external resource allocation required to fill maintenance gaps; and constructing the maintenance resource scheduling sub-model based on the maintenance resource-related decision variables and constraints and the output of the maintenance timing sub-model, the maintenance resource scheduling sub-model being used to determine the scheduling and resource allocation of maintenance operations; wherein the maintenance time period and resource demand information output by the maintenance timing sub-model are used as input parameters for the equipment allocation sub-model and the maintenance resource scheduling sub-model.

[0042] In step S103, the maintenance plan model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types. This decomposition strategy is based on business dimension decoupling to reduce the solution complexity of the high-dimensional overall model. According to the maintenance decision types, maintenance timing decision variables and constraints, equipment allocation decision variables and constraints, and maintenance resource decision variables and constraints are extracted from the maintenance plan model. The maintenance timing decision variables represent the maintenance schedule for each aircraft power unit, the equipment allocation decision variables represent the quantity decision for leasing or purchasing power units, and the maintenance resource decision variables represent the batch allocation of maintenance tasks and the selection of maintenance depots.

[0043] Based on the maintenance timing decision variables and constraints, the maintenance timing sub-model is constructed. The maintenance timing sub-model is used to determine the maintenance time of each aircraft power unit. Its constraints may include a single-cycle repair restriction, which indicates that each power unit is only repaired once within the maintenance cycle; a cycle number restriction, which indicates that the number of operating cycles when the power unit is repaired is within a preset range; and an available quantity restriction, which indicates that the number of available power units per month meets the flight mission requirements.

[0044] Based on the aforementioned equipment allocation decision variables and constraints, a sub-model for equipment allocation is constructed. This sub-model is used to determine the allocation of external resources needed to fill maintenance gaps, such as decisions on the number of leased power units and the number of purchased power units. Its constraints include a leasing ban that prohibits new leases within a specific time period, a lease-return consistency restriction that balances the number of leased-in and leased-out units, and a lease term restriction that requires lease returns to meet the minimum lease duration requirement.

[0045] Based on the aforementioned maintenance resource decision variables and constraints, and the output of the maintenance timing sub-model, a maintenance resource scheduling sub-model is constructed. This sub-model determines the scheduling of maintenance operations and resource allocation. Its constraints include maintenance plant capacity limits representing the monthly maintenance capacity ceiling, and maintenance batch limits representing the control of the number of power units sent for repair simultaneously. The maintenance time period and resource demand information output by the maintenance timing sub-model serve as input parameters to the equipment allocation sub-model and the maintenance resource scheduling sub-model, forming a first and second linkage condition between the sub-models. The first linkage condition indicates that the maintenance time period output by the maintenance timing sub-model serves as input to the maintenance resource scheduling sub-model to determine the detailed maintenance schedule. The second linkage condition indicates that the additional power unit demand output by the maintenance timing sub-model serves as input to the equipment allocation sub-model to determine the leasing or purchasing option. Through this decomposition and linkage mechanism, each sub-model can be solved independently and optimized collaboratively, achieving the generation of a comprehensive optimal maintenance plan under complex constraints.

[0046] In one possible implementation, in step S104, the step of solving based on each sub-model and the relationships between them to generate a device maintenance plan includes: solving the maintenance timing sub-model using a heuristic algorithm to obtain multiple candidate maintenance timing schemes, each candidate maintenance timing scheme including a planned maintenance time period and corresponding resource requirement information; for each candidate maintenance timing scheme, inputting the corresponding planned maintenance time period and resource requirement information into the maintenance resource scheduling sub-model and the device allocation sub-model respectively; if both the maintenance resource scheduling sub-model and the device allocation sub-model can obtain feasible solutions, then the candidate maintenance timing scheme is retained; and so on. Starting with the retained candidate maintenance sequence schemes, the generated candidate solutions are checked to see if they satisfy the parameter linkage relationship between the maintenance sequence sub-model, the equipment allocation sub-model, and the maintenance resource scheduling sub-model. If not, the search for the branch containing the candidate solution is terminated. The current branch node is searched based on the branch and bound method, and the heuristic algorithm is called to perform local optimization on the current branch node to update the feasible solution and the objective function boundary. From all candidate solutions obtained after search and optimization that satisfy all parameter linkage relationships, the solution that makes the comprehensive optimization objective of the maintenance plan model optimal is selected, and combined with the corresponding maintenance resource scheduling scheme and equipment allocation scheme to generate the equipment maintenance plan.

[0047] In step S104, the step of solving the sub-models and their interrelationships to generate a device maintenance plan includes solving the maintenance timing sub-model using a heuristic algorithm to obtain multiple candidate maintenance timing schemes. Each candidate maintenance timing scheme includes a planned maintenance time period and corresponding resource requirement information. This heuristic algorithm uses an adaptive local search strategy to quickly obtain an initial feasible solution, providing a starting point for subsequent accurate solutions. For each candidate maintenance timing scheme, the corresponding planned maintenance time period and resource requirement information are input into the maintenance resource scheduling sub-model and the device allocation sub-model, respectively. Cross-model collaborative verification is performed based on the linkage conditions between the sub-models to check the matching degree of the candidate scheme in terms of maintenance resource capacity and device allocation feasibility. If both the maintenance resource scheduling sub-model and the device allocation sub-model can obtain feasible solutions, the candidate maintenance timing scheme is retained, and infeasible schemes with conflicts are eliminated, thus narrowing the subsequent search space.

[0048] Starting with each retained candidate maintenance timing scheme, the generated candidate solution is iteratively checked whether it satisfies the parameter linkage relationship between the maintenance timing sub-model, the device allocation sub-model, and the maintenance resource scheduling sub-model. This parameter linkage relationship includes the consistency constraint of maintenance time period, the matching constraint of resource demand quantity, and the sequential constraint of timing logic. If the candidate solution violates any linkage relationship, the further search of the branch where the candidate solution is located is terminated, thereby realizing cross-model collaborative pruning to reduce invalid calculations.

[0049] The branch-and-bound algorithm is used to search the current branch node. This involves splitting integer decision variables to generate a sub-model branch tree, determining boundary values ​​through relaxation, and iterative pruning based on these boundary values, systematically exploring the solution space. During the branch-and-bound search, the heuristic algorithm is dynamically invoked to locally optimize the current branch node, supplementing new feasible solutions and updating the lower bound of the objective function, thus accelerating the convergence process. Furthermore, from all candidate solutions obtained after search and optimization that satisfy all parameter linkage relationships, the solution that optimizes the overall goal of the maintenance plan model is selected. This solution is then combined with the corresponding maintenance resource scheduling scheme and equipment allocation scheme to generate the equipment maintenance plan. The plan outputs core information such as maintenance time, maintenance sequence, equipment allocation quantity, and maintenance resource allocation for each power unit, achieving automated formulation of an economically optimal and reliable maintenance plan under complex constraints.

[0050] In one possible implementation, during the execution of the device maintenance plan, progress data and resource consumption data of maintenance operations are collected in real time to monitor whether the maintenance progress lags behind the planned time node and whether the resource consumption exceeds the budget threshold. If the above-mentioned abnormalities occur, each sub-model is automatically triggered to re-solve based on the latest collected data to generate a temporary adjustment plan to correct the maintenance sequence, allocate additional resources, or reallocate maintenance tasks, so as to ensure that the maintenance plan can still be smoothly promoted in case of emergencies.

[0051] The maintenance method for aircraft power units provided in the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0052] The aircraft power plant maintenance method provided in this application aims to minimize costs, rationalize redundancy, and maximize resource utilization. It integrates a rolling optimization mechanism, a model decomposition solution strategy, and a robust multi-objective modeling method. Through multi-module collaboration, it adapts to the actual business needs of aircraft power plant maintenance. The overall system architecture of this method consists of a robust multi-objective and constraint definition module, a data processing and update module, a high-dimensional model decomposition module, a rolling optimization scheduling module, a sub-model collaborative solution module, and a scheme verification and iteration module. These modules work together to achieve full-process optimization of the repair plan.

[0053] Figure 2 This paper illustrates the overall process for developing an aircraft power unit repair plan. Starting from the core business objectives of minimizing repair costs, rationalizing redundancy, and maximizing resource utilization, the process first enters the robust multi-objective model construction phase. This phase primarily clarifies multi-dimensional optimization objectives and various constraints, constructing the core framework of the model from three dimensions: cost, safety, and efficiency. This addresses the problem of traditional models having a single objective and lacking robustness. With minimizing the overall cost of the entire repair process as the core objective, while also considering the rationalization of the average monthly power unit redundancy and maximizing maintenance resource utilization, the multi-objectives are transformed into a solvable comprehensive objective function using the objective weight coefficient method. The weight coefficients can be dynamically adjusted according to business needs.

[0054] The comprehensive objective function can be specifically defined as follows: F=α*Cost+β*1 / Redundancy+γ*1 / Utilization Where α+β+γ=1, α, β, and γ are weighting coefficients oriented by business needs, Cost is the total cost of the repair process, Redundancy is the engine redundancy rate, and Utilization is the utilization rate of maintenance resources.

[0055] To more precisely guide the optimization direction, a penalty term, Penalty, can be added to the objective function to minimize the deviation between the actual number of engine repair cycles and the target number of cycles, thus forming a complete optimization objective:

[0056] The penalty term can be defined as: =

[0057] Where I and J are the engine set and the planned month set, respectively. Let i be the cumulative cycle number of engine i in month j. Let i be the target cycle number when engine i is sent for repair. The variable is a 0-1 decision variable. When it is 1, it means that engine i is sent for repair in month j.

[0058] The constraints include compliance constraints, operational constraints, and maintenance plan constraints. Specifically, these include: the power unit can only be sent for maintenance once per maintenance cycle; there are limits on the number of maintenance cycles for the power unit; the power unit cannot be used during the maintenance period and for a period after maintenance completion; new leased power units are prohibited during specific time periods; the lease-in and lease-out of power units must be consistent; power unit returns must meet lease time requirements; the monthly available power units must meet flight mission requirements; and the number of maintenance cycles for the power unit can be increased through training. These constraints ensure the compliance and safety feasibility of the maintenance plan. For example: Sole repair constraint: = This ensures that each engine is sent for repair only once within the planned period T (total number of months).

[0059] Repair time constraint: Month of engine i's repair (Integer decision variables) are determined by maintenance indication variables: = .

[0060] Cycle count upper and lower limit constraints: The cumulative number of cycles when the engine is sent for repair must be within a preset range. ≤ +

[0061] in, , These are the minimum and maximum allowable number of cycles when engine i is sent for repair, respectively. Its initial loop number, The estimated increase in flight cycles per month for the engine. , Let i be the number of times engine i is trained on the left and right wings of the aircraft in year q (an integer variable). The number of cycles increased per engine training session.

[0062] Repair capacity constraint: The total number of repair requests per month shall not exceed the repair shop's capacity. ≤ ,in This represents the maintenance capacity for month j.

[0063] Retirement time constraints: The engine must be sent for repair before retirement. .

[0064] The monthly availability of power units must meet the constraints of flight mission requirements. Specifically, the monthly availability of power units must be no less than twice the number of aircraft. The output maintenance plan must satisfy this constraint to effectively avoid production delays caused by power unit shortages. The mathematical expression is: ≥

[0065] in, This represents the total number of engines available for the current month. This represents the number of aircraft on the wing in month j. The status is determined by the availability of existing engines, leased engines, purchased engines, and the condition of engines sent for repair. =

[0066] This is a 0-1 variable, indicating whether engine i is available in month j. , These represent the number of engines leased and owned in month j, respectively. The number of additional engines returned in month j. The number of engines purchased and delivered in advance in month j.

[0067] The restriction on adding new leased engines within a specific time period can be defined as prohibiting the addition of new leased engines for a specific year or earlier, further clarifying the scope of the restriction and ensuring compliance of the plan. The restriction is as follows: =0

[0068] For the months in which leasing is prohibited, This represents the number of engines newly leased in month j.

[0069] Building upon existing constraints, robust constraints are constructed by introducing uncertainty parameters to further enhance the adaptability of the solution in actual implementation. These uncertainty parameters primarily include fluctuations in the power unit maintenance cycle and monthly workload. Maintenance cycle fluctuation coefficients and workload fluctuation coefficients are determined based on these parameters. The workload fluctuation coefficient is introduced into the operational constraints, while the maintenance cycle fluctuation coefficient is introduced into the resource constraints. By introducing these fluctuation coefficients, the constraints are optimized to ensure that the solution still meets compliance and safety requirements within the range of uncertainties.

[0070] Specifically, the robustness constraint can be set to ensure that the number of available engines per month is no less than twice the number of aircraft multiplied by a first redundancy coefficient. This first redundancy coefficient is the sum of 1 and the flight volume fluctuation coefficient, to prevent engine shortages caused by sudden increases in flight volume. Its mathematical expression is: ≥

[0071] Where Δ1 is the flight volume fluctuation coefficient. The maintenance robustness constraint is set so that the engine repair time must be multiplied by the maintenance cycle in advance. This second redundancy coefficient is the sum of 1 and the maintenance cycle fluctuation coefficient, preventing the engine from failing to return to service on time due to extended maintenance cycles. This requires a set of Big M-method constraints to ensure that the engine is unavailable during the maintenance period (considering fluctuations) and the post-repair protection period.

[0072] in, This is the maintenance cycle fluctuation coefficient. The baseline maintenance duration (in months) is used. The protection period (in months) after the repair is completed, where M is a very large constant. It is a very small positive number. , , This is a 0-1 auxiliary variable used to indicate the state of engine i in month j relative to its maintenance period. Engine availability variable. The relationship with these auxiliary variables is as follows:

[0073]

[0074] Training iteration limit constraint:

[0075] Training logic control (through auxiliary variables) , accomplish):

[0076] in This is the estimated cumulative number of cycles for engine i in the 12 months prior to its planned retirement (i.e., at esn_retire_date_i-12). This constraint means that if the engine's cycle count has naturally reached the lower limit for maintenance before retirement, no further training is required.

[0077] The various constraints defined by the target and constraint definition module are transformed into mathematical expressions and integrated into the model to achieve accurate mathematical mapping of business scenarios.

[0078] The data processing and update module receives and processes various business data required for repair plan formulation. Its specific tasks include three parts: full data collection, data cleaning and standardization, and dynamic data updates. Full data collection covers power unit status, operating cycle count, warranty contracts, historical maintenance cycle fluctuation data, historical task volume fluctuation data, and repair shop capacity data. During data cleaning and standardization, outlier data is removed, missing data is supplemented, and data of different formats is converted into a unified input format for the model. Simultaneously, statistical analysis is performed on uncertain data to generate characteristic values ​​such as fluctuation ranges and probability distributions. Dynamic data updates establish a real-time data synchronization mechanism, collecting dynamic data such as power unit operating status, task structure adjustments, and capacity changes in real time. This provides the latest data input for rolling optimization, ultimately generating a structured input data set to ensure data accuracy and usability.

[0079] To address the issues of high dimensionality and slow solution in traditional holistic models, the high-dimensional model decomposition module proposes a model decomposition strategy based on business dimension decoupling. This strategy breaks down the high-dimensional holistic mixed-integer programming model into multiple low-dimensional, independently solvable, and collaborative sub-models, reducing solution complexity and improving efficiency. The decomposition is based on business type, specifically divided into a maintenance timing sub-model, a device allocation sub-model, and a maintenance resource scheduling sub-model. These three sub-models correspond to different decision priorities and are equipped with explicit linkage conditions to ensure consistency and synergy among the sub-models.

[0080] The maintenance timing sub-model uses the repair time and repair sequence of a single group of power units as decision variables, focusing on the compliance constraints and cycle number constraints of that group of power units. The unit allocation sub-model uses the number of power units leased and purchased each month as decision variables, determining the leasing and purchase cost constraints and robust operation constraints.

[0081] Its cost calculation is the core, and the rental cost is: =

[0082] in The number of engines returned in month j. The cost of leasing a single engine for one month.

[0083] Purchase cost (considering depreciation) is: =

[0084] in The purchase cost of a single engine. The annual depreciation rate for the purchased engines is given. The maintenance resource scheduling sub-model uses the power unit repair batch and repair shop allocation as decision variables, and combines the repair time period output by the maintenance timing sub-model to determine the maintenance resource utilization constraint.

[0085] Resource utilization rate is calculated as follows: =

[0086] The sub-model linkage condition is set as follows: the maintenance timing sub-model outputs the time period for power unit repair and the number of additional power units required in each stage. The time period for power unit repair is used as the parameter input. The maintenance resource scheduling sub-model solves for the specific repair time point, and the number of additional power units required in each stage is used as the parameter input. The device allocation sub-model solves for the leasing and purchasing scheme.

[0087] The rolling optimization scheduling module includes a phased rolling optimization mechanism, dividing long-term repair plans into multiple rolling optimization phases consisting of a decision period and a forecast period. This addresses the problem that traditional static models cannot adapt to dynamic business scenarios, enabling dynamic iterative updates to repair plans. The phase division divides the entire repair plan cycle into continuous rolling optimization phases. Each rolling optimization phase includes a decision period and a forecast period. The decision period is the planning period for determining execution, and the forecast period is a reference prediction period, the duration of which can be set according to actual business needs. For example, the decision period is typically set to 1 month, the forecast period to 2 months, and long-term repair plans can be set to 3 or 5 years.

[0088] During the current rolling optimization decision-making phase, continuously collected equipment update data includes actual operating status data of the power unit, actual resource occupancy data, and business adjustment data. Before formulating the equipment maintenance plan for the next rolling optimization decision-making phase, the actual equipment status data and actual resource occupancy data at the end of the current rolling optimization decision-making phase are obtained and updated into the equipment maintenance data. This data serves as the initial condition for resolving the maintenance sequence sub-model, equipment allocation sub-model, and maintenance resource scheduling sub-model. Based on the updated data, each sub-model is then resolved to generate the equipment maintenance plan for the next rolling optimization decision-making phase, which is then executed. During the rolling optimization process, plan continuity constraints are introduced, using parameters such as the remaining resources and actual engine status from the previous phase as inputs to the model for the next phase. This ensures that the repair plans, leasing / purchasing plans, and maintenance resource scheduling plans for the preceding and following phases are continuous and conflict-free, avoiding plan gaps caused by rolling optimization.

[0089] The sub-model collaborative solving module employs an open-source optimized solution framework for each decomposed low-dimensional sub-model. It integrates branch-and-bound methods with heuristic algorithms for collaborative sub-model solving, significantly improving efficiency while maintaining accuracy. The overall solution process is optimized to sub-model-level collaborative solving, such as... Figure 3 As shown.

[0090] Figure 3 The diagram illustrates the sub-model-level collaborative solution process. Starting with the conversion of the model into the solution framework format, it sequentially proceeds through sub-model heuristic pre-solution, branch-and-bound precise solution, and dual-algorithm collaborative optimization, ultimately achieving overall solution fusion and outputting a robust optimal solution. Specifically, this includes: model initialization, which converts each sub-model into an input format recognizable by the solution framework, completes parameter initialization, sets the branch-and-bound termination condition, and imports standardized input data and uncertainty feature values; the branch-and-bound termination condition includes a cost error of less than 1% and a sub-model solution time limit; sub-model heuristic pre-solution, which calls heuristic algorithms to approximate solutions for each sub-model, quickly obtaining initial feasible solutions for each sub-model, and simultaneously performing preliminary collaborative verification of the initial feasible solutions based on sub-model linkage conditions, eliminating conflicting solutions.

[0091] Furthermore, the branch and bound method is sequentially applied to each sub-model for precise solution. This involves splitting the integer decision variables of each sub-model, generating sub-model branch trees, relaxing the solution for bounding, and iterative pruning. During the solution process, cross-model collaborative pruning is implemented based on the linkage conditions of the sub-models. If the solution of a sub-model violates the linkage conditions, it is pruned directly without further solution, further improving the solution efficiency. During the branch and bound method solution process, heuristic algorithms are dynamically invoked to perform local optimization on the key branches of each sub-model, supplement new feasible solutions, update the lower bound of the objective function, and reduce the number of iterations. Finally, the optimal solutions of each sub-model are fused based on linkage constraints to generate a robust optimal solution for the overall repair plan, outputting core information such as the repair time, sequence, leasing and purchase plan, and maintenance resource scheduling plan for each power unit.

[0092] In practical applications, this collaborative solution method can shorten the solution time by more than 60% compared to the traditional direct solution of the overall model. It can be adapted to the maintenance plan formulation of a large-scale CFM5B engine fleet, and the output of a single plan formulation is still maintained at the hour level.

[0093] The solution verification and iteration module also includes multi-dimensional solution verification and closed-loop iteration functions. After the model solves and outputs the solution, it verifies it from three dimensions: compliance, robustness, and overall benefits. This ensures that the solution meets actual business needs and simultaneously optimizes the model in a closed loop based on actual execution data. Compliance verification uses a combination of model constraint reverse lookup and manual sampling to verify whether the solution meets all constraints, ensuring full compliance. Robustness verification introduces Monte Carlo simulation, performing multiple random samplings of uncertain parameters to verify the feasibility of the solution under different uncertain scenarios. Overall benefit verification calculates the comprehensive objective function value of the solution and compares it with the historical best solution and manually calculated solutions to ensure that the solution achieves optimal overall benefits.

[0094] If the verification fails, the data is fed back to the data processing and update module to adjust the constraint parameters or weight coefficients of the maintenance plan model and solve it again. The closed-loop iteration feeds back the actual execution data of the solution to the data processing and update module in real time to calibrate model parameters, optimize weight coefficients, and update uncertainty feature values, providing data support for the next rolling optimization and realizing continuous iterative optimization of the model and solution.

[0095] The system functionality provided in this application also includes three core functions: robust multi-scenario comparative analysis, rolling optimization plan visualization, and adaptive optimization of model parameters. Robust multi-scenario comparative analysis supports re-solving after adjusting multiple sets of parameters. Adjustable parameters include capacity demand, maintenance costs, weighting coefficients, and uncertainty fluctuation ranges. It enables comparative analysis of repair plans under different business scenarios, outputting core indicators such as comprehensive objective function values, costs, redundancy rates, and resource utilization rates for each scenario, providing quantitative basis for business decisions.

[0096] The rolling optimization plan visualization displays the decision-making period plan, forecast period plan, leasing and purchasing plan, and maintenance resource scheduling plan at each stage of rolling optimization, supporting real-time adjustment and traceability of the plan; the adaptive optimization of model parameters automatically optimizes parameters such as robust multi-objective weight coefficients and uncertainty fluctuation range based on historical execution data, improving the model's adaptability and solution accuracy, further reducing reliance on human experience, and realizing full automation and standardization of the repair plan formulation process.

[0097] In summary, this application provides a method for maintaining an aircraft power plant. The method involves acquiring maintenance data of the aircraft power plant; constructing a maintenance plan model based on the maintenance data, with the optimization objectives of minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization; decomposing the maintenance plan model into interrelated maintenance sequence sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types; and solving the problem based on each sub-model and the relationships between them to generate and execute an equipment maintenance plan. By constructing a maintenance planning model aimed at minimizing maintenance costs, rationalizing equipment redundancy, and maximizing maintenance resource utilization, a balanced optimization of maintenance plans is achieved across the three dimensions of economy, safety, and efficiency. This avoids resource waste caused by insufficient backup equipment or excessive redundancy due to solely pursuing the lowest cost. By decomposing the maintenance planning model into three interrelated sub-models—maintenance timing, equipment allocation, and maintenance resource scheduling—the solution complexity of high-dimensional complex models is reduced, enabling feasible solutions to be obtained within an acceptable timeframe for the formulation of maintenance plans for large-scale aero-engine fleets. By solving and executing maintenance plans based on each sub-model and their interrelationships, the coordination among the three decision dimensions of maintenance scheduling, equipment allocation, and maintenance resource scheduling is ensured, avoiding internal conflicts in the plans. The automated generation of aero-engine maintenance plans is achieved, reducing reliance on individual experience and improving the repeatability and standardization of plan formulation.

[0098] Based on the same technical concept, embodiments of this application also provide an aircraft power plant maintenance system, such as... Figure 4 As shown, it includes: Data acquisition module 401 is used to acquire equipment maintenance data of aircraft power units; The model building module 402 is used to build a maintenance plan model based on the device maintenance data. The maintenance plan model has the optimization objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. The model decomposition module 403 is used to decompose the maintenance plan model into interrelated maintenance timing sub-models, equipment allocation sub-models and maintenance resource scheduling sub-models according to the maintenance decision type. The plan generation module 404 is used to solve the problem based on each sub-model and the relationships between them, generate a device maintenance plan, and execute it.

[0099] This application also provides an electronic device corresponding to the method provided in the foregoing embodiments. Please refer to... Figure 5The diagram illustrates an electronic device provided by some embodiments of this application. The electronic device 20 may include: a processor 200, a memory 201, a bus 202, and a communication interface 203, wherein the processor 200, the communication interface 203, and the memory 201 are connected via the bus 202; the memory 201 stores a computer program that can run on the processor 200, and when the processor 200 runs the computer program, it executes the method provided by any of the foregoing embodiments of this application.

[0100] The memory 201 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one physical port (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.

[0101] Bus 202 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 201 is used to store programs. After receiving an execution instruction, the processor 200 executes the program. The method disclosed in any of the foregoing embodiments of this application can be applied to the processor 200, or implemented by the processor 200.

[0102] The processor 200 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 200 or by instructions in software form. The processor 200 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 201. The processor 200 reads the information in memory 201 and, in conjunction with its hardware, completes the steps of the above method.

[0103] The electronic devices and methods provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.

[0104] This application also provides a computer-readable storage medium corresponding to the method provided in the foregoing embodiments. Please refer to... Figure 6 The computer-readable storage medium shown is an optical disc 30, on which a computer program (i.e., a program product) is stored, which, when run by a processor, executes the methods provided in any of the foregoing embodiments.

[0105] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0106] The computer-readable storage medium provided in the above embodiments of this application and the method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.

[0107] It should be noted that the above embodiments are illustrative of this application and not limiting of it, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0108] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0109] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for maintaining an aircraft power plant, characterized in that, include: Acquire equipment maintenance data for aircraft power units; the equipment maintenance data includes fixed equipment data, which includes maintenance cost accounting data, available equipment quantity data, and maintenance resource occupancy data. Based on the device maintenance data, a maintenance plan model is constructed, with the optimization objectives of minimizing maintenance costs, rationalizing device redundancy, and maximizing maintenance resource utilization. The maintenance planning model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types; The solution is obtained based on each sub-model and the relationships between them, and a device maintenance plan is generated and executed. The process of constructing a maintenance plan model based on the device maintenance data includes: determining the maintenance cost minimization target, the device redundancy rate rationalization target, and the maintenance resource utilization maximization target based on preset maintenance business requirements; determining constraints based on the maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data, including cost constraints, redundancy constraints, and resource constraints; obtaining a comprehensive optimization target based on the maintenance cost minimization target, device redundancy rate rationalization target, and maintenance resource utilization maximization target using a weighted coefficient method; and constructing the maintenance plan model based on the device maintenance data, the comprehensive optimization target, and the constraints. The maintenance planning model is decomposed into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types. This includes: parsing maintenance timing-related decision variables and constraints, equipment allocation-related decision variables and constraints, and maintenance resource-related decision variables and constraints from the maintenance planning model according to the maintenance decision types; constructing the maintenance timing sub-model based on the maintenance timing-related decision variables and constraints, which is used to determine the maintenance time for each aero-engine unit; constructing the equipment allocation sub-model based on the equipment allocation-related decision variables and constraints, which is used to determine the external resource allocation required to fill maintenance gaps; and constructing the maintenance resource scheduling sub-model based on the maintenance resource-related decision variables and constraints, and the output of the maintenance timing sub-model, which is used to determine the scheduling and resource allocation of maintenance operations. The maintenance time periods and resource demand information output by the maintenance timing sub-model serve as input parameters for the equipment allocation sub-model and the maintenance resource scheduling sub-model. The process of solving the sub-models and their interrelationships to generate a device maintenance plan includes: solving the maintenance timing sub-model using a heuristic algorithm to obtain multiple candidate maintenance timing schemes, each candidate scheme including a planned maintenance time period and corresponding resource requirements; for each candidate scheme, inputting the corresponding planned maintenance time period and resource requirements into the maintenance resource scheduling sub-model and the device allocation sub-model respectively; if both the maintenance resource scheduling sub-model and the device allocation sub-model can obtain feasible solutions, then retaining the candidate maintenance timing scheme; and using each retained candidate maintenance timing scheme... Starting with the case, the generated candidate solutions are checked to see if they satisfy the parameter linkage relationship between the maintenance timing sub-model, the equipment allocation sub-model, and the maintenance resource scheduling sub-model. If not, the search for the branch containing the candidate solution is terminated. The current branch node is searched based on the branch and bound method, and the heuristic algorithm is called to perform local optimization on the current branch node to update the feasible solution and the boundary of the objective function. From all candidate solutions obtained after search and optimization that satisfy all parameter linkage relationships, the solution that makes the comprehensive optimization objective of the maintenance plan model optimal is selected, and combined with the corresponding maintenance resource scheduling scheme and equipment allocation scheme to generate the equipment maintenance plan.

2. The method according to claim 1, characterized in that, The device maintenance data includes device update data; the device update data includes the latest status data and current operational requirements data of the aircraft power unit. After the maintenance plan for the generating device is completed and executed, it also includes: The entire maintenance plan cycle is divided into several consecutive rolling optimization phases. Each rolling optimization phase includes a decision period and a prediction period. The decision period refers to the planning period for determining execution, and the prediction period refers to the planning period following the decision period for providing reference information. The planning length of the prediction period is not less than the planning length of the decision period. During the decision-making process of the current rolling optimization phase, device update data is continuously collected, and the device maintenance data is updated based on the device update data. Based on the updated equipment maintenance data, the maintenance timing sub-model, equipment allocation sub-model, and maintenance resource scheduling sub-model are solved to generate and execute the equipment maintenance plan for the next rolling optimization decision period.

3. The method according to claim 2, characterized in that, During the decision-making process of the current rolling optimization phase, device update data is continuously collected, and the device maintenance data is updated based on the device update data, including: Before generating the device maintenance plan for the next rolling optimization phase decision period, obtain the actual device status data and actual resource usage data at the end of the current rolling optimization phase decision period; The actual device status data and actual resource occupancy data are updated in the device maintenance data and used as input parameters for solving the maintenance timing sub-model, device allocation sub-model, and maintenance resource scheduling sub-model.

4. The method according to claim 1, characterized in that, The device maintenance data also includes device fluctuation data, which includes historical maintenance cycle fluctuation data and historical flight volume fluctuation data. The determination of the constraints also includes: Based on the historical maintenance cycle fluctuation data and the historical flight volume fluctuation data, the maintenance cycle fluctuation coefficient and the flight volume fluctuation coefficient are determined respectively. The flight volume fluctuation coefficient is introduced into the redundancy constraint, and the maintenance cycle fluctuation coefficient is introduced into the resource constraint to determine the constraints required for the maintenance plan model.

5. An aircraft power plant maintenance system, characterized in that, include: The data acquisition module is used to acquire equipment maintenance data of the aircraft power unit; The model building module is used to build a maintenance plan model based on the device maintenance data. The maintenance plan model aims to minimize maintenance costs, rationalize device redundancy, and maximize the utilization of maintenance resources. The device maintenance data includes fixed device data, which includes maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data. The process of constructing a maintenance plan model based on the device maintenance data includes: determining the maintenance cost minimization target, the device redundancy rate rationalization target, and the maintenance resource utilization maximization target based on preset maintenance business requirements; determining constraints based on the maintenance cost accounting data, available device quantity data, and maintenance resource occupancy data, including cost constraints, redundancy constraints, and resource constraints; obtaining a comprehensive optimization target based on the maintenance cost minimization target, device redundancy rate rationalization target, and maintenance resource utilization maximization target using a weighted coefficient method; and constructing the maintenance plan model based on the device maintenance data, the comprehensive optimization target, and the constraints. The model decomposition module is used to decompose the maintenance plan model into interrelated maintenance timing sub-models, equipment allocation sub-models, and maintenance resource scheduling sub-models according to maintenance decision types. This decomposition includes: parsing maintenance timing-related decision variables and constraints, equipment allocation-related decision variables and constraints, and maintenance resource-related decision variables and constraints from the maintenance plan model according to the maintenance decision types; and constructing the maintenance timing sub-models based on the maintenance timing-related decision variables and constraints. The timing sub-model is used to determine the maintenance time of each aero-engine unit; based on the unit allocation decision variables and constraints, the unit allocation sub-model is constructed, which is used to determine the allocation of external resources required to fill maintenance gaps; based on the maintenance resource decision variables and constraints, and the output of the maintenance timing sub-model, the maintenance resource scheduling sub-model is constructed, which is used to determine the scheduling of maintenance operations and resource allocation; wherein, the maintenance time period and resource demand information output by the maintenance timing sub-model are used as input parameters for the unit allocation sub-model and the maintenance resource scheduling sub-model; The plan generation module is used to solve for and execute a device maintenance plan based on each sub-model and the relationships between them. The process of solving for and generating the device maintenance plan based on each sub-model and the relationships between them includes: solving the maintenance timing sub-model using a heuristic algorithm to obtain multiple candidate maintenance timing schemes, each candidate scheme including a planned maintenance time period and corresponding resource requirement information; for each candidate maintenance timing scheme, inputting the corresponding planned maintenance time period and resource requirement information into the maintenance resource scheduling sub-model and the device allocation sub-model respectively; if both the maintenance resource scheduling sub-model and the device allocation sub-model can obtain feasible solutions, then the feasible solution is retained. The process involves: describing candidate maintenance timing schemes; starting with each retained candidate maintenance timing scheme, verifying whether the generated candidate solutions satisfy the parameter linkage relationship between the maintenance timing sub-model, the equipment allocation sub-model, and the maintenance resource scheduling sub-model; if not, terminating the search for the branch containing the candidate solution; searching the current branch node based on the branch and bound method, and calling the heuristic algorithm to perform local optimization on the current branch node to update the feasible solution and the objective function boundary; from all candidate solutions obtained after search and optimization that satisfy all parameter linkage relationships, selecting the solution that makes the comprehensive optimization objective of the maintenance plan model optimal, and combining it with the corresponding maintenance resource scheduling scheme and equipment allocation scheme to generate the equipment maintenance plan.

6. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method as claimed in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, It stores computer-readable instructions that can be executed by a processor to implement the method as described in any one of claims 1-4.

Images