Part optimization method, device, equipment, medium and product for engine

Abstract

Embodiments of the present application disclose a part optimization method, device, equipment, medium and product for an engine, the method comprising: generating a candidate exchange mode list corresponding to each to-be-replaced engine according to S to-be-replaced engines; creating a global conflict matrix for indicating a logical relationship between any two candidate exchange modes; sorting each candidate exchange mode in the S candidate exchange mode lists in descending order of a minimum cycle optimization amount of each candidate exchange mode to obtain a candidate exchange mode set; constructing an initial search tree through the global conflict matrix and the candidate exchange mode set, performing traversal processing on the initial search tree, and generating an initial batch part replacement scheme when the traversal processing reaches a first leaf node in the initial search tree; and performing backtracking processing on the initial search tree until an optimal batch part replacement scheme is obtained. The present application can improve the generation efficiency of the optimal batch part replacement scheme.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, medium and product for optimizing engine parts. Background Technology

[0002] As the culmination of modern industrial technology, aero engines are hailed as the "flower of industry." Their operational reliability, service life, and maintenance efficiency directly determine the operating costs, flight safety, and market competitiveness of airlines and aircraft operators, making them core equipment in the aviation operation system. With the rapid development of the air transport industry, airline fleets are constantly expanding, and the number of engines in service is continuously increasing. Engines of different service ages and operating conditions coexist, and the wear and tear of engine parts (such as fans / low-pressure compressors, high-pressure compressors, combustion chambers, high-pressure turbines, and low-pressure turbines) varies significantly. To extend the overall service life of engines, reduce maintenance costs, and improve the overall operational efficiency of the fleet, parts replacement has become a crucial method in aero engine maintenance. By replacing well-maintained parts with severely worn parts, engine performance can be quickly restored, avoiding the need for entire engine shutdowns due to single-part failures, reducing downtime and spare engine leasing expenses, and maximizing the remaining service life of each part to achieve maximum asset value. Therefore, how to scientifically and rationally plan engine parts replacement schemes has become a key requirement for airlines and aircraft operators to improve maintenance levels and control operating costs. In existing technologies, the implementation of engine component interchangeability mainly relies on manual experience. The specific process involves maintenance personnel manually recording and statistically analyzing the operating parameters of each engine (such as runtime, cycle count, exhaust temperature margin, etc.), the wear status and remaining lifespan of each component, and combining this with their accumulated maintenance experience to determine which engines can interchange components, which components to interchange, and the appropriate interchange sequence. Finally, a component interchange plan is formulated and executed. However, this manual experience-based component interchange method is no longer sufficient to meet the refined and efficient requirements of modern aero-engine maintenance. Specifically: First, manual decision-making is highly subjective, has low accuracy, and is prone to errors. Second, it is inefficient and cannot meet the maintenance needs of large-scale fleets. As the fleet size expands, the number of engines and components increases significantly. Manually statistically analyzing the status information of each engine and component requires substantial manpower and time, the formulation of interchange plans is time-consuming, it is difficult to quickly respond to engine maintenance needs, and it cannot effectively improve overall operational efficiency. In summary, existing methods for interchangeing engine parts rely on manual decision-making, which suffers from low accuracy and inefficiency, and cannot meet the needs of airlines and aircraft operators for refined, efficient, and low-cost parts interchangeability. Summary of the Invention

[0003] This application provides a method, apparatus, device, medium, and product for optimizing engine parts, which can automate and precisely optimize engine part replacement and improve the efficiency of generating optimal batch part replacement plans.

[0004] One embodiment of this application provides a method for optimizing engine parts, the method comprising: Get S engines with the same engine model that need to be replaced; S is an integer greater than 1. Based on each engine part in the S engines to be replaced, generate a candidate exchange method list for each engine to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information on replacing one or more engine parts in an engine to be replaced. Based on each candidate exchange method in the list of S candidate exchange methods, create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods; Obtain the minimum loop optimization amount for each candidate swap method in the list of S candidate swap methods. Sort each candidate swap method in the list of S candidate swap methods in descending order of the minimum loop optimization amount to obtain the set of candidate swap methods. An initial search tree is constructed using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. The initial search tree is traversed. When the traversal reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. Based on the initial batch parts replacement scheme, the initial search tree is backtracked until the optimal batch parts replacement scheme for the S engines to be replaced is obtained.

[0005] One embodiment of this application provides a component optimization device for an engine, the device comprising: The engine acquisition module is used to acquire S engines with the same engine model to be replaced; S is an integer greater than 1. The candidate exchange method determination module is used to generate a candidate exchange method list for each of the S engines to be replaced, based on each engine part in the S engines to be replaced. The candidate exchange methods in the candidate exchange method list are used to indicate information on replacing one or more engine parts in an engine to be replaced. The conflict detection module is used to create a global conflict matrix that indicates the logical relationship between any two candidate exchange methods based on each candidate exchange method in the list of S candidate exchange methods. The sorting module is used to obtain the minimum loop optimization amount corresponding to each candidate swap method in the S candidate swap method list, and sort each candidate swap method in the S candidate swap method list in descending order of the minimum loop optimization amount of each candidate swap method to obtain a set of candidate swap methods. The traversal processing module is used to construct an initial search tree by using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. The initial search tree is traversed. When the traversal reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. The backtracking module is used to backtrack the initial search tree based on the initial batch parts replacement plan until the optimal batch parts replacement plan for the S engines to be replaced is obtained.

[0006] In one alternative implementation, the S replacement engines include replacement engine A i , where i is a positive integer less than or equal to S; when the candidate exchange method determination module generates a candidate exchange method list for each of the S engines to be replaced based on each engine part, the candidate exchange method determination module specifically performs the following operations: For replacing engine A i The M engine parts are traversed and combined to obtain T parts combinations; M and T are both positive integers; a parts combination must include at least one of the M engine parts. Based on T parts combinations and S-1 remaining engine parts, generate a sublist of candidate exchange methods for each of the T parts combinations; S-1 remaining engines refer to the S engines to be replaced excluding engine A. i Other engines to be replaced; The T candidate swapping sublists are combined to form the engine A to be replaced. i The corresponding list of candidate swapping methods.

[0007] In one alternative implementation, the T component combinations include component combination B. j, j is a positive integer less than or equal to T; the candidate exchange method determination module is used to generate a sublist of candidate exchange methods corresponding to each of the T part combinations based on the T part combinations and the engine parts in the S-1 remaining engines. Specifically, the candidate exchange method determination module is used to perform the following operations: From the remaining S-1 engines, obtain the parts combination B respectively. j P engine parts with the same part type are considered as S-1 candidate part combinations; P is a positive integer less than or equal to M. According to part assembly B j Combine with candidate part S-1 to generate part combination B. j The corresponding G candidate swaps; G is a positive integer less than or equal to S-1; Assemble part B j The corresponding G candidate exchange methods are determined to be part combination B. j The corresponding candidate swapping sublist.

[0008] In one optional implementation, the list of S candidate swaps includes N candidate swaps, where N is a positive integer; the N candidate swaps include candidate swap C. k Candidate swapping method C z Both k and z are positive integers less than or equal to N; when the conflict detection module creates a global conflict matrix to indicate the logical relationship between any two candidate exchange methods based on each candidate exchange method in the S candidate exchange method list, the conflict detection module specifically performs the following operations: If candidate exchange method C k Engine parts and candidate exchange methods C z If one or more engine parts have the same part identifier, then candidate exchange method C is determined. k Candidate swapping method C z The logical relationship between them is mutual exclusion; If candidate exchange method C k Engine parts and candidate exchange methods C z If no engine part in the configuration has the same part identifier, then candidate exchange method C is determined. k Candidate swapping method C z The logical relationship between them is a compatibility relationship; When the logical relationship between any two candidate exchange methods among N candidate exchange methods is obtained, a global conflict matrix is ​​generated based on the logical relationship between any two candidate exchange methods.

[0009] In one optional implementation, the list of S candidate swaps includes N candidate swaps, where N is a positive integer; the N candidate swaps include candidate swap C. k k is a positive integer less than or equal to N; when the sorting module obtains the minimum loop optimization amount for each candidate swap method in the S candidate swap method list, the sorting module specifically performs the following operations: Obtain candidate swapping method C k The remaining cycle number of each engine part in the target engine to be replaced is determined as the minimum remaining cycle number of the target engine to be replaced; S engines to be replaced include the target engine to be replaced. Obtain candidate swapping method C k The remaining cycle number for each candidate part; candidate exchange method C k The candidate parts are selected through candidate exchange method C. k The corresponding combination of candidate parts determines the outcome; Obtain candidate swapping method C k The remaining cycle number for each unreplaced part; unreplaced parts refer to the target engine to be replaced, excluding candidate exchange method C. k Engine parts other than the released parts; Candidate exchange method C k The released part is determined based on one of the T part combinations corresponding to the target engine to be replaced; candidate exchange method C k Release components and candidate exchange methods C k The candidate parts in the list have the same part type; The remaining cycle number corresponding to each candidate part and the remaining cycle number corresponding to each unreplaced part are determined as the set of remaining cycle numbers. The minimum value in the set of remaining cycle numbers is determined as the second minimum remaining cycle number of the target engine to be replaced. The difference between the first minimum remaining cycle number and the second minimum remaining cycle number is determined as the candidate swap method C. k The corresponding minimum loop optimization amount.

[0010] In one optional implementation, the candidate swap set includes N candidate swaps, where N is a positive integer. When constructing the initial search tree using the global conflict matrix and the order of each candidate swap in the candidate swap set, the traversal processing module specifically performs the following operations: Create a root node with initial node state information; Based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set, a w-th layer child node is created for the w-1 layer target node; w is a positive integer less than or equal to N; the multiple target logical relationships include the logical relationship between the w-th candidate exchange method and each candidate exchange method in the w-1 layer selected exchange method set; when w is 1, the w-1 layer target node is the root node; the w-th layer child node is the child node of the w-1 layer target node, and the w-th layer child node includes a selection node for indicating the selection of the w-th candidate exchange method, and / or a skip node for indicating skipping the w-th candidate exchange method; the w-1-th candidate exchange method in the candidate exchange method set has a corresponding relationship with the w-1 layer target node; the w-1 layer target node is determined in the w-1 layer child nodes; the w-1 layer selected exchange method set includes the candidate exchange methods selected in the node path from the root node to the w-1 layer target node; w is a positive integer less than or equal to N; When the Nth level child node is obtained, an initial search tree is formed based on the nodes between the root node and the Nth level child node.

[0011] In one optional implementation, the traversal processing module is used to create a child node at the w-th level for a target node at the (w-1)-th level based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set. Specifically, the traversal processing module performs the following operations: Determine the target node at level (w-1) from the child nodes at level (w-1); If multiple target logical relationships in the global conflict matrix are all compatible, then based on the w-th candidate exchange method in the candidate exchange method set, for the target node of the (w-1)th layer, create the w-th selection node to indicate the selection of the w-th candidate exchange method, and the w-th skip node to indicate skipping the w-th candidate exchange method. Then, determine the w-th selection node and the w-th skip node as the w-th layer child nodes. If one or more of the target logical relationships in the global conflict matrix are mutually exclusive, then the first set of released part identifiers in the node status information of the target node at level w-1 is detected. If the first set of released part identifiers includes the part identifier of the candidate part in the candidate exchange method, then it is a target node at level w-1. A selection node for indicating the selection of the candidate exchange method at level w and a skip node for indicating skipping the candidate exchange method at level w are created. The selection node at level w and the skip node at level w are determined as child nodes at level w. If the first set of released part identifiers does not include the part identifier of the candidate part in the w-th candidate exchange method, then according to the w-th candidate exchange method in the set of candidate exchange methods, a w-th skip node is created for the target node of the (w-1)-th layer to indicate skipping the w-th candidate exchange method, and the w-th skip node is determined as the w-th layer child node.

[0012] In one alternative implementation, when the traversal processing module determines the target node at level (w-1) from the child nodes at level (w-1), the traversal processing module specifically performs the following operations: If the (w-1)th layer child nodes include the (w-1)th selected node and the (w-1)th skipped node, then the (w-1)th selected node is determined as the (w-1)th layer target node; If the (w-1)th layer of child nodes includes only the (w-1)th skipped node, then the (w-1)th skipped node is determined as the (w-1)th layer of target node.

[0013] In one alternative implementation, when the traversal processing module creates the w-th selection node to indicate the selection of the w-th candidate exchange method, the traversal processing module specifically performs the following operations: Each candidate swap in the (w-1)th selected swap set, and the wth candidate swap, form the wth selected swap set; Based on each candidate exchange method in the w-th selected exchange method set, a set of used part identifiers is generated to indicate the used engine parts, and a set of released part identifiers is generated to indicate the released parts; the set of used part identifiers includes the part identifiers of the engine parts for each candidate exchange method in the w-th selected exchange method set; the set of released part identifiers includes the part identifiers of the released parts for each candidate exchange method in the w-th selected exchange method set; the engine parts in the w-th candidate exchange method include the released parts in the w-th candidate exchange method. Based on each candidate swap method in the w-th selected swap method set, determine the cumulative loop optimization amount corresponding to selecting the w-th candidate swap method; Obtain the index position of the w-th candidate swap method in the candidate swap method set, and determine the search tree depth based on the index position of the w-th candidate swap method; The first node status information is composed of the set of used part identifiers, the set of released part identifiers, the cumulative loop optimization amount corresponding to the selection of the w-th candidate exchange method, the index position of the w-th candidate exchange method, and the search tree depth. Based on the status information of the first node, create the w-th selection node.

[0014] In one alternative implementation, when the traversal processing module creates the w-th skip node to indicate skipping the w-th candidate swap, the traversal processing module specifically performs the following operations: Obtain the node state information of the target node at layer w-1; Get the index position of the w-th candidate swap method in the candidate swap method set; The depth of the search tree is determined based on the index position of the w-th candidate swap. Based on the index position of the w-th candidate exchange method and the depth of the search tree, the node state information of the target node at the (w-1)-th layer is updated to obtain the second node state information. Based on the status information of the second node, create the w-th skip node.

[0015] In one optional implementation, the backtracking module is used to backtrack the initial search tree based on the initial batch parts replacement plan until the optimal batch parts replacement plan for the S engines to be replaced is obtained. Specifically, the backtracking module is used to perform the following operations: Obtain the total loop optimization amount corresponding to the initial batch part replacement plan; Obtain the timeout thresholds corresponding to the S engines to be replaced; In the initial search tree, the r-th backtracking node is obtained, along with the search execution time corresponding to it. If the search execution time is less than or equal to the timeout threshold, an upper bound pruning threshold and a remaining potential pruning threshold are generated based on the node state information and the remaining candidate swapping methods corresponding to the r-th backtracking node. The remaining candidate swapping methods corresponding to the r-th backtracking node refer to the candidate swapping methods whose index position is after the index position in the node state information of the r-th backtracking node. When r=1, the (r-1)-th backtracking node is the first leaf node in the initial search tree, the current optimal loop optimization amount corresponding to the (r-1)-th backtracking node is the total loop optimization amount, and the current optimal solution corresponding to the (r-1)-th backtracking node is the initial batch part replacement scheme. When r is greater than 1, the r-th backtracking node is determined based on the (r-1)-th backtracking node. r is a positive integer. If the current optimal loop optimization amount corresponding to the r-th backtracking node is less than the upper bound pruning threshold and less than the remaining potential pruning threshold, then according to the remaining candidate exchange method corresponding to the r-th backtracking node, a branch subtree of the r-th backtracking node is generated. According to the branch subtree of the r-th backtracking node, the current optimal solution corresponding to the (r-1)-th backtracking node is updated to obtain the current optimal solution corresponding to the r-th backtracking node. The loop optimization amount of the current optimal solution corresponding to the r-th backtracking node is determined as the current optimal loop optimization amount corresponding to the r-th backtracking node. If the current optimal loop optimization amount corresponding to the r-th backtracking node is greater than or equal to the upper bound pruning threshold and greater than or equal to the remaining potential pruning threshold, then continue to obtain the current optimal loop optimization amount corresponding to the (r+1)-th backtracking node. When there is no next backtracking node for the r-th backtracking node, the current optimal solution corresponding to the r-th backtracking node is determined as the optimal batch parts replacement scheme for the S engines to be replaced.

[0016] In one optional implementation, when the backtracking module obtains the r-th backtracking node, it specifically performs the following operations: If the (r-1)th backtracking node has an unvisited sibling node, then the unvisited sibling node of the (r-1)th backtracking node is determined as the rth backtracking node. If the (r-1)th backtracking node does not have any unvisited sibling nodes, then backtrack to the parent node of the (r-1)th backtracking node and determine the unvisited sibling node of the parent node of the (r-1)th backtracking node as the rth backtracking node.

[0017] In one optional implementation, the backtracking module updates the current optimal solution corresponding to the (r-1)th backtracking node based on the branch subtree of the r-th backtracking node. When obtaining the current optimal solution corresponding to the r-th backtracking node, the backtracking module specifically performs the following operations: Traverse the subtree of the r-th backtracking node. When the first leaf node in the subtree of the r-th backtracking node is reached, generate an updated batch parts replacement plan based on the node status information of the first leaf node in the subtree of the r-th backtracking node. Based on the updated batch parts replacement scheme, backtracking is performed on the branch subtree of the r-th backtracking node until the optimal batch parts replacement scheme corresponding to the r-th backtracking node is obtained. The optimal batch parts replacement scheme corresponding to the r-th backtracking node is determined as the current optimal solution corresponding to the r-th backtracking node.

[0018] In one optional implementation, when the backtracking module generates the upper bound pruning threshold and the remaining potential pruning threshold based on the node state information corresponding to the r-th backtracking node and the remaining candidate swapping methods corresponding to the r-th backtracking node, the backtracking module specifically performs the following operations: Based on the selected exchange method set, the released part identifier set, and the global conflict matrix in the node status information corresponding to the r-th backtracking node, the remaining candidate exchange methods corresponding to the r-th backtracking node are filtered to obtain Q remaining candidate exchange methods; Q is a positive integer. Based on the engine identifiers of the engines to be replaced corresponding to the Q remaining candidate exchange methods, the Q remaining candidate exchange methods are grouped to obtain Y remaining candidate exchange method groups; Y is a positive integer less than or equal to Q; the remaining candidate exchange methods in each remaining candidate exchange method group correspond to the same engine identifier; Obtain the minimum loop optimization amount corresponding to each of the Q remaining candidate swapping methods. Based on the minimum loop optimization amount of the remaining candidate swapping methods in each group of remaining candidate swapping methods, obtain the minimum loop optimization amount with the largest value from each of the Y remaining candidate swapping method groups. The upper bound pruning threshold is generated based on the sum of the minimum loop optimization quantities of Y nodes with the largest values ​​and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node. The remaining potential pruning threshold is generated based on the sum of the Q minimum loop optimization quantities and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node.

[0019] One embodiment of this application provides a computer device, including a processor, a memory, and an input / output interface; The processor is connected to a memory and an input / output interface, respectively. The input / output interface is used to receive and output data, the memory is used to store computer programs, and the processor is used to call the computer programs so that the computer device containing the processor executes the method in one aspect of the embodiments of this application.

[0020] One aspect of this application provides a computer-readable storage medium storing a computer program adapted to be loaded and executed by a processor, so that a computer device having the processor performs the method of one aspect of this application.

[0021] One aspect of this application provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in various optional embodiments of this application. In other words, when the computer instructions are executed by the processor, they implement the methods provided in various optional embodiments of this application.

[0022] Implementing the embodiments of this application will have the following beneficial effects: In this embodiment, S engines with the same engine model are obtained; S is an integer greater than 1. Based on each engine part in the S engines to be replaced, a candidate swapping method list is generated for each engine to be replaced. The candidate swapping methods in the list indicate information for replacing one or more engine parts in an engine to be replaced. Based on each candidate swapping method in the S candidate swapping method lists, a global conflict matrix is ​​created to indicate the logical relationship between any two candidate swapping methods. The minimum loop optimization amount corresponding to each candidate swapping method in the S candidate swapping method lists is obtained, and the minimum loop optimization amounts of each candidate swapping method are sorted from largest to smallest. The process involves sorting each candidate exchange method in the S candidate exchange method lists in a smaller order to obtain a candidate exchange method set. An initial search tree is constructed using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. This initial search tree is traversed, and when the traversal reaches the first leaf node, an initial batch parts replacement plan is generated based on the node state information of the first leaf node. The node state information includes the candidate exchange methods selected along the path from the root node to the first leaf node of the initial search tree. Based on the initial batch parts replacement plan, the initial search tree is backtracked until the optimal batch parts replacement plan for the S engines to be replaced is obtained. Through this process, a candidate exchange method list is generated for each engine to be replaced, thus generating all possible candidate exchange methods. This avoids the problems of limited experience and unreasonable exchange methods in traditional manual decision-making, which can lead to a single or unreasonable exchange method. It broadens the range of parts exchange options and improves the objectivity and comprehensiveness of the generated candidate exchange methods. By constructing a global conflict matrix, the logical relationship between any two candidate exchange methods can be clearly defined. This avoids the difficulty of comprehensively sorting out the logical relationships between multiple candidate methods during manual decision-making, which can lead to problems of incoordination due to conflicts during solution execution. The global conflict matrix provides a clear basis for subsequent solution selection, improving the rationality and feasibility of solution formulation and avoiding rework due to unidentified conflicts, thereby indirectly improving operation and maintenance efficiency. Rapidly generating initial batch parts replacement plans and iterating backtracking on the initial search tree shortens the formulation cycle of the optimal batch parts replacement plan for S engines to be replaced and improves the generation efficiency of the optimal batch parts replacement plan. Furthermore, by sorting each candidate exchange method to obtain a set of candidate exchange methods, it is possible to prioritize and retain candidate exchange methods with better optimization effects when generating subsequent initial batch parts replacement plans, achieving rapid locking of initial batch parts replacement plans and significantly improving the generation efficiency of initial batch parts replacement plans, thereby indirectly improving the generation efficiency of the optimal batch parts replacement plan. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a network interaction architecture diagram provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating a scenario of a component optimization method for an engine provided in an embodiment of this application. Figure 3 This is a flowchart illustrating a method for optimizing engine parts provided in an embodiment of this application. Figure 1 ; Figure 4 This is a schematic diagram of a backtracking process provided in an embodiment of this application. Figure 1 ; Figure 5 This is a flowchart illustrating a method for optimizing engine parts provided in an embodiment of this application. Figure 2 ; Figure 6 This is a schematic diagram of a backtracking process provided in an embodiment of this application. Figure 2 ; Figure 7 This is a schematic diagram of a parts optimization device for an engine provided in an embodiment of this application; Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0026] If this application requires the collection of object data (such as user data), a prompt interface or pop-up window will be displayed before and during the collection process. This prompt interface or pop-up window is used to inform the user that certain data is being collected. The data acquisition steps will only begin after the user confirms the prompt interface or pop-up window; otherwise, the process will end. Furthermore, the acquired user data will be used in reasonable and legal scenarios or for legitimate purposes. Optionally, in scenarios where user data needs to be used but user authorization has not been obtained, authorization can be requested from the user, and the user data can only be used after authorization is granted.

[0027] It is understood that, in the specific embodiments of this application, the user data involved requires user permission or consent when the following embodiments of this application are applied to specific products or technologies, and the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant regions.

[0028] In the embodiments of this application, please refer to Figure 1 , Figure 1 This is a network interaction architecture diagram provided in an embodiment of this application, such as... Figure 1 As shown, the system includes a server 101 and a cluster of service devices. The cluster of service devices may include service devices 102a, 102b, 102c, ..., 102n. Communication connections can exist between the service devices in the cluster; for example, there is a communication connection between service devices 102a and 102b, and between service devices 102a and 102c. Simultaneously, any service device in the cluster can have a communication connection with the server 101; for example, there is a communication connection between service device 102a and server 101. Optionally, the communication connection between service device 102a and any other service device (e.g., service device 102b) can be achieved through a communication connection between service device 102a and server 101, and a communication connection between server 101 and service device 102b. The communication connection method is not limited; it can be established directly or indirectly through wired communication, wireless communication, or other methods. This application does not impose any restrictions on this method.

[0029] It should be understood that the engine component optimization method mentioned in this application can be executed by any one of the business devices or server 101 in the business device cluster alone, or by any one of the business devices and server 101 in the business device cluster together, without any limitation.

[0030] The following example illustrates the engine parts optimization method executed jointly by business device 102b and server 101. Business device 102b can obtain the target engine model and retrieve S replacement engines from the engine database that match the target engine model. It can be understood that the S replacement engines share the same engine model, which is the target engine model, and S is an integer greater than 1. It should be noted that the engine database is accessible to business device 102b. The engine database can be a dedicated software system for managing engine data objects, recording abstract information such as engine identification, engine parts, technical specifications, and historical maintenance records through table structures or documents. The replacement engines and engine parts mentioned here are not physical entities, but rather identifiers or information used to indicate physical entities. In other words, a replacement engine is not the physical entity itself, but rather information used to indicate the physical entity of the replacement engine, such as identification information, simulation image information, or engine unique name information.

[0031] Furthermore, business device 102b can send S engines to be replaced to server 101. Server 101 can automatically analyze and process the S engines to be replaced according to the engine-specific parts optimization method mentioned in this application (see the detailed description below for specifics) to obtain the optimal batch parts replacement scheme for the S engines. Optionally, server 101 can call an application program to automatically analyze and process the S engines to be replaced to obtain the optimal batch parts replacement scheme for the S engines. It should be noted that the application program mentioned here refers to a program that encapsulates the engine-specific parts optimization method to achieve automatic analysis of the optimal batch parts replacement scheme and obtain the optimal batch parts replacement scheme. This application program can provide users with analysis functions for exchanging engine parts for a batch of engines to be replaced, and automatically analyze and identify which engines to be replaced and in what manner the engine parts should be exchanged, thereby maximizing engine life and improving the overall operating efficiency of the S engines to be replaced.

[0032] Server 101 can return the optimal batch parts replacement plan to business device 102b. Business device 102b can then simulate the assembly of S engines to be replaced according to the optimal batch parts replacement plan, obtaining the target optimized engines corresponding to the S engines to be replaced. The simulated assembly refers to the process of virtually matching, combining, and assembling candidate parts with the engines to be replaced according to the candidate exchange method in the optimal batch parts replacement plan, without performing actual physical disassembly and assembly operations, to obtain the optimized structure and state of the target optimized engines.

[0033] It is understood that the business equipment and server mentioned in the embodiments of this application can also be a type of computer equipment. Specifically, the business equipment mentioned above can be an electronic device, including but not limited to mobile phones, tablets, desktop computers, laptops, handheld computers, augmented reality / virtual reality (AR / VR) devices, wearable devices, and other mobile internet devices (MIDs) with network access capabilities. Figure 1 As shown, the service device can be a mobile phone (as shown in service device 102a), a desktop computer (as shown in service device 102b), a tablet computer (as shown in service device 102c), and a laptop computer (as shown in service device 102n), etc. Figure 1 Only a portion of the devices are listed. The servers mentioned above can be standalone physical servers, server clusters consisting of multiple physical servers, or distributed systems.

[0034] For details, please see Figure 2 , Figure 2 This is a schematic diagram illustrating a scenario for a component optimization method for an engine provided in an embodiment of this application. Taking a computer device as the executing entity as an example, the computer device can be, for example, […]. Figure 1 The shown device or server 101 can be any one of the service devices or servers 101 combined. For example... Figure 2 As shown, computer device 201 can obtain the target engine model 202 of the engine to be optimized, and engine database 203. Based on the target engine model 202, computer device 201 can retrieve engines with the same engine model from engine database 203, and identify S engines to be replaced from the retrieved multiple engines.

[0035] Furthermore, the computer device 201 can generate a candidate exchange method list for each of the S engines to be replaced, based on each engine part among them. The candidate exchange methods in the list indicate information for replacing one or more engine parts within an engine to be replaced. For example, the S engines to be replaced include engine E1, engine E2, ..., engine ES. The candidate exchange method list for engine E2 can be as follows: Figure 2As shown in the candidate exchange method list 204, engine E2 includes four engine parts: part a2, part b2, part c2, and part d2. The candidate exchange method list 204 should include information on replacing one or more engine parts in engine E2. The one or more engine parts in engine E2 can include 15 different representations: "part a2", "part b2", "part c2", "part d2", "part a2, part b2", "part a2, part c2", "part a2, part d2", "part b2, part c2", "part b2, part d2", "part a2, part b2, part c2", "part a2, part b2, part d2", "part a2, part c2, part d2", "part b2, part c2, part d2", and "part a2, part b2, part c2, part d2".

[0036] In the candidate exchange method list 204, there are S-1 candidate exchange methods for part a2, including "Part E2-a2 for Part E1-a1", ..., "Part E2-a2 for Part ES-aS". Taking the part exchange method "Part E2-a2 for Part E1-a1" as an example, engine E2 acts as the source engine (i.e., the receiving engine), and engine E1 acts as the donor engine. Executing this part exchange method means that part a1 needs to be removed from engine E1, part a2 needs to be removed from engine E2, and the removed part a1 needs to be installed into engine E2. It should be noted that part a2 removed from engine E2 is not installed into engine E1 immediately, but is used as a release part of engine E2 for subsequent cascaded exchanges.

[0037] It should be noted that the removal and installation of parts described here are logical process descriptions, not physical operations on the actual engine or engine parts. Here, part a2 and part E2-a2 both refer to engine part a in engine E2, part a1 and part E1-a1 both refer to engine part a in engine E1, and part ES-aS refers to engine part a in engine ES. Engine part a refers to an engine part of a specific part type, such as a fan / low-pressure compressor. Parts a1, a2, and aS are used to indicate the fan / low-pressure compressor engine part located in a different engine.

[0038] In the candidate exchange method list 204, there are S-1 candidate exchange methods for part b2, including "part E2-b2 exchanged for part E1-b1", ..., "part E2-b2 exchanged for part ES-bS". For "parts a2, b2, c2, and d2, i.e., parts a2 / b2 / c2 / d2 in the candidate exchange method list 204", there are S-1 candidate exchange methods, including "part E2-a2 / b2 / c2 / d2 exchanged for part E1-a1 / b1 / c1 / d1", ..., "part E2-a2 / b2 / c2 / d2 exchanged for part ES-a1 / b1 / c1 / d1".

[0039] Furthermore, the computer device 201 can create a global conflict matrix 205, based on each candidate exchange method in the S candidate exchange method list, to indicate the logical relationship between any two candidate exchange methods. The S candidate exchange method list includes N candidate exchange methods, where N is a positive integer. The global conflict matrix 205 indicates the logical relationship between any two candidate exchange methods from the N candidate exchange methods. The logical relationship includes a mutually exclusive relationship indicating a conflict between the two candidate exchange methods, and a compatible relationship indicating that the two candidate exchange methods do not conflict. As shown in the global conflict matrix 205, the N candidate exchange methods include candidate exchange method C1, candidate exchange method C2, ..., candidate exchange method C... N The logical relationship between candidate swapping method C1 and candidate swapping method C2 is mutually exclusive. N The logical relationship between them is a compatibility relationship; candidate swapping method C2 and candidate swapping method C... N The logical relationship between them is one of compatibility.

[0040] Because an engine part taken from one engine to be replaced cannot be simultaneously replaced in a different engine, or simultaneously used by different candidate exchange methods. For example, if two candidate exchange methods are "part E2-a2 for part E1-a1" and "part E3-a3 for part E1-a1", part a1 taken from engine E1 cannot be simultaneously replaced in engine E2 and engine E3. The logical relationship between these two candidate exchange methods is mutual exclusion.

[0041] Alternatively, the two candidate exchange methods are "part E2-a2 / b2 for part E1-a1 / b1" and "part E2-a2 for part E1-a1". In the first candidate exchange method, parts a1 and b1 taken from engine E1 are exchanged for parts E2. In the second candidate exchange method, part a1 taken from engine E1 is exchanged for parts E2. Both candidate exchange methods use part a1 from engine E1, and the logical relationship between these two candidate exchange methods is mutually exclusive. That is, the logical relationship between candidate exchange method C1 and candidate exchange method C2 is mutually exclusive; executing candidate exchange method C1 means that candidate exchange method C2 cannot be executed, and vice versa.

[0042] Furthermore, the computer device 201 can obtain the minimum cycle optimization amount corresponding to each of the S candidate swapping methods in the candidate swapping method list. It then sorts each candidate swapping method in the S candidate swapping method list according to the minimum cycle optimization amount in descending order, obtaining a candidate swapping method set 206. As shown in the candidate swapping method set 206, the order of the candidate swapping methods is now candidate swapping method C5, candidate swapping method C4, candidate swapping method C1, ..., candidate swapping method C2. That is, the minimum cycle optimization amount of candidate swapping method C5 is greater than that of candidate swapping method C4, the minimum cycle optimization amount of candidate swapping method C4 is greater than that of candidate swapping method C1, and the minimum cycle optimization amount of candidate swapping method C2 is the smallest.

[0043] Computer device 201 can construct an initial search tree 207 based on the global conflict matrix 205 and the order of each candidate exchange method in the candidate exchange method set 206. The initial search tree 207 can be traversed. When the traversal reaches the first leaf node in the initial search tree 207, an initial batch parts replacement scheme 208 is generated based on the node state information of the first leaf node. The node state information includes the candidate exchange methods selected along the node path from the root node of the initial search tree 207 to the first leaf node. In other words, the initial batch parts replacement scheme 208 consists of the candidate exchange methods selected along the node path from the root node of the initial search tree 207 to the first leaf node.

[0044] The computer device 201 can backtrack the initial search tree 207 based on the initial batch parts replacement scheme 208 until the optimal batch parts replacement scheme 209 for the S engines to be replaced is obtained.

[0045] Through the above process, based on engine parts, a list of candidate exchange methods is generated for each engine to be replaced, thus generating all possible candidate exchange methods. This avoids the problems of limited experience and unreasonable exchange methods in traditional manual decision-making, which leads to a single exchange method. It broadens the range of parts exchange options and improves the objectivity and comprehensiveness of the generated candidate exchange methods. By constructing a global conflict matrix, the logical relationship between any two candidate exchange methods can be clearly defined. This avoids the difficulty in comprehensively sorting out the logical relationships between multiple candidate methods during manual decision-making, which can lead to problems of uncoordinated solutions during implementation. The global conflict matrix provides a clear basis for subsequent solution selection, improving the rationality and feasibility of solution formulation and avoiding rework due to unidentified conflicts, thereby indirectly improving operation and maintenance efficiency. The rapid generation of initial batch parts replacement plans and the backtracking iteration of the initial search tree shorten the formulation cycle of the optimal batch parts replacement plan for S engines to be replaced and improve the generation efficiency of the optimal batch parts replacement plan. Furthermore, by sorting each candidate exchange method, a set of candidate exchange methods is obtained. This allows for the priority retention of candidate exchange methods with better optimization effects when generating initial batch part replacement schemes. This enables the rapid locking of initial batch part replacement schemes, significantly improving the generation efficiency of initial batch part replacement schemes, and indirectly improving the generation efficiency of optimal batch part replacement schemes.

[0046] Further, please see Figure 3 , Figure 3 This is a flowchart illustrating a method for optimizing engine parts provided in an embodiment of this application. Figure 1 This method for optimizing engine parts can be executed by computer equipment, such as... Figure 1 The shown business device or server 101 can also be a device composed of any business device and server 101. The engine component optimization method can include at least the following steps S301-S306: Step S301: Obtain S engines with the same engine model to be replaced; S is an integer greater than 1.

[0047] In this embodiment, the computer device acquires the target engine model and an engine database. Based on the target engine model, it selects engines with the same model number from the engine database as replacement engines, resulting in S replacement engines with the same model number. Here, S is an integer greater than 1. The target engine model can be manually selected or determined according to preset selection rules. For example, the computer device can group all engines in the engine database according to their model number based on the minimum remaining cycle count of each engine in the database, obtaining multiple engine combinations with different model numbers. It then obtains the average minimum remaining cycle count for each engine combination and determines the engine model corresponding to the combination with the smallest average minimum remaining cycle count as the target engine model. The minimum remaining cycle count refers to the minimum value among the remaining cycle counts of multiple engine parts within an engine. The remaining cycle count of an engine part refers to the remaining flight cycles that the engine part can safely continue operating in its current state; it is a core quantitative indicator for measuring the remaining usable life of the engine part. The overall lifespan of an engine can be determined by the engine part with the fewest remaining cycle count, i.e., by the minimum remaining cycle count.

[0048] Optionally, the computer equipment can retrieve a batch of engines with different engine models from the engine database. Then, based on the engine model, this batch of engines with different engine models can be grouped into multiple engine groups, each with a different engine model. Engines within an engine group share the same engine model. The computer equipment can identify any given engine group as containing S engines with the same engine model to be replaced. The process of retrieving this batch of engines with different engine models can be as follows: retrieving engines from the engine database whose minimum remaining cycle count is less than a minimum remaining cycle count threshold; retrieving engines from the engine database whose maintenance cycle is within a maintenance cycle period; or retrieving a batch of engines with different engine models based on a specific aircraft type or optimization potential. For example, if the specific aircraft type is a narrow-body aircraft, the computer equipment can select all engines operating on that narrow-body aircraft from the engine database as a batch of engines with different engine models. The maintenance cycle period can be pre-set, such as "the next 3 months," meaning the period from the current system time to 3 months later.

[0049] It should be noted that replacement engines with the same engine model can include, but are not limited to, engines with identical engine model identifiers (such as engines all having the model number CFM56-7B22), as well as engines from the same series that have equivalent physical interfaces and performance and are interchangeable. For example, engines with the models CFM56-7B22, CFM56-7B24, and CFM56-7B26 are all CFM56-7B series engines, and these three engine models can be considered to be engines with the same engine model.

[0050] Step S302: Based on each engine part in the S engines to be replaced, generate a candidate exchange method list for each engine to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information on replacing one or more engine parts in an engine to be replaced.

[0051] In this embodiment, the computer device can generate a component combination corresponding to each of the S engines to be replaced, based on each engine component among the S engines to be replaced, and then generate a candidate replacement method list corresponding to each engine to be replaced based on the component combination corresponding to each engine to be replaced. The candidate replacement methods in the candidate replacement method list are used to indicate information for replacing one or more engine components in an engine to be replaced.

[0052] Specifically, taking one of the S engines to be replaced (referred to as engine E1) as an example, the computer equipment can generate M part combinations corresponding to each of the M engine parts in engine E1. It can be understood that the part combinations corresponding to each engine part can include single-part combinations and multi-part combinations. For example, taking M engine parts including part 1 and part 2 as an example, the single-part combinations corresponding to engine E1 can include "(part 1)" and "(part 2)", while the multi-part combinations can include "(part 1, part 2)".

[0053] The M engine parts of an engine can include, but are not limited to, the fan / low-pressure compressor (LPC), high-pressure compressor (HPC), combustion chamber and high-pressure turbine (CC&HPT), and low-pressure turbine (LPT). HPC and CC&HPT are engine parts subject to CSPR (Cycle Since Part Replacement), meaning that HPC and CC&HPT have a maximum operating cycle limit after replacement, and their remaining lifespan is determined by the cumulative operating cycle count. In other words, once the cumulative operating cycle count reaches the limit, they can no longer be used. For example, the CSPR limit (operating cycle threshold) for a narrow-body engine is 5,000 cycles. When the cumulative operating cycle count of an HPC part used in that narrow-body engine reaches 5,000, the part can be considered unusable and unable to continue maintenance.

[0054] Furthermore, the computer equipment can generate a sublist of candidate exchange methods for each part combination corresponding to engine E1 based on each part combination and the engine parts in the other engines. The computer equipment can then combine these sublists of candidate exchange methods for each part combination to form a list of candidate exchange methods for engine E1. Here, "the other engines" refers to the S-1 engines to be replaced, excluding the engine to be replaced in this example (engine E1). Taking the single part combination "(part 1)" in the above example as an example, the sublist of candidate exchange methods corresponding to this single part combination "(part 1)" can include candidate exchange methods generated by exchanging part 1 in engine E1 with part 1 in each of the S-1 engines to be replaced, such as "part 1 in engine E1 for part 1 in engine E2", "part 1 in engine E1 for part 1 in engine E3", etc. It should be noted that engines E2 and E3 both belong to the S-1 engines to be replaced.

[0055] Step S303: Based on each candidate exchange method in the list of S candidate exchange methods, create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods.

[0056] In this embodiment, the list of S candidate exchange methods includes N candidate exchange methods, where N is a positive integer. The logical relationships include mutual exclusion and compatibility. A computer device can select any two candidate exchange methods from the N candidate exchange methods. If one or more engine parts with the same part identifier exist among the engine parts in these two candidate exchange methods, the computer device can determine that the logical relationship between these two candidate exchange methods is mutual exclusion. Conversely, if no engine parts with the same part identifier exist among the engine parts in these two candidate exchange methods, the computer device can determine that the logical relationship between these two candidate exchange methods is compatibility.

[0057] Furthermore, when the computer device obtains the logical relationship between every two candidate exchange methods among the N candidate exchange methods, it can generate a global conflict matrix based on the logical relationship between every two candidate exchange methods.

[0058] Specifically, the N candidate swapping methods include candidate swapping method C. k Candidate swapping method C z Both k and z are positive integers less than or equal to N. If candidate swapping method C... k Engine parts and candidate exchange methods C z If one or more engine parts have the same part identifier, then candidate exchange method C is determined. k Candidate swapping method C z The logical relationship between them is mutual exclusion. For example, candidate exchange method C... k This is represented as "part E2-a2 exchanged for part E1-a1", with candidate exchange method C. z This is represented as "part E3-a3 for part E1-a1". Among these two candidate exchange methods, there is an engine part with the same part identifier, namely part E1-a1. The computer equipment can determine candidate exchange method C. k Candidate swapping method C z The logical relationship between them is mutual exclusion, that is, candidate exchange method C k Candidate swapping method C z Only one option can be executed. It should be noted that the part identifier of an engine part is a unique label assigned to each engine part; that is, the part identifier of an engine part uniquely identifies that engine part.

[0059] Optionally, if candidate swap method C k Engine parts and candidate exchange methods C z If no engine part in the configuration has the same part identifier, then candidate exchange method C is determined. k Candidate swapping method Cz The logical relationship between them is a compatibility relationship. For example, candidate exchange method C... k This is represented as "part E2-a2 exchanged for part E1-a1", with candidate exchange method C. z This is represented as "part E3-a3 for part E4-a4". Since there are no engine parts with the same part identifier among these two candidate exchange methods, the computer equipment can determine candidate exchange method C. k Candidate swapping method C z The logical relationship between them is a compatibility relationship, that is, candidate exchange method C k Candidate swapping method C z They can be executed together.

[0060] An alternative representation of the global conflict matrix can be found in Table 1: Table 1

[0061] As shown in Table 1, "True" indicates that the two corresponding component exchange methods conflict, meaning they are mutually exclusive. "False" indicates that the two corresponding component exchange methods do not conflict, meaning they are compatible. Table 1 shows that the logical relationship between candidate exchange method C1 and candidate exchange method C2 is mutually exclusive; they cannot be performed simultaneously. That is, if candidate exchange method C1 is selected, candidate exchange method C2 cannot be selected; conversely, if candidate exchange method C2 is selected, candidate exchange method C1 cannot be selected. The logical relationship between candidate exchange method C1 and candidate exchange method C2 is mutually exclusive. N The logical relationship between them is one of compatibility; they can be performed simultaneously. Candidate swapping method C2 and candidate swapping method C... N The logical relationship between them is one of compatibility, meaning they can be performed simultaneously.

[0062] Optionally, the computer device can also convert the global conflict matrix into a global conflict list. This involves retrieving two mutually exclusive candidate swaps from the global conflict matrix, constructing conflict data pairs, and storing these pairs in a list format to obtain the global conflict list. Subsequently, the computer device does not need to query the global conflict matrix for the logical relationship between two candidate swaps; it can directly query the global conflict list, thus facilitating the rapid retrieval of the logical relationship between two candidate swaps.

[0063] This step constructs a global conflict matrix, enabling clear quantification and visualization of the conflict relationships between candidate exchange methods. This avoids the problem that traditional manual decision-making struggles to comprehensively analyze the conflict relationships between multiple candidate exchange methods, which can easily lead to uncoordinated solutions during execution. Furthermore, the global conflict matrix provides a clear basis for subsequent candidate exchange method selection, improving the rationality and feasibility of the optimal batch parts replacement solution. It also avoids rework due to unidentified conflicts, indirectly improving operational efficiency.

[0064] Step S304: Obtain the minimum loop optimization amount corresponding to each candidate exchange method in the S candidate exchange method list. Sort each candidate exchange method in the S candidate exchange method list in descending order of the minimum loop optimization amount of each candidate exchange method to obtain the candidate exchange method set.

[0065] In this embodiment, the list of S candidate exchange methods includes N candidate exchange methods, where N is a positive integer. The computer device can obtain the minimum loop optimization amount corresponding to each of the N candidate exchange methods. A candidate exchange method includes a released part and a candidate part. The released part can refer to the engine part removed from the source engine in the candidate exchange method, and the candidate part can refer to the engine part provided by the donor engine in the candidate exchange method, i.e., the engine part that needs to be replaced by the source engine. For example, a candidate exchange method is represented as "part E2-a2 for part E1-a1", where engine E2 is the source engine (i.e., the receiving engine), engine E1 is the donor engine, the released part is "part E2-a2", and the candidate part is "part E1-a1".

[0066] Specifically, the computer equipment can obtain the remaining cycle number corresponding to each of the S engine parts to be replaced. The minimum value among the remaining cycle numbers corresponding to each of the M engine parts in one engine to be replaced (i.e., the minimum of the M remaining cycle numbers) is determined as the first minimum remaining cycle number for that engine. This allows the determination of the first minimum remaining cycle number for each of the S engines to be replaced.

[0067] The computer device can obtain the remaining cycle count of candidate parts for each candidate exchange method, and the remaining cycle count of unreplaced parts for each candidate exchange method. It can then select the minimum remaining cycle count from both the remaining cycle counts of candidate parts and unreplaced parts for each candidate exchange method, and define this minimum remaining cycle count as the second minimum remaining cycle count for that candidate exchange method. The computer device can then determine the difference between the first minimum remaining cycle count and the second minimum remaining cycle count for each candidate exchange method as the minimum cycle optimization amount for that candidate exchange method.

[0068] In this context, the unreplaced parts corresponding to a candidate exchange method refer to the engine parts in the source engine corresponding to that candidate exchange method, excluding the released parts. Continuing with the example of the part exchange method "part E2-a2 for part E1-a1", the source engine is engine E2, which includes M engine parts: part a2 (remaining cycle count: 500), part b2 (remaining cycle count: 800), part c2 (remaining cycle count: 1800), and part d2 (remaining cycle count: 2000). The released part is a2, and the unreplaced parts include parts b2, c2, and d2. The candidate part is part a1 (remaining cycle count: 1000). In this case, the second minimum remaining cycle count for this part exchange method is the remaining cycle count corresponding to part b2, which is 800.

[0069] It should be noted that the first minimum remaining cycle number corresponding to a candidate exchange method refers to the first minimum remaining cycle number of the engine to be replaced corresponding to the candidate exchange method. The engine to be replaced corresponding to the candidate exchange method can refer to the source engine in the candidate exchange method.

[0070] In the example above, the engine to be replaced corresponding to the candidate exchange method is engine E2. The first minimum remaining cycle number corresponding to this candidate exchange method is the first minimum remaining cycle number of engine E2, which is the remaining cycle number of part a2 in engine E2, 500. The minimum cycle optimization amount corresponding to this candidate exchange method is the difference between the first minimum remaining cycle number (i.e., 500) and the second minimum remaining cycle number (i.e., 800) corresponding to this candidate exchange method, which is 300.

[0071] Furthermore, the computer equipment can sort the N candidate swapping methods in descending order of the minimum cyclic optimization amount of each candidate swapping method to obtain a set of candidate swapping methods.

[0072] Step S305: Construct an initial search tree by using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. Traverse the initial search tree. When the traversal reaches the first leaf node in the initial search tree, generate an initial batch part replacement plan based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node.

[0073] In this embodiment of the application, the computer device can construct an initial search tree by using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. Specifically, the computer device can create a root node. The node state information of the root node can be represented as "{swap[],used{},avail{},improvement:0,index:0}", where "swap[]" indicates the selected candidate swap method, that is, the set of selected swap methods corresponding to the current node, which is empty at this time; "used{}" indicates the set of engine parts (including released parts and candidate parts) corresponding to the selected candidate swap method, that is, the set of used part identifiers, which is empty at this time; "avail{}" indicates the set of released parts corresponding to the selected candidate swap method, that is, the set of released part identifiers, which is empty at this time; "improvement:0" indicates the cumulative value of the minimum loop optimization amount of the selected candidate swap method (hereinafter referred to as the cumulative loop optimization amount), which is 0 at this time; "index:0" indicates the index position of the currently selected candidate swap method in the set of candidate swap methods, or indicates the depth of the search tree, that is, the depth (or level) of the corresponding node in the initial search tree.

[0074] The computer device can obtain the first candidate swap from the set of candidate swaps. Based on the first candidate swap, it creates a selection node for the root node that selects the first candidate swap, and a skip node that skips the first candidate swap. Figure 2Taking the candidate swapping method set 206 in the embodiment as an example, the first candidate swapping method is candidate swapping method C5. If candidate swapping method C5 is represented as "E3-a3 swaps E4-a4", and the minimum loop optimization amount is 2000, then the node status information of the selection node (hereinafter referred to as selection node C5) created by the computer device to select candidate swapping method C5 can be represented as "{swap[C5],used{E3-a3,E4-a4},avail{E3-a3},improvement:2000,index:1}". Here, "swap[C5]" indicates that the current selected node's selected swap method set includes candidate swap method C5; "used{E3-a3,E4-a4}" indicates the set of used part identifiers after selecting candidate swap method C5, including used parts E3-a3 and E4-a4; "avail{E3-a3}" indicates the set of released part identifiers after selecting candidate swap method C5, including released part E3-a3, meaning that part E3-a3 can be used for cascading in subsequent candidate swap methods. The node status information of the skip node (hereinafter referred to as skip node C5) created by the computer device to skip candidate swap method C5 can be represented as "{swap[],used{},avail{},improvement:0,index:1}".

[0075] Furthermore, the computer device can create child nodes for the selected node C5. It can be understood that if the candidate switching method C4 and candidate switching method C5 are compatible, the computer device can create two child nodes for the selected node C5: a selected node C4 and a skipped node C4. The specific implementation process for creating the selected node C4 and skipped node C4 can be found in the detailed description of creating the selected node C5 and skipped node C5 above, and will not be repeated here. If candidate swap C4 is represented as "E2-b2 swapped for E4-b4", and the minimum loop optimization is 1800, then the node state information of selected node C4 can be represented as "{swap[C5,C4],used{E3-a3,E4-a4,E2-b2,E4-b4},avail{E3-a3,E2-b2},improvement:3800,index:2}"; the node state information of skipped node C4 can be represented as "{swap[C5],used{E3-a3,E4-a4},avail{E3-a3},improvement:2000,index:2}". In other words, except for the index, the rest of the node state information of skipped node C4 is the same as the node state information of selected node C5.

[0076] Optionally, if candidate swapping method C4 and candidate swapping method C5 are mutually exclusive, the computer device can create a child node that skips node C4 for selected node C5.

[0077] Furthermore, if the selected node C5 includes two child nodes, the selected node C4 and the skipped node C4, the computer device continues to create a child node for the selected node C4 according to the candidate exchange method C1 in the candidate exchange method set 206. If the selected node C5 includes only one child node, the skipped node C4, the computer device continues to create a child node for the skipped node C4 according to the candidate exchange method C1 in the candidate exchange method set 206.

[0078] This process continues until the computer device creates a child node corresponding to the second-to-last candidate exchange method (selection node or skip node) in the candidate exchange method set 206, i.e., candidate exchange method C2, thus obtaining the initial search tree.

[0079] The computer device can traverse the initial search tree. When the traversal reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node state information of the first leaf node. The node state information includes the candidate swapping methods selected in the node path from the root node to the first leaf node of the initial search tree. The traversal process can be a depth-first traversal.

[0080] For example, when the computer device creates a selection node C2 and a skip node C2 for the selection node corresponding to the second-to-last candidate exchange method based on the last candidate exchange method in the candidate exchange method set 206, namely candidate exchange method C2, the first leaf node in the initial search tree is the selection node C2. At this time, the initial batch part replacement scheme includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. That is to say, at this time, the initial batch part replacement scheme includes at least candidate exchange method C5, the second-to-last candidate exchange method, and candidate exchange method C2.

[0081] When the computer device creates a skip node C2 for the selection node corresponding to the second-to-last candidate exchange method based on the last candidate exchange method in the candidate exchange method set 206, namely candidate exchange method C2, the first leaf node in the initial search tree is the skip node C2. At this time, the initial batch part replacement scheme includes at least candidate exchange method C5 and the second-to-last candidate exchange method.

[0082] Step S306: Based on the initial batch parts replacement plan, backtrack the initial search tree until the optimal batch parts replacement plan for the S engines to be replaced is obtained.

[0083] In this embodiment, the computer device can perform backtracking processing on the initial search tree based on the initial batch parts replacement plan. Specifically, the computer device can obtain the total loop optimization amount corresponding to the initial batch parts replacement plan, and obtain the timeout thresholds corresponding to the S engines to be replaced. The computer device can use the initial batch parts replacement plan as the current optimal solution. Then, in the initial search tree, recursively return from the first leaf node to the root node, treating unvisited nodes as backtracking nodes in turn, and continuing the exploration downwards based on the backtracking nodes. That is, based on the backtracking node and the remaining candidate swapping methods corresponding to the backtracking node, analyze the candidate solutions existing in the branch subtree formed by the remaining candidate swapping methods corresponding to the backtracking node. If the total loop optimization amount of the candidate solutions is greater than the total loop optimization amount of the current optimal solution, then the computer device can determine the candidate solution as the new current optimal solution. Furthermore, the computer device can continue backtracking from the node corresponding to the current optimal solution, updating the current optimal solution step by step, until the backtracking process ends, that is, there is no next backtracking node, or in other words, when backtracking to the root node, the computer device can determine the current optimal solution obtained at this time as the optimal batch parts replacement scheme for S engines to be replaced.

[0084] Please see also Figure 4 , Figure 4 This is a schematic diagram of a backtracking process provided in an embodiment of this application. Figure 1 Assuming the initial search tree is obtained based on the global conflict matrix and the candidate swap set of 206, as shown below... Figure 4 As shown in the initial search tree 401, there are N candidate swaps, including candidate swaps C1 to C6, i.e., N=6. Node A is the root node, node C5 is the selected node C5, node C5' is the skipped node C5, node C4 is the selected node C4, node C4' is the skipped node C4, node C1' is the skipped node C1, node C3 is the selected node C3, node C3' is the skipped node C3, node C6' is the skipped node C6, node C2 is the selected node C2, and node C2' is the skipped node C2. That is to say, at this time, the order of the N candidate swaps in the candidate swap set 206 is "candidate swap C5, candidate swap C4, candidate swap C1, candidate swap C3, candidate swap C6, candidate swap C2".

[0085] At this point, the computer traverses the initial search tree 401, obtaining the initial batch part replacement schemes as "candidate exchange method C5, candidate exchange method C4, candidate exchange method C3, and candidate exchange method C2". The computer backtracks through the initial search tree 401, backtracking to the first backtracking node, node C2'. At this point, the candidate solutions are "candidate exchange method C5, candidate exchange method C4, and candidate exchange method C3". If the total cyclic optimization amount corresponding to the candidate solution is greater than or equal to the total cyclic optimization amount of the current optimal solution (i.e., the initial batch part replacement scheme), then the candidate solution is taken as the new current optimal solution. If the total cyclic optimization amount corresponding to the candidate solution is less than the total cyclic optimization amount of the current optimal solution (i.e., the initial batch part replacement scheme), then the computer can continue to keep the initial batch part replacement scheme as the current optimal solution.

[0086] Assuming the current optimal solution is still the initial batch parts replacement scheme, the computer device continues to backtrack. At this time, the next unvisited node is node C3', that is, the second backtracking node is node C3'. The computer device can construct a branch subtree formed by the remaining candidate exchange methods corresponding to the second backtracking node, and obtain a branch subtree 402a as shown in the initial search tree 402. Then, based on the candidate solutions formed by the branch subtree 402a, it can be determined whether to update the current optimal solution.

[0087] Furthermore, the computer equipment can backtrack through branch subtree 402a based on the current optimal solution until the optimal batch part replacement scheme corresponding to the second backtracking node is obtained. The optimal batch part replacement scheme corresponding to the second backtracking node is then determined as the current optimal solution corresponding to the second backtracking node. The computer equipment can obtain the following by backtracking through branch subtree 402a: Figure 4 The branch subtree 403a shown in the initial search tree 403.

[0088] Here, it is assumed that candidate swap method C6 is a non-conflicting candidate swap method among the selected options. The computer device can create nodes C6 and C6' for node C3' based on candidate swap method C6. Further assuming that candidate swap method C2 conflicts with one or more of the selected candidate swap methods, node C2' can be created for node C6 based on candidate swap method C2 (i.e., candidate swap method C2 is not selected). This yields the candidate solution (i.e., the updated batch part replacement scheme) corresponding to branch subtree 402a, which is now "candidate swap method C5, candidate swap method C4, candidate swap method C6". If the total loop optimization amount of this candidate solution is less than the total loop optimization amount of the current optimal solution, then the current optimal solution remains unchanged, still being "candidate swap method C5, candidate swap method C4, candidate swap method C3, candidate swap method C2".

[0089] The computer then backtracks to node C6' and creates nodes C2 and C2' for node C6' based on the remaining candidate swapping methods, i.e., candidate swapping method C2. At this point, the second candidate solution may include "candidate swapping method C5, candidate swapping method C4, and candidate swapping method C2". The total loop optimization amount of the second candidate solution is less than the total loop optimization amount of the current optimal solution, and the current optimal solution remains unchanged.

[0090] The computer then backtracks to node C2'. At this point, there are no candidate swaps after candidate swap C2, and the third candidate solution is obtained as "candidate swap C5, candidate swap C4". The total loop optimization amount of the third candidate solution is less than the total loop optimization amount of the current optimal solution, and the current optimal solution remains unchanged.

[0091] In other words, none of the three candidate solutions corresponding to subtree 403a (the first candidate solution corresponding to subtree 403a is the candidate solution corresponding to subtree 402a) can update the current optimal solution. The computer device can continue to backtrack to the third backtracking node, namely node C4'. Based on the candidate solutions corresponding to the subtrees formed by the remaining candidate exchange methods corresponding to node C4', it continues to determine whether to update the current optimal solution until it backtracks to the root node, ends the backtracking process of the initial search tree, and determines the current optimal solution as the optimal batch parts replacement scheme for S engines to be replaced.

[0092] Further, please see Figure 5 , Figure 5 This is a flowchart illustrating a method for optimizing engine parts provided in an embodiment of this application. Figure 2 This method for optimizing engine parts can be executed by computer equipment, such as... Figure 1 The shown business device or server 101 can also be a device composed of any business device and server 101. The engine component optimization method can include at least the following steps S501-S509: Step S501: Obtain S engines with the same engine model to be replaced; S is an integer greater than 1.

[0093] In this embodiment of the application, the specific implementation process of step S501 can be found in the above description. Figure 3 The specific details of step S301 in the embodiments will not be repeated here.

[0094] Step S502: Based on each engine part in the S engines to be replaced, generate a list of candidate replacement methods for each engine to be replaced.

[0095] In this embodiment of the application, the S engines to be replaced include engine A. i , where i is a positive integer less than or equal to S. The computer equipment can handle the replacement of engine A. i The M engine parts are iterated and combined to obtain T part combinations. Here, M and T are both positive integers; a part combination must include at least one of the M engine parts.

[0096] Specifically, computer equipment can be used to replace engine A. i The M engine parts are iterated and combined sequentially to generate a combination containing any one engine part, a combination containing any two engine parts, and so on, until a combination containing all M engine parts is generated, resulting in T combination combinations. Where T=2 M -1. For example, when M=4, the number of combinations of parts including any one engine part is 4, the number of combinations of parts including any two engine parts is 6, the number of combinations of parts including any three engine parts is 4, and the number of combinations of parts including any four engine parts is 1. That is, T parts combinations equal 15 parts combinations (T=2). M -1=2 4 -1=15).

[0097] Optionally, the computer equipment can obtain the engine A to be replaced. i The remaining cycle counts of M engine parts are used to determine the engine part with the smallest remaining cycle count as the priority replacement part. The computer can generate T part combinations based on the priority replacement part, or the priority replacement part combined with other engine parts. Each part combination includes the priority replacement part. Other engine parts refer to engine A to be replaced. i The engine parts in question are M engine parts excluding those that are prioritized for replacement. Taking M engine parts a, b, c, and d as examples, with part a being the priority replacement part (M=4), the T part combinations are (part a), (part a, part b), (part a, part c), (part a, part d), (part a, part b, part c), (part a, part b, part d), (part a, part c, part d), and (part a, part b, part c, part d), resulting in 8 part combinations (T=2). M-1 =2 4-1= 8).

[0098] The computer equipment can generate a sublist of candidate exchange methods for each of the T part combinations based on T part combinations and S-1 remaining engine parts. Here, S-1 remaining engines refer to the S engines to be replaced excluding engine A. i Other engines to be replaced besides those mentioned above. Specifically, T parts assemblies include parts assembly B. j Let j be a positive integer less than or equal to T. The computer device can obtain the parts combination B from each of the S-1 remaining engines. j There are P candidate engine parts with the same part type, which are grouped into S-1 candidate part combinations. Here, P is a positive integer less than or equal to M. The computer device can then determine the part combination B based on this combination. j Combine with candidate part S-1 to generate part combination B. j The corresponding G candidate swaps; G is a positive integer less than or equal to S-1.

[0099] Specifically, take a component combination (E1-a1, E1-b1) as component combination B. j Taking this as an example, the computer equipment can generate S-1 candidate exchange methods corresponding to the part combination (E1-a1, E1-b1) based on the part combination (E1-a1, E1-b1) and parts a and b from the remaining S-1 engines. The S-1 candidate part combinations include (E2-a2, E2-b2), (E3-a3, E3-b3), ..., (ES-aS, ES-bS). The S-1 candidate exchange methods include (E1-a1, E1-b1 exchanged for E2-a2, E2-b2), (E1-a1, E1-b1 exchanged for E3-a3, E3-b3), (E1-a1, E1-b1 exchanged for E4-a4, E4-b4), ..., (E1-a1, E1-b1 exchanged for ES-aS, ES-bS). Furthermore, the computer equipment can process part combination B... j The corresponding S-1 candidate exchange methods are filtered to obtain part combination B. j The corresponding G candidate swapping methods. The filtering process may include, but is not limited to, filtering based on a runnable loop count threshold, and verifying date availability.

[0100] Among them, the runnable cycle number threshold filtering refers to, if part combination B j If P engine parts in the system include engine parts restricted by CSPR (i.e., HPC and CC&HPT), then the computer equipment needs to obtain the engine A to be replaced. iThe operable cycle count threshold (i.e., CSPR limit value) is set. Taking one of the S-1 candidate exchange methods as an example, the computer device can obtain the cumulative number of running cycles of the candidate parts in the candidate exchange method. If the cumulative number of running cycles of one or more candidate parts is greater than or equal to the operable cycle count threshold, the candidate exchange method is deleted from the S-1 candidate exchange methods.

[0101] Among them, date availability verification refers to the process where, if the part reference date (as_of_date) of a candidate part in a candidate exchange method is greater than the date of the engine A to be replaced, then... i If the engine reference date (engine.as_of_date) is specified, then the candidate exchange method is removed from the S-1 candidate exchange methods.

[0102] It should be noted that the reference date is used to indicate that the status or data of the engine parts / engine is valid on or before this date. Screening candidate exchange methods that have passed date availability verification is to ensure that the reference date of the candidate parts selected for that exchange method is no later than the reference date of the corresponding engine to be replaced, thereby avoiding matching an engine part in a "future state" with an engine in a "past state".

[0103] For example, engine A to be replaced i The engine reference date is May 10, xx year. The reference date of the candidate part in one candidate exchange method is May 8, xx year. The reference date of the candidate part in another candidate exchange method is May 12, xx year. The computer equipment can delete the candidate exchange method with the reference date of May 12, xx year and determine the candidate exchange method with the reference date of May 8, xx year as one of the G candidate exchange methods.

[0104] In other words, component assembly B j The cumulative number of running cycles for each candidate part in the corresponding G candidate exchange methods is less than the threshold for the number of running cycles, and the part reference date for each candidate part in each candidate exchange method is less than or equal to the engine A to be replaced. i The engine's base date.

[0105] Computer equipment can assemble parts B j The corresponding G candidate exchange methods are determined to be part combination B. j The corresponding candidate exchange method sublist. Then, the T parts can be combined to form the engine A to be replaced, corresponding to the candidate exchange method sublist. i The corresponding list of candidate swapping methods.

[0106] The candidate exchange method list is used to indicate information on replacing one or more engine parts in an engine to be replaced.

[0107] Step S503: Based on each candidate exchange method in the list of S candidate exchange methods, create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods.

[0108] In this embodiment of the application, the specific implementation process of step S503 can be found in the above description. Figure 3 The specific description of step S303 in the embodiments will not be repeated here.

[0109] Step S504: Obtain the minimum loop optimization amount corresponding to each candidate exchange method in the S candidate exchange method list. Sort each candidate exchange method in the S candidate exchange method list in descending order of the minimum loop optimization amount of each candidate exchange method to obtain the candidate exchange method set.

[0110] In this embodiment of the application, the list of S candidate swapping methods includes N candidate swapping methods, where N is a positive integer; the N candidate swapping methods include candidate swapping method C. k k is a positive integer less than or equal to N. The computer device can obtain candidate swapping methods C. k The minimum remaining cycle count among all remaining cycle counts of each engine part in the target engine to be replaced is determined as the first minimum remaining cycle count of the target engine to be replaced. Here, the S engines to be replaced include the target engine to be replaced; that is, the candidate replacement method C... k The corresponding engine to be replaced is called the target engine to be replaced. Among them, candidate replacement method C... k The corresponding engine to be replaced refers to candidate replacement method C. k The source engine, i.e., the engine to be replaced that receives engine parts, is the candidate exchange method C. k If it is represented as "part E2-a2 replaces part E1-a1", then the target engine to be replaced is engine E2.

[0111] Computer devices can obtain candidate switching methods C k The remaining cycle number for each candidate part. Among them, candidate exchange method C... k The candidate parts are selected through candidate exchange method C. k The candidate component combination is used to determine this. For example, if candidate exchange method C... k This is represented as "exchanging parts E2-a2, E2-b2 for parts E1-a1, E1-b1", with candidate exchange method C.k The corresponding candidate part combination is (E1-a1, E1-b1), and the candidate exchange method is C. k The candidate parts are part E1-a1 (remaining cycles: 1000) and part E1-b1 (remaining cycles: 1200). The computer device can obtain the remaining cycles for part E1-a1 and part E1-b1 respectively.

[0112] Computer devices can obtain candidate switching methods C k The remaining cycle number for each unreplaced part. Here, unreplaced parts refer to all parts of the target engine to be replaced, excluding candidate exchange method C. k Engine parts other than the released parts; Candidate exchange method C k The released part is determined based on one of the T part combinations corresponding to the target engine to be replaced; candidate exchange method C k Release components and candidate exchange methods C k The candidate parts in the list have the same part type. Continuing with the example above, candidate exchange method C... k This is represented as "exchanging parts E2-a2, E2-b2 for parts E1-a1, E1-b1", where the released parts include parts E2-a2 and E2-b2, and the candidate exchange method is C. k The corresponding unreplaced parts include the two remaining engine parts in engine E2, namely part E2-c2 (remaining cycle count: 1000) and part E2-d2 (remaining cycle count: 2000). In other words, the computer equipment can obtain the remaining cycle counts for part E2-c2 and part E2-d2 respectively.

[0113] The computer equipment can determine the remaining cycle number for each candidate part and the remaining cycle number for each unreplaced part as a set of remaining cycle numbers. The minimum value in the set of remaining cycle numbers is determined as the second minimum remaining cycle number of the target engine to be replaced. Continuing with the example above, the set of remaining cycle numbers now includes the remaining cycle numbers for the four engine parts E1-a1, E1-b1, E2-c2, and E2-d2, represented as (1000, 1200, 1000, 2000). In this case, the minimum value in the set of remaining cycle numbers is 1000, which is the second minimum remaining cycle number.

[0114] The computer device can determine the difference between the first minimum remaining cycle number and the second minimum remaining cycle number as the candidate swap method C. k The corresponding minimum loop optimization amount. If candidate swap method C k The corresponding first minimum remaining cycle number is 500, then candidate swap method Ck The corresponding minimum loop optimization amount is 1000-500=500.

[0115] Furthermore, when the computer device obtains the minimum cycle optimization amount corresponding to each of the N candidate swapping methods, it can sort the N candidate swapping methods in descending order of their minimum cycle optimization amounts to obtain a set of candidate swapping methods.

[0116] Step S505: Construct an initial search tree by using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. Traverse the initial search tree. When the traversal reaches the first leaf node in the initial search tree, generate an initial batch part replacement plan based on the node status information of the first leaf node.

[0117] In this embodiment of the application, the candidate exchange method set includes N candidate exchange methods, where N is a positive integer. The computer device can create a root node with initial node state information; based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set, a w-th layer child node is created for the (w-1)-th layer target node; when the N-th layer child node is obtained, an initial search tree can be formed based on the nodes between the root node and the N-th layer child node.

[0118] The initial node state information refers to the node state information corresponding to the root node. One possible way to represent the initial node state information is as shown above: "{swap[],used{},avail{},improvement:0,index:0}".

[0119] Where w is a positive integer less than or equal to N; multiple target logical relationships include the logical relationship between the w-th candidate exchange method and each candidate exchange method in the set of w-1 selected exchange methods; when w is 1, the w-1 level target node is the root node; the w-th level child node is the child node of the w-1 level target node, and the w-th level child node includes a selection node for indicating the selection of the w-th candidate exchange method, and / or a skip node for indicating skipping the w-th candidate exchange method; the w-1 level candidate exchange method in the set of candidate exchange methods has a corresponding relationship with the w-1 level target node; the w-1 level target node is determined in the w-1 level child nodes; the w-1 level selected exchange method set includes the candidate exchange methods selected in the node path from the root node to the w-1 level target node; w is a positive integer less than or equal to N.

[0120] The specific implementation process of the computer device creating a layer-w child node based on multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set can be as follows: The computer device can determine the layer-w-1 target node from the layer-w-1 child nodes. If the layer-w-1 child nodes include a w-1 selection node and a w-1 skip node, the computer device can determine the w-1 selection node as the layer-w-1 target node. Optionally, if the layer-w-1 child nodes only include a w-1 skip node, the computer device can determine the w-1 skip node as the layer-w-1 target node. Figure 4 As shown in the initial search tree 401, when the (w-1)th level child node is a level 1 child node (i.e., w=2), the level 1 child nodes include node C5 (i.e., the selected node C5, which is the first selected node) and node C5' (i.e., the skipped node C5, which is the first skipped node). At this time, the computer device can determine node C5 as the level 1 target node. When the (w-1)th level child node is a level 3 child node (i.e., w=4), the level 3 child nodes only include node C1' (i.e., the skipped node C1, which is the third skipped node). At this time, the computer device can determine node C1' as the level 3 target node.

[0121] It should be noted that if the target node at level (w-1) is of the selection type, meaning it is the w-1th selected node, then the set of selected swap methods includes the w-1th candidate swap method. Optionally, if the target node at level (w-1) is of the skip type, meaning it is the w-1th skipped node, then the set of selected swap methods does not include the w-1th candidate swap method. For example... Figure 4 As shown in the initial search tree 401, the target node of the second layer is the second selected node, i.e., node C4. At this time, the (w-1)th selected swap set is the second selected swap set, which is the candidate swap set selected in the node path from the root node to the target node of the second layer (node ​​C4), including "candidate swap C5" and "candidate swap C4". Assume that the target node of the second layer is the second skipped node, i.e., node C4' (i.e., assume node C4 does not exist). At this time, the (w-1)th selected swap set is the second selected swap set, which is the candidate swap set selected in the node path from the root node to the target node of the second layer (node ​​C4'), including "candidate swap C5" but excluding candidate swap C4.

[0122] Furthermore, if the logical relationships between multiple targets in the global conflict matrix are all compatible, that is, if the engine parts for each candidate exchange method in the (w-1)th selected exchange method set do not include the engine parts for the w-th candidate exchange method, then the computer device can create a w-th selection node for indicating the selection of the w-th candidate exchange method and a w-th skip node for indicating skipping the w-th candidate exchange method, based on the w-th candidate exchange method in the candidate exchange method set. The w-th selection node and the w-th skip node are then determined as the w-th layer child nodes. The w-1st layer target node is used as... Figure 4 Taking node C1' as an example (where w=4), the multiple target logical relationships include the logical relationship between the fourth candidate exchange method (i.e., candidate exchange method C3) and each candidate exchange method in the third selected exchange method set (including candidate exchange method C5 and candidate exchange method C4). If all the target logical relationships are compatible, that is, if the engine parts in candidate exchange methods C5 and C4 are represented as used{E3-a3,E4-a4,E3-b3,E6-b6}, then the fourth candidate exchange method (i.e., candidate exchange method C3) is represented as "E2-d2 exchanged for E5-d5". The used{E3-a3,E4-a4,E3-b3,E6-b6} does not include E2-d2 and E5-d5. In other words, at this time, candidate exchange method C3 can be selected. The computer device can create a fourth selection node (i.e., node C3) for the target node (i.e. node C1') of the third layer (i.e. node C1') to indicate the selection of the fourth candidate exchange method (i.e., candidate exchange method C3), and a fourth skip node (i.e. node C3') to indicate skipping the fourth candidate exchange method. The fourth selection node and the fourth skip node are determined as the fourth layer child nodes.

[0123] Optionally, if one or more of the target logical relationships in the global conflict matrix are mutually exclusive, the computer device can detect the first set of released part identifiers in the node status information of the (w-1)th layer target node. If the first set of released part identifiers is found to include the part identifier of a candidate part in the w-th candidate exchange method, the computer device can create a w-th selection node for the w-1th layer target node to indicate the selection of the w-th candidate exchange method, and a w-th skip node for indicating skipping the w-th candidate exchange method. The w-th selection node and the w-th skip node are then identified as w-th layer child nodes. Here, the first set of released part identifiers refers to the set of released part identifiers in the node status information of the (w-1)th layer target node.

[0124] Continuing with the target node at layer w-1 as Figure 4Taking node C1' as an example (where w=4), multiple target logical relationships exist, including the logical relationship between the fourth candidate exchange method (i.e., candidate exchange method C3) and each candidate exchange method in the third selected exchange method set (including candidate exchange methods C5 and C4). In this case, if one or more target logical relationships are mutually exclusive—that is, candidate exchange method C5 and candidate exchange method C3 are mutually exclusive, or candidate exchange method C4 and candidate exchange method C3 are mutually exclusive, or candidate exchange method C4 and candidate exchange method C3, and candidate exchange method C5 and candidate exchange method C3 are all mutually exclusive—then... In other words, if the engine parts in candidate exchange method C5 and candidate exchange method C4 are represented as used{E3-a3,E4-a4,E3-b3,E6-b6}, and assuming that the fourth candidate exchange method (i.e., candidate exchange method C3) is represented as "E2-a2 exchange E3-a3" (i.e., candidate exchange method C5 and candidate exchange method C3 are mutually exclusive), and used{E3-a3,E4-a4,E3-b3,E6-b6} includes the engine parts in candidate exchange method C3, such as E3-a3, then the computer device can detect the first set of released part identifiers in the node status information of the target node (i.e., node C1') at layer w-1. At this time, the first set of released part identifiers is represented as "avail{E3-a3,E3-b3}". The computer device can detect that the part identifier of the candidate part in the fourth candidate exchange method is included in the first set of released part identifiers, that is, “avail{E3-a3,E3-b3}” includes “E3-a3”. In other words, the candidate exchange method C3 can be selected at this time. Then, the computer device can create a fourth selection node for node C1' to indicate the selection of candidate exchange method C3, and a fourth skip node to indicate skipping candidate exchange method C3. The fourth selection node and the fourth skip node are determined as the fourth-level child nodes, that is, node C3 and node C3' in the initial search tree 401.

[0125] Further optional, if the first set of released part identifiers does not include the part identifier of the candidate part in the w-th candidate exchange method, then according to the w-th candidate exchange method in the set of candidate exchange methods, a w-th skip node is created for the target node of the (w-1)-th layer to indicate skipping the w-th candidate exchange method, and the w-th skip node is determined as the w-th layer child node. For example, suppose the fourth candidate exchange method (i.e., candidate exchange method C3) is represented as "E3-a3 exchange for E5-a5" (i.e., candidate exchange method C5 and candidate exchange method C3 are mutually exclusive). The used{E3-a3,E4-a4,E3-b3,E6-b6} includes the engine parts in candidate exchange method C3, such as E3-a3. At this time, the computer device can detect that the first release part identifier set "avail{E3-a3,E3-b3}" does not include the part identifier of the candidate parts in candidate exchange method C3, i.e., E5-a5. That is to say, candidate exchange method C3 cannot be selected at this time. Then, the computer device can create a fourth skip node to indicate skipping candidate exchange method C3, which can be determined as the fourth level child node. That is, at this time, only node C3' should exist in the initial search tree 401, and node C3 should not exist.

[0126] The specific implementation process of creating the w-th selection node for indicating the selection of the w-th candidate exchange method can be as follows: The computer device can combine each candidate exchange method in the (w-1)-th selected exchange method set with the w-th candidate exchange method to form the w-th selected exchange method set; based on each candidate exchange method in the w-th selected exchange method set, it can generate a used part identifier set for indicating used engine parts and a released part identifier set for indicating released parts. Specifically, the computer device can add the engine parts of each candidate exchange method in the w-th selected exchange method set to the initial used part identifier set, i.e., "used{}", to obtain the used part identifier set for indicating used engine parts. The computer device can add the released parts of each candidate exchange method in the w-th selected exchange method set to the initial released part identifier set, i.e., "avail{}", to obtain the released part identifier set for indicating released parts. Alternatively, the computer device can directly inherit the set of used part identifiers from the node status information of the (w-1)th layer target node, add the engine part from the w-th candidate exchange method to the set of used part identifiers in the node status information of the (w-1)th layer target node, and obtain the set of used part identifiers for the w-th selected node to be created. Similarly, the set of released part identifiers for the w-th selected node to be created can be determined based on the set of released part identifiers in the node status information of the (w-1)th layer target node, which will not be elaborated here.

[0127] Through the above steps, the cascaded interchange of engine parts can be realized, thereby improving the utilization rate of engine parts and avoiding resource idleness. Cascaded interchange allows the use of released modules in the set of released part identifiers, which greatly expands the range of candidate exchange methods. It enables candidate exchange methods that could not be executed due to engine parts being occupied to be implemented, making it easier to find the optimal batch part replacement scheme with higher benefits.

[0128] Furthermore, the computer device can determine the cumulative loop optimization amount corresponding to selecting the w-th candidate exchange method based on each candidate exchange method in the w-th selected exchange method set. Specifically, the computer device can determine the cumulative loop optimization amount corresponding to selecting the w-th candidate exchange method as the sum of the minimum loop optimization amounts of each candidate exchange method in the w-th selected exchange method set. The computer device can obtain the index position of the w-th candidate exchange method in the candidate exchange method set, determine the search tree depth based on the index position of the w-th candidate exchange method; combine the used part identifier set, the released part identifier set, the cumulative loop optimization amount corresponding to selecting the w-th candidate exchange method, the index position of the w-th candidate exchange method, and the search tree depth to form the first node status information; and create the w-th selection node based on the first node status information.

[0129] The set of used part identifiers includes the part identifiers of engine parts for each candidate exchange method in the w-th selected exchange method set; the set of released part identifiers includes the part identifiers of released parts for each candidate exchange method in the w-th selected exchange method set; and the engine parts in the w-th candidate exchange method include the released parts in the w-th candidate exchange method.

[0130] Continuing with the example of candidate exchange methods C5 and C4, where the engine parts are represented as used{E3-a3,E4-a4,E3-b3,E6-b6} and candidate exchange method C3 is represented as "E2-a2 for E3-a3", the computer device can generate the following: The w-th selected exchange method set, composed of the (w-1)-th selected exchange method set and the w-th candidate exchange method, includes "candidate exchange method C5, candidate exchange method C4, and candidate exchange method C3". The set of used part identifiers used to indicate used engine parts can be represented as "used{E3-a3,E4-a4,E3-b3,E6-b6,E2-a2}"; the generated set of release part identifiers used to indicate released parts changes from "avail{E3-a3,E3-b3}" to "avail{E3-b3,E2-a2}", meaning that part E3-a3 is used by engine E2 corresponding to candidate exchange method C3, and engine E2 releases part E2-a2. In other words, the released parts released by candidate exchange method C5 can be used by candidate exchange method C3, thus achieving cascaded exchange. At this point, the cumulative loop optimization amount corresponding to the w-th candidate exchange method is the sum of the minimum loop optimization amounts corresponding to the three candidate exchange methods "candidate exchange method C5, candidate exchange method C4, and candidate exchange method C3" (assumed to be 5000).

[0131] The index positions in the candidate swap set can start from 0, meaning the index position of the first candidate swap is 0, i.e., index[0]. The initial search tree depth can also start from 0, meaning the search tree depth of the root node is 0. In other words, taking the example above, the index position of candidate swap C5 in the candidate swap set is index[0], and the index position of candidate swap C3 in the candidate swap set is index[3]. The search tree depth corresponding to candidate swap C3 is also 3.

[0132] It is understandable that the w-th selected node, which is currently the 4th selected node, can have its corresponding first node state information represented as "{swap[C5,C4,C3],used{E3-a3,E4-a4,E3-b3,E6-b6,E2-a2},avail{E3-b3,E2-a2},improvement:5000,index:3,depth:3}". Here, depth:3 indicates the depth of the search tree.

[0133] The specific implementation process of the computer device creating the w-th skip node to indicate skipping the w-th candidate exchange method can be as follows: the computer device can obtain the node status information of the target node at the (w-1)-th layer; obtain the index position of the w-th candidate exchange method in the candidate exchange method set; determine the search tree depth based on the index position of the w-th candidate exchange method; update the node status information of the target node at the (w-1)-th layer based on the index position of the w-th candidate exchange method and the search tree depth to obtain the second node status information; and create the w-th skip node based on the second node status information. If the node state information of the target node at level w-1 is represented as “{swap[C5,C4],used{E3-a3,E4-a4,E3-b3,E6-b6},avail{E3-a3,E3-b3},improvement:3800,index:2,depth:2}”, then the state information of the second node can be represented as “{swap[C5,C4],used{E3-a3,E4-a4,E3-b3,E6-b6},avail{E3-a3,E3-b3},improvement:3800,index:3,depth:3}”, where the index positions and search tree depths are different.

[0134] Step S506: Obtain the total loop optimization amount corresponding to the initial batch parts replacement scheme; obtain the timeout threshold corresponding to the S engines to be replaced; in the initial search tree, obtain the r-th backtracking node and obtain the search execution time corresponding to the r-th backtracking node.

[0135] In this embodiment, the computer device can determine the total cyclic optimization amount corresponding to the initial batch parts replacement scheme by summing the minimum cyclic optimization amounts of each candidate exchange method selected in the initial batch parts replacement scheme. The computer device can obtain the timeout thresholds corresponding to the S engines to be replaced. An optional method for determining the timeout thresholds can be found in formula (1): Formula (1) As shown in formula (1), N is the number of candidate exchange methods in the candidate exchange method set. This is used to indicate the timeout threshold, and its unit can be milliseconds (ms). That is, when the number of candidate exchange methods in the candidate exchange method set is greater than 60,000, the corresponding timeout threshold is 600,000 ms, or 10 minutes (min). In other words, the search execution time will not exceed 10 minutes. hour, It is determined based on the value of N. When When the timeout threshold is 120,000ms, or 2 minutes (min), the minimum search execution time is 2 minutes.

[0136] Furthermore, the computer device can obtain the r-th backtracking node in the initial search tree. Specifically, if the (r-1)-th backtracking node has an unvisited sibling node, the computer device can identify that unvisited sibling node as the r-th backtracking node. If the (r-1)-th backtracking node does not have an unvisited sibling node, the computer device can backtrack to the parent node of the (r-1)-th backtracking node and identify its unvisited sibling node as the r-th backtracking node. If the parent node of the (r-1)-th backtracking node also does not have an unvisited sibling node, the backtracking continues upwards until an unvisited sibling node is obtained, which is then identified as the r-th backtracking node. Here, an unvisited sibling node refers to a sibling node of the (r-1)-th backtracking node that has not yet been considered or evaluated for feasibility and effectiveness in the current search path (i.e., the currently constructed solution).

[0137] Please see again Figure 4 In the initial search tree 401, when r=1, the (r-1)th backtracking node is the first leaf node in the initial search tree 401, i.e., node C2. At this time, node C2 has an unvisited sibling node, i.e., node C2'. Therefore, the computer device can determine node C2' as the r-th backtracking node, i.e., the first backtracking node is node C2'. Further, when r=2, the (r-1)th backtracking node is node C2', and node C2''s sibling node is node C2. Node C2 has been visited, i.e., node C2' does not have an unvisited sibling node. The computer device can backtrack to the parent node of node C2', i.e., node C6'. If the parent node of node C2' (i.e., node C6') has an unvisited sibling node, the computer device can determine the unvisited sibling node of the parent node of node C2' as the second backtracking node. Since node C6' in the initial search tree 401 has no unvisited sibling node, the computer device can continue to backtrack to the parent node of node C6'. At this time, the parent node of node C6' (i.e., node C3) has a sibling node, i.e., node C3'. The computer device can identify node C3' as the second backtracking node.

[0138] It should be noted that the remaining candidate exchange methods corresponding to the r-th backtracking node refer to the candidate exchange methods in the candidate exchange method set whose index position is located after the index position in the node state information of the r-th backtracking node; when r=1, the (r-1)-th backtracking node is the first leaf node in the initial search tree, the current optimal loop optimization quantity corresponding to the (r-1)-th backtracking node is the total loop optimization quantity, and the current optimal solution corresponding to the (r-1)-th backtracking node is the initial batch part replacement scheme; when r is greater than 1, the r-th backtracking node is determined based on the (r-1)-th backtracking node; r is a positive integer.

[0139] Once the computer device obtains the r-th backtracking node, it can acquire the search execution time corresponding to that node, which is the current system time and the execution start time. The time difference between the current system time and the execution start time is determined as the search execution time corresponding to the r-th backtracking node. The execution start time refers to the time at which the initial search tree is constructed using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set.

[0140] Step S507: If the search execution time is less than or equal to the timeout threshold, then generate the upper bound pruning threshold and the remaining potential pruning threshold based on the node state information corresponding to the r-th backtracking node and the remaining candidate exchange method corresponding to the r-th backtracking node.

[0141] In this embodiment, if the search execution time is less than or equal to the timeout threshold (i.e., no execution timeout), the computer device can filter the remaining candidate exchange methods corresponding to the r-th backtracking node based on the selected exchange method set, the released part identifier set, and the global conflict matrix in the node status information corresponding to the r-th backtracking node, obtaining Q remaining candidate exchange methods. Here, Q is a positive integer. Specifically, the computer device can obtain H remaining candidate exchange methods corresponding to the r-th backtracking node. If, in the global conflict matrix, any one of the H remaining candidate exchange methods has a mutually exclusive logical relationship with one or more candidate exchange methods in the selected exchange method set corresponding to the r-th backtracking node, and there is no part identifier for a candidate part in that remaining candidate exchange method in the released part identifier set corresponding to the r-th backtracking node, then the computer device can delete that remaining candidate exchange method from the H remaining candidate exchange methods, obtaining Q remaining candidate exchange methods. Here, H is an integer greater than or equal to Q.

[0142] For example, the H remaining candidate exchange methods include 3 remaining candidate exchange methods, namely candidate exchange method 1, candidate exchange method 2 and candidate exchange method 3. Assume that the logical relationship between candidate exchange method 1 and one or more candidate exchange methods in the selected exchange method set corresponding to the r-th backtracking node is mutually exclusive, and there is no part identifier of the candidate part in candidate exchange method 1 in the release part identifier set corresponding to the r-th backtracking node. The computer device can delete candidate exchange method 1 from the H remaining candidate exchange methods to obtain Q remaining candidate exchange methods. That is, at this time, the Q remaining candidate exchange methods include candidate exchange method 2 and candidate exchange method 3.

[0143] Furthermore, the computer equipment can group the Q remaining candidate exchange methods according to the engine identifiers of the engines to be replaced, resulting in Y groups of remaining candidate exchange methods. The computer equipment can obtain the minimum loop optimization amount corresponding to each of the Q remaining candidate exchange methods. Based on the minimum loop optimization amount of the remaining candidate exchange methods in each group, it can obtain the minimum loop optimization amount with the largest value from each of the Y remaining candidate exchange method groups. Based on the sum of the Y minimum loop optimization amounts with the largest values ​​and the cumulative loop optimization amount in the node state information corresponding to the r-th backtracking node, an upper bound pruning threshold is generated.

[0144] Where Y is a positive integer less than or equal to Q; the remaining candidate exchange methods in each remaining candidate exchange method group correspond to the same engine identifier.

[0145] Continuing with the aforementioned Q remaining candidate swapping methods, including candidate swapping method 2 and candidate swapping method 3, assuming the engine identifier of the engine to be replaced corresponding to candidate swapping method 2 is X-01, and the engine identifier of the engine to be replaced corresponding to candidate swapping method 3 is X-02, meaning candidate swapping method 2 and candidate swapping method 3 correspond to two different engines to be replaced, then there are Y remaining candidate swapping method groups, forming two groups. Each remaining candidate swapping method group includes one candidate swapping method. The minimum loop optimization amount with the largest value corresponding to one remaining candidate swapping method group is the minimum loop optimization amount of candidate swapping method 2, and the minimum loop optimization amount with the largest value corresponding to the other remaining candidate swapping method group is the minimum loop optimization amount of candidate swapping method 3. Further assuming the cumulative loop optimization amount in the node state information corresponding to the r-th backtracking node is 2000, the minimum loop optimization amount of candidate swapping method 2 is 500, and the minimum loop optimization amount of candidate swapping method 3 is 400, then the upper bound pruning threshold is 2000 + 500 + 400 = 2900.

[0146] Assume that the engine identifier for the engine to be replaced corresponding to candidate swap method 2 is X-01, and the engine identifier for the engine to be replaced corresponding to candidate swap method 3 is also X-01. That is, candidate swap method 2 and candidate swap method 3 correspond to the same engine to be replaced. In this case, the Y remaining candidate swap method groups form one remaining candidate swap method group, including candidate swap method 2 and candidate swap method 3. The minimum cycle optimization amount with the largest value in this remaining candidate swap method group is the minimum cycle optimization amount for candidate swap method 2, which is 500. The upper bound pruning threshold is then 2000 + 500 = 2500.

[0147] Furthermore, the computer device can generate the remaining potential pruning threshold based on the sum of the Q minimum cyclic optimization quantities and the cumulative cyclic optimization quantity in the node state information corresponding to the r-th backtracking node. That is, the remaining potential pruning threshold is determined by adding the sum of the Q minimum cyclic optimization quantities to the sum of the cumulative cyclic optimization quantities in the node state information corresponding to the r backtracking nodes. Continuing with the example of the Q remaining candidate swapping methods including candidate swapping method 2 and candidate swapping method 3, the remaining potential pruning threshold is now 2000 + 500 + 400 = 2900.

[0148] Optionally, if the search execution time exceeds the timeout threshold, the backtracking process is stopped, and subsequent steps S508 and S509 are not executed. The computer device can directly determine the current optimal solution corresponding to the (r-1)th backtracking node as the optimal batch parts replacement scheme for the S engines to be replaced.

[0149] Step S508: If the current optimal loop optimization amount corresponding to the r-th backtracking node is less than the upper bound pruning threshold and less than the remaining potential pruning threshold, then according to the remaining candidate exchange method corresponding to the r-th backtracking node, generate the branch subtree of the r-th backtracking node. According to the branch subtree of the r-th backtracking node, update the current optimal solution corresponding to the (r-1)-th backtracking node to obtain the current optimal solution corresponding to the r-th backtracking node. The loop optimization amount of the current optimal solution corresponding to the r-th backtracking node is determined as the current optimal loop optimization amount corresponding to the r-th backtracking node.

[0150] In this embodiment, if the current optimal loop optimization amount corresponding to the r-th backtracking node is less than the upper bound pruning threshold and less than the remaining potential pruning threshold, the computer device can generate a branch subtree of the r-th backtracking node according to the remaining candidate swapping method corresponding to the r-th backtracking node. The specific implementation process of the computer device generating the branch subtree of the r-th backtracking node can refer to the specific implementation process of generating the initial search tree, and will not be repeated here.

[0151] Furthermore, the computer device can update the current optimal solution corresponding to the (r-1)th backtracking node based on the branch subtree of the r-th backtracking node, thus obtaining the current optimal solution corresponding to the r-th backtracking node. Specifically, the computer device can traverse the branch subtree of the r-th backtracking node. When traversing to the first leaf node in the branch subtree of the r-th backtracking node, an updated batch parts replacement plan is generated based on the node state information of the first leaf node in the branch subtree of the r-th backtracking node. Then, based on the updated batch parts replacement plan, the branch subtree of the r-th backtracking node is backtracked until the optimal batch parts replacement plan corresponding to the r-th backtracking node is obtained. The optimal batch parts replacement plan corresponding to the r-th backtracking node is then determined as the current optimal solution corresponding to the r-th backtracking node.

[0152] Please see also Figure 6 , Figure 6 This is a schematic diagram of a backtracking process provided in an embodiment of this application. Figure 2 ,like Figure 6 As shown, subtree 403a is the subtree of the second backtracking node. Subtree 403a can be traversed. When the traversal reaches the first leaf node in the subtree of the second backtracking node (i.e., the first node C2' from left to right in subtree 403a), an updated batch parts replacement plan is generated based on the node state information of the first leaf node in subtree 403a. Specifically, the computer device can determine the candidate exchange methods selected from the root node to the first leaf node in subtree 403a as the current candidate replacement plan. That is, the current candidate replacement plan includes "candidate exchange method C5, candidate exchange method C4, and candidate exchange method C6". The computer device can obtain the total loop optimization amount of the current candidate replacement plan. If the total loop optimization amount of the current candidate replacement plan is greater than or equal to the current optimal solution corresponding to the r-th backtracking node (which is the current optimal solution corresponding to the 1st backtracking node, i.e., the initial batch parts replacement plan), then the computer device can determine the current candidate replacement plan as the updated batch parts replacement plan. If the total loop optimization amount of the current candidate replacement scheme is less than the current optimal solution corresponding to the (r-1)th backtracking node (which is the current optimal solution corresponding to the 1st backtracking node, i.e. the initial batch part replacement scheme), then the computer equipment can determine the initial batch part replacement scheme as the updated batch part replacement scheme.

[0153] Furthermore, the computer equipment can backtrack the branch subtree of the r-th backtracking node (where r=2) based on the updated batch parts replacement plan, until the optimal batch parts replacement plan corresponding to the r-th backtracking node is obtained. This optimal batch parts replacement plan is then determined as the current optimal solution corresponding to the r-th backtracking node. The specific implementation process of backtracking the branch subtree of the r-th backtracking node can be found in the detailed description of backtracking the initial search tree, and will not be repeated here.

[0154] Understandably, if the total loop optimization amount of the three current candidate replacement schemes corresponding to branch subtree 403a is less than the total loop optimization amount of the current optimal solution corresponding to the (r-1)th backtracking node, it means that none of the three current candidate replacement schemes corresponding to branch subtree 403a is better than the current optimal solution corresponding to the (r-1)th backtracking node. In this case, the computer device can keep the current optimal solution corresponding to the (r-1)th backtracking node unchanged, that is, the current optimal solution corresponding to the rth backtracking node is the same as the current optimal solution corresponding to the (r-1)th backtracking node.

[0155] Furthermore, the computer device continues to backtrack through the initial search tree 403 to obtain the third backtracking node, namely node C4'. Based on the remaining candidate swapping methods corresponding to the third backtracking node, the computer device generates a branch subtree for the third backtracking node, resulting in... Figure 6 The initial search tree 601 shows a branch subtree 601a. ​​The computer traverses branch subtree 601a to obtain the updated batch part replacement scheme corresponding to the third backtracking node, including the candidate swapping methods selected from node C2' in branch subtree 601a to the root node. The computer can backtrack through branch subtree 601a based on the updated batch part replacement scheme corresponding to the third backtracking node until the optimal batch part replacement scheme corresponding to the third backtracking node is obtained. This optimal batch part replacement scheme is then determined as the current optimal solution corresponding to the third backtracking node.

[0156] It should be noted that when backtracking to node 601b in subtree 601a, if the current optimal cycle optimization amount corresponding to the third backtracking node is greater than or equal to the upper bound pruning threshold of node 601b, and is greater than or equal to the remaining potential pruning threshold of node 601b, then the computer can directly continue to obtain the current optimal cycle optimization amount corresponding to the fourth backtracking node. That is, the subtree obtained after backtracking the subtree of the third backtracking node can be as shown in subtree 602a in the initial search tree 602. The computer device does not continue to create child nodes for node 601b, that is, it performs pruning on the subtree under node 601b.

[0157] When backtracking to node 601b in subtree 601a, if the current optimal loop optimization amount corresponding to the third backtracking node is less than the upper bound pruning threshold of node 601b and less than the remaining potential pruning threshold of node 601b, the computer device continues to generate subtrees of node 601b and continues to traverse and backtrack the subtrees of node 601b until the optimal batch part replacement scheme corresponding to the third backtracking node is obtained. The optimal batch part replacement scheme corresponding to the third backtracking node is determined as the current optimal solution corresponding to the third backtracking node. At this time, the subtree obtained by the computer device after backtracking the subtree of the third backtracking node can be shown as subtree 603a in the initial search tree 603.

[0158] In this manner, the computer device continues to backtrack the initial search tree 603, that is, continues to obtain the current optimal loop optimization quantity corresponding to the next backtracking node, until the backtracking process of the initial search tree 603 is completed, which will not be elaborated here.

[0159] Optionally, if the current optimal loop optimization value corresponding to the r-th backtracking node is greater than or equal to the upper bound pruning threshold, and greater than or equal to the remaining potential pruning threshold, then the computer device continues to obtain the current optimal loop optimization value corresponding to the (r+1)-th backtracking node. Please also refer to... Figure 4 ,like Figure 4 As shown, if the current optimal loop optimization value corresponding to the second backtracking node (i.e., node C3') is greater than or equal to the upper bound pruning threshold corresponding to the second backtracking node, and greater than or equal to the remaining potential pruning threshold corresponding to the second backtracking node, the computer device can directly backtrack to the third backtracking node, i.e., node C4', and continue to obtain the current optimal loop optimization value corresponding to the third backtracking node. In other words, there should be no branch subtree 402a in the initial search tree 402.

[0160] Through the above steps, the upper bound pruning process of the initial search tree is implemented based on the upper bound pruning threshold, and the remaining potential pruning process is implemented based on the remaining potential pruning threshold. Search branches that cannot possibly be better than the current optimal solution are eliminated in advance, thereby reducing the search space of the initial search tree, avoiding invalid backtracking, significantly reducing the number of search nodes, and lowering the algorithm time and computational resource consumption. Furthermore, an optimistic estimation method is used to construct the upper bound pruning threshold and the remaining potential pruning threshold. These thresholds are always no less than the actual obtainable benefits, ensuring a safe and conservative pruning process and preventing the erroneous pruning of branches containing the optimal solution. This improves the efficiency of generating the optimal batch part replacement scheme without sacrificing solution accuracy.

[0161] The computer device can determine the cyclic optimization amount of the current optimal solution corresponding to the r-th backtracking node as the current optimal cyclic optimization amount corresponding to the r-th backtracking node.

[0162] Optionally, the computer device can also generate a unique node state key for each node in the initial search tree and cache the node state key of each node in the initial search tree to a node state key set. A node's node state key can be determined by the set of used part identifiers and the set of released part identifiers in the node state information, as well as the index position of the candidate swap method corresponding to the node in the candidate swap method set. An optional representation of the node state key can be "stateKey=`${startIndex}:used[${usedModuleIds}]:avail[${availableModuleIds}]". Here, stateKey is the node state key corresponding to the node, startIndex is the index position of the candidate swap method corresponding to the node, usedModuleIds indicates the set of used part identifiers in the node state information, and availableModuleIds indicates the set of released part identifiers in the node state information. Optionally, the computer device can also concatenate the set of used part identifiers, the set of released part identifiers, and the index position of the candidate exchange method corresponding to the node in the candidate exchange method set in the node status information to obtain concatenated information, and perform hash processing on the concatenated information to obtain the node status key.

[0163] During the backtracking process of the initial search tree, when the computer device backtracks to a target node, if the target node's state key matches a candidate node's state key cached in the node state key set, the computer device can directly obtain the traversal data and backtracking data of each child node under the node corresponding to the candidate node's state key. Based on this data, the computer device generates the current optimal solution for the target node and the current optimal loop optimization amount corresponding to the current optimal solution. Here, traversal data refers to the data generated by the computer device during the traversal of the branch subtrees of the node corresponding to the candidate node's state key, and backtracking data refers to the data generated by the computer device during the backtracking of the branch subtrees of the node corresponding to the candidate node's state key.

[0164] Step S509: When there is no next backtracking node for the r-th backtracking node, the current optimal solution corresponding to the r-th backtracking node is determined as the optimal batch parts replacement scheme for the S engines to be replaced.

[0165] In this embodiment of the application, when the computer device backtracks the initial search tree to the root node, that is, when there is no next backtracking node after the r-th backtracking node, the computer device can determine the current optimal solution corresponding to the r-th backtracking node as the optimal batch parts replacement scheme for the S engines to be replaced.

[0166] Through the above process, based on engine parts, a list of candidate exchange methods is generated for each engine to be replaced, thus generating all possible candidate exchange methods. This avoids the problems of limited experience and unreasonable exchange methods in traditional manual decision-making, which leads to a single exchange method. It broadens the range of parts exchange options and improves the objectivity and comprehensiveness of the generated candidate exchange methods. By constructing a global conflict matrix, the logical relationship between any two candidate exchange methods can be clearly defined. This avoids the difficulty in comprehensively sorting out the logical relationships between multiple candidate methods during manual decision-making, which can lead to problems of uncoordinated solutions during implementation. The global conflict matrix provides a clear basis for subsequent solution selection, improving the rationality and feasibility of solution formulation and avoiding rework due to unidentified conflicts, thereby indirectly improving operation and maintenance efficiency. The rapid generation of initial batch parts replacement plans and the backtracking iteration of the initial search tree shorten the formulation cycle of the optimal batch parts replacement plan for S engines to be replaced and improve the generation efficiency of the optimal batch parts replacement plan. Furthermore, by sorting each candidate swap method to obtain a set of candidate swap methods, it is possible to prioritize and retain candidate swap methods with better optimization effects when generating the initial batch of part replacement schemes. This enables rapid locking of the initial batch of part replacement schemes, significantly improving the generation efficiency of the initial batch of part replacement schemes, and indirectly improving the generation efficiency of the optimal batch of part replacement schemes. Moreover, through upper bound pruning and residual potential pruning, search branches that cannot be better than the current optimal solution are eliminated in advance. This reduces the search space of the initial search tree, avoids invalid backtracking, significantly reduces the number of search nodes, and lowers the algorithm time and computational resource consumption.

[0167] Further, please see Figure 7 , Figure 7 This is a schematic diagram of a component optimization device for an engine provided in an embodiment of this application. This component optimization device can be a computer program (including program code, etc.) running on a computer device; for example, it can be an application software. The component optimization device can be used to execute corresponding steps in the method provided in the embodiments of this application. Figure 7 As shown, the engine-specific parts optimization device 700 can be used for Figure 3 and Figure 5Specifically, the computer device in the corresponding embodiment may include: an engine acquisition module 11, a candidate exchange method determination module 12, a conflict detection module 13, a sorting processing module 14, a traversal processing module 15, and a backtracking processing module 16.

[0168] Engine acquisition module 11 is used to acquire S engines with the same engine model to be replaced; S is an integer greater than 1. The candidate exchange method determination module 12 is used to generate a candidate exchange method list for each engine part in the S engines to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information for replacing one or more engine parts in an engine to be replaced. The conflict detection module 13 is used to create a global conflict matrix that indicates the logical relationship between any two candidate exchange methods based on each candidate exchange method in the list of S candidate exchange methods. The sorting module 14 is used to obtain the minimum loop optimization amount corresponding to each candidate swap method in the S candidate swap method list, and sort each candidate swap method in the S candidate swap method list in descending order of the minimum loop optimization amount of each candidate swap method to obtain a set of candidate swap methods. The traversal processing module 15 is used to construct an initial search tree by using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set, and to traverse the initial search tree. When the traversal process reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. The backtracking module 16 is used to backtrack the initial search tree based on the initial batch parts replacement scheme until the optimal batch parts replacement scheme for the S engines to be replaced is obtained.

[0169] In one alternative implementation, the S replacement engines include replacement engine A i , where i is a positive integer less than or equal to S; when the candidate exchange method determination module 12 generates a candidate exchange method list for each engine to be replaced based on each engine part in the S engines to be replaced, the candidate exchange method determination module 12 is specifically used to perform the following operations: For replacing engine A i The M engine parts are traversed and combined to obtain T parts combinations; M and T are both positive integers; a parts combination must include at least one of the M engine parts. Based on T parts combinations and S-1 remaining engine parts, generate a sublist of candidate exchange methods for each of the T parts combinations; S-1 remaining engines refer to the S engines to be replaced excluding engine A. i Other engines to be replaced; The T candidate swapping sublists are combined to form the engine A to be replaced. i The corresponding list of candidate swapping methods.

[0170] In one alternative implementation, the T component combinations include component combination B. j , j is a positive integer less than or equal to T; when the candidate exchange method determination module 12 generates a candidate exchange method sublist corresponding to each of the T part combinations based on the T part combinations and the engine parts in the S-1 remaining engines, the candidate exchange method determination module 12 is specifically used to perform the following operations: From the remaining S-1 engines, obtain the parts combination B respectively. j P engine parts with the same part type are considered as S-1 candidate part combinations; P is a positive integer less than or equal to M. According to part assembly B j Combine with candidate part S-1 to generate part combination B. j The corresponding G candidate swaps; G is a positive integer less than or equal to S-1; Assemble part B j The corresponding G candidate exchange methods are determined to be part combination B. j The corresponding candidate swapping sublist.

[0171] In one optional implementation, the list of S candidate swaps includes N candidate swaps, where N is a positive integer; the N candidate swaps include candidate swap C. k Candidate swapping method C z Both k and z are positive integers less than or equal to N; when the conflict detection module 13 is used to create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods based on each candidate exchange method in the S candidate exchange method list, the conflict detection module 13 is specifically used to perform the following operations: If candidate exchange method C k Engine parts and candidate exchange methods C z If one or more engine parts have the same part identifier, then candidate exchange method C is determined. k Candidate swapping method C z The logical relationship between them is mutual exclusion; If candidate exchange method C kEngine parts and candidate exchange methods C z If no engine part in the configuration has the same part identifier, then candidate exchange method C is determined. k Candidate swapping method C z The logical relationship between them is a compatibility relationship; When the logical relationship between any two candidate exchange methods among N candidate exchange methods is obtained, a global conflict matrix is ​​generated based on the logical relationship between any two candidate exchange methods.

[0172] In one optional implementation, the list of S candidate swaps includes N candidate swaps, where N is a positive integer; the N candidate swaps include candidate swap C. k k is a positive integer less than or equal to N; when sorting module 14 obtains the minimum loop optimization amount corresponding to each candidate swap method in the S candidate swap method list, sorting module 14 is specifically used to perform the following operations: Obtain candidate swapping method C k The remaining cycle number of each engine part in the target engine to be replaced is determined as the minimum remaining cycle number of the target engine to be replaced; S engines to be replaced include the target engine to be replaced. Obtain candidate swapping method C k The remaining cycle number for each candidate part; candidate exchange method C k The candidate parts are selected through candidate exchange method C. k The corresponding combination of candidate parts determines the outcome; Obtain candidate swapping method C k The remaining cycle number for each unreplaced part; unreplaced parts refer to the target engine to be replaced, excluding candidate exchange method C. k Engine parts other than the released parts; Candidate exchange method C k The released part is determined based on one of the T part combinations corresponding to the target engine to be replaced; candidate exchange method C k Release components and candidate exchange methods C k The candidate parts in the list have the same part type; The remaining cycle number corresponding to each candidate part and the remaining cycle number corresponding to each unreplaced part are determined as the set of remaining cycle numbers. The minimum value in the set of remaining cycle numbers is determined as the second minimum remaining cycle number of the target engine to be replaced. The difference between the first minimum remaining cycle number and the second minimum remaining cycle number is determined as the candidate swap method C. kThe corresponding minimum loop optimization amount.

[0173] In one optional implementation, the candidate swap set includes N candidate swaps, where N is a positive integer; the traversal processing module 15 is used to construct the initial search tree using the global conflict matrix and the order of each candidate swap in the candidate swap set. Specifically, the traversal processing module 15 performs the following operations: Create a root node with initial node state information; Based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set, a w-th layer child node is created for the w-1 layer target node; w is a positive integer less than or equal to N; the multiple target logical relationships include the logical relationship between the w-th candidate exchange method and each candidate exchange method in the w-1 layer selected exchange method set; when w is 1, the w-1 layer target node is the root node; the w-th layer child node is the child node of the w-1 layer target node, and the w-th layer child node includes a selection node for indicating the selection of the w-th candidate exchange method, and / or a skip node for indicating skipping the w-th candidate exchange method; the w-1-th candidate exchange method in the candidate exchange method set has a corresponding relationship with the w-1 layer target node; the w-1 layer target node is determined in the w-1 layer child nodes; the w-1 layer selected exchange method set includes the candidate exchange methods selected in the node path from the root node to the w-1 layer target node; w is a positive integer less than or equal to N; When the Nth level child node is obtained, an initial search tree is formed based on the nodes between the root node and the Nth level child node.

[0174] In one optional implementation, the traversal processing module 15 is used to create a child node at the w-th layer for a target node at the (w-1)-th layer based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set. Specifically, the traversal processing module 15 is used to perform the following operations: Determine the target node at level (w-1) from the child nodes at level (w-1); If multiple target logical relationships in the global conflict matrix are all compatible, then based on the w-th candidate exchange method in the candidate exchange method set, for the target node of the (w-1)th layer, create the w-th selection node to indicate the selection of the w-th candidate exchange method, and the w-th skip node to indicate skipping the w-th candidate exchange method. Then, determine the w-th selection node and the w-th skip node as the w-th layer child nodes. If one or more of the target logical relationships in the global conflict matrix are mutually exclusive, then the first set of released part identifiers in the node status information of the target node at level w-1 is detected. If the first set of released part identifiers includes the part identifier of the candidate part in the candidate exchange method, then it is a target node at level w-1. A selection node for indicating the selection of the candidate exchange method at level w and a skip node for indicating skipping the candidate exchange method at level w are created. The selection node at level w and the skip node at level w are determined as child nodes at level w. If the first set of released part identifiers does not include the part identifier of the candidate part in the w-th candidate exchange method, then according to the w-th candidate exchange method in the set of candidate exchange methods, a w-th skip node is created for the target node of the (w-1)-th layer to indicate skipping the w-th candidate exchange method, and the w-th skip node is determined as the w-th layer child node.

[0175] In one alternative implementation, when the traversal processing module 15 determines the target node at level w-1 from the child nodes at level w-1, the traversal processing module 15 specifically performs the following operations: If the (w-1)th layer child nodes include the (w-1)th selected node and the (w-1)th skipped node, then the (w-1)th selected node is determined as the (w-1)th layer target node; If the (w-1)th layer of child nodes includes only the (w-1)th skipped node, then the (w-1)th skipped node is determined as the (w-1)th layer of target node.

[0176] In one alternative implementation, when the traversal processing module 15 creates the w-th selection node to indicate the selection of the w-th candidate exchange method, the traversal processing module 15 specifically performs the following operations: Each candidate swap in the (w-1)th selected swap set, and the wth candidate swap, form the wth selected swap set; Based on each candidate exchange method in the w-th selected exchange method set, a set of used part identifiers is generated to indicate the used engine parts, and a set of released part identifiers is generated to indicate the released parts; the set of used part identifiers includes the part identifiers of the engine parts for each candidate exchange method in the w-th selected exchange method set; the set of released part identifiers includes the part identifiers of the released parts for each candidate exchange method in the w-th selected exchange method set; the engine parts in the w-th candidate exchange method include the released parts in the w-th candidate exchange method. Based on each candidate swap method in the w-th selected swap method set, determine the cumulative loop optimization amount corresponding to selecting the w-th candidate swap method; Obtain the index position of the w-th candidate swap method in the candidate swap method set, and determine the search tree depth based on the index position of the w-th candidate swap method; The first node status information is composed of the set of used part identifiers, the set of released part identifiers, the cumulative loop optimization amount corresponding to the selection of the w-th candidate exchange method, the index position of the w-th candidate exchange method, and the search tree depth. Based on the status information of the first node, create the w-th selection node.

[0177] In one alternative implementation, when the traversal processing module 15 creates the w-th skip node to indicate skipping the w-th candidate swap, the traversal processing module 15 specifically performs the following operations: Obtain the node state information of the target node at layer w-1; Get the index position of the w-th candidate swap method in the candidate swap method set; The depth of the search tree is determined based on the index position of the w-th candidate swap. Based on the index position of the w-th candidate exchange method and the depth of the search tree, the node state information of the target node at the (w-1)-th layer is updated to obtain the second node state information. Based on the status information of the second node, create the w-th skip node.

[0178] In one optional implementation, the backtracking module 16 is used to backtrack the initial search tree based on the initial batch parts replacement scheme until the optimal batch parts replacement scheme for the S engines to be replaced is obtained. Specifically, the backtracking module 16 is used to perform the following operations: Obtain the total loop optimization amount corresponding to the initial batch part replacement plan; Obtain the timeout thresholds corresponding to the S engines to be replaced; In the initial search tree, the r-th backtracking node is obtained, along with the search execution time corresponding to it. If the search execution time is less than or equal to the timeout threshold, an upper bound pruning threshold and a remaining potential pruning threshold are generated based on the node state information and the remaining candidate swapping methods corresponding to the r-th backtracking node. The remaining candidate swapping methods corresponding to the r-th backtracking node refer to the candidate swapping methods whose index position is after the index position in the node state information of the r-th backtracking node. When r=1, the (r-1)-th backtracking node is the first leaf node in the initial search tree, the current optimal loop optimization amount corresponding to the (r-1)-th backtracking node is the total loop optimization amount, and the current optimal solution corresponding to the (r-1)-th backtracking node is the initial batch part replacement scheme. When r is greater than 1, the r-th backtracking node is determined based on the (r-1)-th backtracking node. r is a positive integer. If the current optimal loop optimization amount corresponding to the r-th backtracking node is less than the upper bound pruning threshold and less than the remaining potential pruning threshold, then according to the remaining candidate exchange method corresponding to the r-th backtracking node, a branch subtree of the r-th backtracking node is generated. According to the branch subtree of the r-th backtracking node, the current optimal solution corresponding to the (r-1)-th backtracking node is updated to obtain the current optimal solution corresponding to the r-th backtracking node. The loop optimization amount of the current optimal solution corresponding to the r-th backtracking node is determined as the current optimal loop optimization amount corresponding to the r-th backtracking node. If the current optimal loop optimization amount corresponding to the r-th backtracking node is greater than or equal to the upper bound pruning threshold and greater than or equal to the remaining potential pruning threshold, then continue to obtain the current optimal loop optimization amount corresponding to the (r+1)-th backtracking node. When there is no next backtracking node for the r-th backtracking node, the current optimal solution corresponding to the r-th backtracking node is determined as the optimal batch parts replacement scheme for the S engines to be replaced.

[0179] In one optional implementation, when the backtracking module 16 obtains the r-th backtracking node, the backtracking module 16 is specifically used to perform the following operations: If the (r-1)th backtracking node has an unvisited sibling node, then the unvisited sibling node of the (r-1)th backtracking node is determined as the rth backtracking node. If the (r-1)th backtracking node does not have any unvisited sibling nodes, then backtrack to the parent node of the (r-1)th backtracking node and determine the unvisited sibling node of the parent node of the (r-1)th backtracking node as the rth backtracking node.

[0180] In one optional implementation, the backtracking processing module 16 is used to update the current optimal solution corresponding to the (r-1)th backtracking node based on the branch subtree of the r-th backtracking node. When the current optimal solution corresponding to the r-th backtracking node is obtained, the backtracking processing module 16 is specifically used to perform the following operations: Traverse the subtree of the r-th backtracking node. When the first leaf node in the subtree of the r-th backtracking node is reached, generate an updated batch parts replacement plan based on the node status information of the first leaf node in the subtree of the r-th backtracking node. Based on the updated batch parts replacement scheme, backtracking is performed on the branch subtree of the r-th backtracking node until the optimal batch parts replacement scheme corresponding to the r-th backtracking node is obtained. The optimal batch parts replacement scheme corresponding to the r-th backtracking node is determined as the current optimal solution corresponding to the r-th backtracking node.

[0181] In one optional implementation, when the backtracking module 16 generates the upper bound pruning threshold and the remaining potential pruning threshold based on the node state information corresponding to the r-th backtracking node and the remaining candidate swapping method corresponding to the r-th backtracking node, the backtracking module 16 is specifically used to perform the following operations: Based on the selected exchange method set, the released part identifier set, and the global conflict matrix in the node status information corresponding to the r-th backtracking node, the remaining candidate exchange methods corresponding to the r-th backtracking node are filtered to obtain Q remaining candidate exchange methods; Q is a positive integer. Based on the engine identifiers of the engines to be replaced corresponding to the Q remaining candidate exchange methods, the Q remaining candidate exchange methods are grouped to obtain Y remaining candidate exchange method groups; Y is a positive integer less than or equal to Q; the remaining candidate exchange methods in each remaining candidate exchange method group correspond to the same engine identifier; Obtain the minimum loop optimization amount corresponding to each of the Q remaining candidate swapping methods. Based on the minimum loop optimization amount of the remaining candidate swapping methods in each group of remaining candidate swapping methods, obtain the minimum loop optimization amount with the largest value from each of the Y remaining candidate swapping method groups. The upper bound pruning threshold is generated based on the sum of the minimum loop optimization quantities of Y nodes with the largest values ​​and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node. The remaining potential pruning threshold is generated based on the sum of the Q minimum loop optimization quantities and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node.

[0182] Through the above process, based on engine parts, a list of candidate exchange methods is generated for each engine to be replaced, thus generating all possible candidate exchange methods. This avoids the problems of limited experience and unreasonable exchange methods in traditional manual decision-making, which leads to a single exchange method. It broadens the range of parts exchange options and improves the objectivity and comprehensiveness of the generated candidate exchange methods. By constructing a global conflict matrix, the logical relationship between any two candidate exchange methods can be clearly defined. This avoids the difficulty in comprehensively sorting out the logical relationships between multiple candidate methods during manual decision-making, which can lead to problems of uncoordinated solutions during implementation. The global conflict matrix provides a clear basis for subsequent solution selection, improving the rationality and feasibility of solution formulation and avoiding rework due to unidentified conflicts, thereby indirectly improving operation and maintenance efficiency. The rapid generation of initial batch parts replacement plans and the backtracking iteration of the initial search tree shorten the formulation cycle of the optimal batch parts replacement plan for S engines to be replaced and improve the generation efficiency of the optimal batch parts replacement plan. Furthermore, by sorting each candidate swap method to obtain a set of candidate swap methods, it is possible to prioritize and retain candidate swap methods with better optimization effects when generating the initial batch of part replacement schemes. This enables rapid locking of the initial batch of part replacement schemes, significantly improving the generation efficiency of the initial batch of part replacement schemes, and indirectly improving the generation efficiency of obtaining the optimal batch of part replacement schemes. Moreover, through upper bound pruning and residual potential pruning, search branches that cannot be better than the current optimal solution are eliminated in advance, thereby reducing the search space of the initial search tree, avoiding invalid backtracking, significantly reducing the number of search nodes, and lowering the algorithm time and computational resource consumption.

[0183] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 8 As shown, the computer device 800 in this embodiment may include a processor 801, a network interface 804, and a memory 805. Furthermore, the computer device 800 may also include a user interface 803 and at least one communication bus 802. The communication bus 802 is used to enable communication between these components. The user interface 803 may include a display screen and a keyboard; optionally, the user interface 803 may also include a standard wired interface or a wireless interface. The network interface 804 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 805 may be a high-speed RAM or non-volatile memory, such as at least one disk storage device. Optionally, the memory 805 may also be at least one storage device located remotely from the processor 801. Figure 8As shown, the memory 805, which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a device control application.

[0184] Network interface 804 provides network communication elements; user interface 803 is mainly used to provide an input interface for users; and processor 801 can be used to call the device control application stored in memory 805 to perform the following operations: Get S engines with the same engine model that need to be replaced; S is an integer greater than 1. Based on each engine part in the S engines to be replaced, generate a candidate exchange method list for each engine to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information on replacing one or more engine parts in an engine to be replaced. Based on each candidate exchange method in the list of S candidate exchange methods, create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods; Obtain the minimum loop optimization amount for each candidate swap method in the list of S candidate swap methods. Sort each candidate swap method in the list of S candidate swap methods in descending order of the minimum loop optimization amount to obtain the set of candidate swap methods. An initial search tree is constructed using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. The initial search tree is traversed. When the traversal reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. Based on the initial batch parts replacement scheme, the initial search tree is backtracked until the optimal batch parts replacement scheme for the S engines to be replaced is obtained.

[0185] Furthermore, it should be noted that embodiments of this application also provide a computer-readable storage medium storing a computer program adapted to be loaded and executed by the processor. Figure 3 or Figure 5 For details on the methods provided in each step, please refer to the document. Figure 3 or Figure 5The implementation methods provided for each step are not repeated here. Furthermore, the beneficial effects of using the same method are also not repeated. For technical details not disclosed in the computer-readable storage medium embodiments involved in this application, please refer to the description of the method embodiments of this application. As an example, a computer program may be deployed to execute on a single computer device, or on multiple computer devices located in one location, or on multiple computer devices distributed across multiple locations and interconnected via a communication network.

[0186] The computer-readable storage medium can be the apparatus provided in any of the foregoing embodiments or the internal storage unit of the computer device, such as the hard disk or memory of the computer device. The computer-readable storage medium can also be an external storage device of the computer device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0187] This application also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform... Figure 3 or Figure 5 The methods provided are among the various optional methods available in the code, so they will not be elaborated upon here.

[0188] The terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other step units inherent to these processes, methods, apparatuses, products, or devices.

[0189] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0190] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0191] The methods and related apparatus provided in this application are described with reference to the method flowcharts and / or structural diagrams provided in this application. Specifically, each block of the method flowchart and / or structural diagram, as well as combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable device to create a machine, such that the instructions, which execute via the processor of the computer or other programmable device, generate instructions for implementing the process. Figure 1 A schematic diagram of one or more processes and / or structures. Figure 1 The computer program instructions may be stored in a computer-readable storage medium that can direct a computer or other programmable device to function in a particular manner, causing the instructions stored in the computer-readable storage medium to produce an article of manufacture including the instruction means, or to be transmitted via a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The instruction means is implemented in the process. Figure 1 A schematic diagram of one or more processes and / or structures. Figure 1The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 A process or multiple processes and / or structures illustrate the steps of the functions specified in one or more boxes.

[0192] The steps in the method of this application embodiment can be adjusted, combined, or deleted according to actual needs.

[0193] The modules in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0194] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A method for optimizing engine parts, characterized in that, The method includes: Get S engines with the same engine model that need to be replaced; S is an integer greater than 1. Based on each engine part in the S engines to be replaced, a candidate exchange method list is generated for each engine to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information on replacing one or more engine parts in an engine to be replaced. Based on each candidate exchange method in the list of S candidate exchange methods, create a global conflict matrix to indicate the logical relationship between any two candidate exchange methods; Obtain the minimum loop optimization amount corresponding to each candidate exchange method in the S candidate exchange method list. Sort each candidate exchange method in the S candidate exchange method list in descending order of the minimum loop optimization amount of each candidate exchange method to obtain a candidate exchange method set. The minimum loop optimization amount corresponding to a candidate exchange method is used to characterize the difference between the minimum remaining number of cycles before the engine to be replaced is executed and the minimum remaining number of cycles after the execution of the candidate exchange method. An initial search tree is constructed using the global conflict matrix and the order of each candidate exchange method in the candidate exchange method set. The initial search tree is traversed. When the traversal reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. Based on the initial batch parts replacement scheme, the initial search tree is backtracked until the optimal batch parts replacement scheme for the S engines to be replaced is obtained.

2. The method according to claim 1, characterized in that, The S engines to be replaced include engine A. i , where i is a positive integer less than or equal to S; The step of generating a candidate exchange method list for each of the S engines to be replaced, based on each engine part, includes: For the engine A to be replaced i The M engine parts are traversed and combined to obtain T parts combinations; M and T are both positive integers; a parts combination must include at least one of the M engine parts. Based on the T parts combinations and the engine parts in the S-1 remaining engines, generate a sublist of candidate exchange methods corresponding to each of the T parts combinations; the S-1 remaining engines refer to the S engines to be replaced excluding engine A. i Other engines to be replaced; The T candidate exchange method sublists are used to form the engine A to be replaced. i The corresponding list of candidate swapping methods.

3. The method according to claim 2, characterized in that, The T component combinations include component combination B. j j is a positive integer less than or equal to T; The step of generating a candidate exchange method sublist corresponding to each of the T part combinations and the engine parts in the S-1 remaining engines includes: From the S-1 remaining engines, obtain the parts that are combined with the aforementioned component B. j P engine parts in the model have the same part type as candidate parts, which are combined as S-1 candidate parts; P is a positive integer less than or equal to M. According to the component assembly B j Combined with the candidate parts S-1, the part combination B is generated. j The corresponding G candidate swaps; G is a positive integer less than or equal to S-1; Assemble the parts B j The corresponding G candidate exchange methods are determined to be part combination B. j The corresponding candidate swapping sublist.

4. The method according to claim 1, characterized in that, The list of S candidate swapping methods includes N candidate swapping methods, where N is a positive integer; the N candidate swapping methods include candidate swapping method C. k Candidate swapping method C z Both k and z are positive integers less than or equal to N; The step of creating a global conflict matrix to indicate the logical relationship between any two candidate exchange methods based on each of the S candidate exchange method lists includes: If the candidate exchange method C k Engine components and the candidate exchange method C z If, among the engine parts, there are one or more engine parts with the same part identifier, then the candidate exchange method C is determined. k and the candidate exchange method C z The logical relationship between them is mutual exclusion; If the candidate exchange method C k Engine components and the candidate exchange method C z If no engine part in the specified engine parts has the same part identifier, then the candidate exchange method C is determined. k and the candidate exchange method C z The logical relationship between them is a compatibility relationship; When the logical relationship between any two candidate exchange methods among the N candidate exchange methods is obtained, a global conflict matrix is ​​generated based on the logical relationship between the two candidate exchange methods.

5. The method according to claim 1, characterized in that, The list of S candidate swapping methods includes N candidate swapping methods, where N is a positive integer; the N candidate swapping methods include candidate swapping method C. k k is a positive integer less than or equal to N; The step of obtaining the minimum loop optimization amount corresponding to each candidate swap method in the list of S candidate swap methods includes: Obtain the candidate exchange method C k The remaining cycle number of each engine part in the target engine to be replaced is determined as the minimum remaining cycle number of the target engine to be replaced; the S engines to be replaced include the target engine to be replaced. Obtain the candidate exchange method C k The remaining cycle number corresponding to each candidate part; the candidate exchange method C k The candidate parts are selected through the candidate exchange method C. k The corresponding combination of candidate parts determines the outcome; Obtain the candidate exchange method C k The remaining cycle number for each unreplaced part; the unreplaced parts refer to the target engine to be replaced, excluding candidate exchange method C. k Engine parts other than the release parts; the candidate exchange method C k The released component is determined based on one of the T component combinations corresponding to the target engine to be replaced; the candidate exchange method C k The release component and the candidate exchange method C k The candidate parts in the list have the same part type; The remaining cycle number corresponding to each candidate part and the remaining cycle number corresponding to each unreplaced part are determined as the set of remaining cycle numbers, and the minimum value in the set of remaining cycle numbers is determined as the second minimum remaining cycle number of the target engine to be replaced. The difference between the first minimum remaining cycle number and the second minimum remaining cycle number is determined as the candidate swap method C. k The corresponding minimum loop optimization amount.

6. The method according to claim 1, characterized in that, The candidate exchange method set includes N candidate exchange methods, where N is a positive integer; The step of constructing an initial search tree using the global conflict matrix and the arrangement order of each candidate exchange method in the candidate exchange method set includes: Create a root node with initial node state information; Based on the multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set, a w-th layer child node is created for the (w-1)-th layer target node; w is a positive integer less than or equal to N; the multiple target logical relationships include the logical relationship between the w-th candidate exchange method and each candidate exchange method in the (w-1)-th selected exchange method set; when w is 1, the (w-1)-th layer target node is the root node; the w-th layer child node is the child node of the (w-1)-th layer target node, and the w-th layer child node... The nodes include selection nodes for indicating the selection of the w-th candidate swap method, and / or skip nodes for indicating skipping the w-th candidate swap method; the (w-1)-th candidate swap method in the candidate swap method set corresponds to the (w-1)-th layer target node; the (w-1)-th layer target node is determined among the (w-1)-th layer child nodes; the (w-1)-th selected swap method set includes the candidate swap methods selected in the node path from the root node to the (w-1)-th layer target node; w is a positive integer less than or equal to N; When the Nth-level child node is obtained, an initial search tree is formed based on the nodes between the root node and the Nth-level child node.

7. The method according to claim 6, characterized in that, The step of creating a layer-w child node for a layer-w (w-1) target node based on multiple target logical relationships in the global conflict matrix and the w-th candidate exchange method in the candidate exchange method set includes: Determine the target node at level (w-1) from the child nodes at level (w-1); If all target logical relationships in the global conflict matrix are compatible, then according to the w-th candidate exchange method in the candidate exchange method set, for the target node of the (w-1)-th layer, a w-th selection node for indicating the selection of the w-th candidate exchange method and a w-th skip node for indicating skipping the w-th candidate exchange method are created, and the w-th selection node and the w-th skip node are determined as the w-th layer child nodes; If one or more of the target logical relationships in the global conflict matrix are mutually exclusive, then the first set of released part identifiers in the node status information of the target node at layer w-1 is detected. If the first set of released part identifiers is detected to include the part identifier of the candidate part in the w-th candidate exchange method, then a w-th selection node for indicating the selection of the w-th candidate exchange method and a w-th skip node for indicating skipping the w-th candidate exchange method are created for the target node at layer w-1. The w-th selection node and the w-th skip node are determined as the w-th layer child node. If the first set of released part identifiers does not include the part identifier of the candidate part in the w-th candidate exchange method, then according to the w-th candidate exchange method in the set of candidate exchange methods, a w-th skip node is created for the target node of the (w-1)-th layer to indicate skipping the w-th candidate exchange method, and the w-th skip node is determined as a child node of the w-th layer.

8. The method according to claim 7, characterized in that, Determining the target node at level (w-1) from the child nodes at level (w-1) includes: If the (w-1)th layer child nodes include the (w-1)th selected node and the (w-1)th skipped node, then the (w-1)th selected node is determined as the (w-1)th layer target node; If the (w-1)th layer of child nodes includes only the (w-1)th skipped node, then the (w-1)th skipped node is determined as the (w-1)th layer of target node.

9. The method according to claim 7, characterized in that, The creation of the w-th selection node for indicating the selection of the w-th candidate exchange method includes: Each candidate swap method in the (w-1)th selected swap method set, and the wth candidate swap method, form the wth selected swap method set; Based on each candidate exchange method in the w-th selected exchange method set, a set of used part identifiers is generated to indicate used engine parts, and a set of released part identifiers is generated to indicate released parts; the set of used part identifiers includes the part identifiers of engine parts for each candidate exchange method in the w-th selected exchange method set; the set of released part identifiers includes the part identifiers of released parts for each candidate exchange method in the w-th selected exchange method set; the engine parts in the w-th candidate exchange method include the released parts in the w-th candidate exchange method. Based on each candidate swap method in the set of selected swap methods, determine the cumulative loop optimization amount corresponding to selecting the w-th candidate swap method; the cumulative loop optimization amount corresponding to the w-th candidate swap method is used to indicate the sum of the minimum loop optimization amounts of each candidate swap method in the set of selected swap methods. Obtain the index position of the w-th candidate swap method in the candidate swap method set, and determine the search tree depth based on the index position of the w-th candidate swap method; The first node status information is composed of the set of used part identifiers, the set of released part identifiers, the cumulative loop optimization amount corresponding to the selection of the w-th candidate exchange method, the index position of the w-th candidate exchange method, and the search tree depth. Based on the status information of the first node, create the w-th selection node.

10. The method according to claim 7, characterized in that, The creation of the w-th skip node for indicating skipping the w-th candidate exchange method includes: Obtain the node state information of the target node at the (w-1)th layer; Obtain the index position of the w-th candidate exchange method in the candidate exchange method set; The search tree depth is determined based on the index position of the w-th candidate exchange method; Based on the index position of the w-th candidate exchange method and the depth of the search tree, the node state information of the target node at the (w-1)-th layer is updated to obtain the second node state information; Based on the status information of the second node, create the w-th skip node.

11. The method according to claim 1, characterized in that, The process of backtracking the initial search tree based on the initial batch parts replacement scheme until the optimal batch parts replacement scheme for the S engines to be replaced is obtained includes: Obtain the total cycle optimization amount corresponding to the initial batch part replacement scheme; Obtain the timeout threshold values ​​corresponding to the S engines to be replaced; In the initial search tree, the r-th backtracking node is obtained, and the search execution time corresponding to the r-th backtracking node is obtained. If the search execution time is less than or equal to the timeout threshold, an upper bound pruning threshold and a remaining potential pruning threshold are generated based on the node state information corresponding to the r-th backtracking node and the remaining candidate exchange methods corresponding to the r-th backtracking node. The remaining candidate exchange methods corresponding to the r-th backtracking node refer to the candidate exchange methods in the candidate exchange method set whose index position is after the index position in the node state information of the r-th backtracking node. When r=1, the (r-1)-th backtracking node is the first leaf node in the initial search tree, the current optimal loop optimization amount corresponding to the (r-1)-th backtracking node is the total loop optimization amount, and the current optimal solution corresponding to the (r-1)-th backtracking node is the initial batch part replacement scheme. When r is greater than 1, the r-th backtracking node is based on the (r-1)-th backtracking node. The backtracking node is determined; r is a positive integer; the upper bound pruning threshold is used to characterize the sum of the cumulative loop optimization amount in the node state information corresponding to the r-th backtracking node and the minimum loop optimization amount with the largest value corresponding to the Y remaining candidate exchange method groups; the Y remaining candidate exchange method groups are obtained by grouping Q remaining candidate exchange methods according to the engine identifier of the engine to be replaced; Q is a positive integer, Y is a positive integer less than or equal to Q; the Q remaining candidate exchange methods are obtained by filtering the remaining candidate exchange methods corresponding to the r-th backtracking node according to the selected exchange method set, the released part identifier set and the global conflict matrix in the node state information corresponding to the r-th backtracking node; the remaining potential pruning threshold is used to characterize the sum of the cumulative loop optimization amount in the node state information corresponding to the r-th backtracking node and the minimum loop optimization amount corresponding to the Q remaining candidate exchange methods respectively; If the current optimal loop optimization amount corresponding to the r-th backtracking node is less than the upper bound pruning threshold and less than the remaining potential pruning threshold, then according to the remaining candidate exchange method corresponding to the r-th backtracking node, a branch subtree of the r-th backtracking node is generated. According to the branch subtree of the r-th backtracking node, the current optimal solution corresponding to the (r-1)-th backtracking node is updated to obtain the current optimal solution corresponding to the r-th backtracking node. The loop optimization amount of the current optimal solution corresponding to the r-th backtracking node is determined as the current optimal loop optimization amount corresponding to the r-th backtracking node. If the current optimal loop optimization amount corresponding to the r-th backtracking node is greater than or equal to the upper bound pruning threshold and greater than or equal to the remaining potential pruning threshold, then continue to obtain the current optimal loop optimization amount corresponding to the (r+1)-th backtracking node. When there is no next backtracking node for the r-th backtracking node, the current optimal solution corresponding to the r-th backtracking node is determined as the optimal batch parts replacement scheme for the S engines to be replaced.

12. The method according to claim 11, characterized in that, Obtaining the r-th backtracking node includes: If the (r-1)th backtracking node has an unvisited sibling node, then the unvisited sibling node of the (r-1)th backtracking node is determined as the rth backtracking node. If the (r-1)th backtracking node does not have any unvisited sibling nodes, then backtrack to the parent node of the (r-1)th backtracking node, and determine the unvisited sibling node of the parent node of the (r-1)th backtracking node as the rth backtracking node.

13. The method according to claim 11, characterized in that, The step of updating the current optimal solution corresponding to the (r-1)th backtracking node based on the branch subtree of the r-th backtracking node to obtain the current optimal solution corresponding to the r-th backtracking node includes: The branch subtree of the r-th backtracking node is traversed. When the first leaf node in the branch subtree of the r-th backtracking node is reached, an updated batch parts replacement plan is generated based on the node status information of the first leaf node in the branch subtree of the r-th backtracking node. According to the updated batch part replacement scheme, the branch subtree of the r-th backtracking node is backtracked until the optimal batch part replacement scheme corresponding to the r-th backtracking node is obtained. The optimal batch part replacement scheme corresponding to the r-th backtracking node is determined as the current optimal solution corresponding to the r-th backtracking node.

14. The method according to claim 11, characterized in that, The step of generating an upper bound pruning threshold and a remaining potential pruning threshold based on the node state information corresponding to the r-th backtracking node and the remaining candidate exchange method corresponding to the r-th backtracking node includes: Based on the selected exchange method set, the released part identifier set in the node status information corresponding to the r-th backtracking node, and the global conflict matrix, the remaining candidate exchange methods corresponding to the r-th backtracking node are filtered to obtain Q remaining candidate exchange methods; Q is a positive integer. Based on the engine identifiers of the engines to be replaced corresponding to the Q remaining candidate exchange methods, the Q remaining candidate exchange methods are grouped to obtain Y remaining candidate exchange method groups; Y is a positive integer less than or equal to Q; the remaining candidate exchange methods in each remaining candidate exchange method group correspond to the same engine identifier; Obtain the minimum loop optimization amount corresponding to each of the Q remaining candidate swapping methods. Based on the minimum loop optimization amount of the remaining candidate swapping methods in each group of remaining candidate swapping methods, obtain the minimum loop optimization amount with the largest value from each of the Y remaining candidate swapping method groups. The upper bound pruning threshold is generated based on the sum of the minimum loop optimization quantities of Y nodes with the largest values, and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node. The remaining potential pruning threshold is generated based on the sum of the Q minimum loop optimization quantities and the cumulative loop optimization quantity in the node state information corresponding to the r-th backtracking node.

15. A component optimization device for an engine, characterized in that, The device includes: The engine acquisition module is used to acquire S engines with the same engine model to be replaced; S is an integer greater than 1. The candidate exchange method determination module is used to generate a candidate exchange method list for each of the S engines to be replaced, based on each engine part in the S engines to be replaced; the candidate exchange method in the candidate exchange method list is used to indicate information for replacing one or more engine parts in an engine to be replaced. The conflict detection module is used to create a global conflict matrix that indicates the logical relationship between any two candidate exchange methods, based on each candidate exchange method in the list of S candidate exchange methods. The sorting module is used to obtain the minimum loop optimization amount corresponding to each candidate exchange method in the S candidate exchange method list, and sort each candidate exchange method in the S candidate exchange method list in descending order of the minimum loop optimization amount of each candidate exchange method to obtain a set of candidate exchange methods; the minimum loop optimization amount corresponding to a candidate exchange method is used to characterize the difference between the minimum remaining number of cycles before the engine to be replaced by the candidate exchange method is executed and the minimum remaining number of cycles after the candidate exchange method is executed. The traversal processing module is used to construct an initial search tree based on the global conflict matrix and the arrangement order of each candidate exchange method in the candidate exchange method set, and to traverse the initial search tree. When the traversal process reaches the first leaf node in the initial search tree, an initial batch part replacement plan is generated based on the node status information of the first leaf node. The node status information includes the candidate exchange methods selected in the node path from the root node of the initial search tree to the first leaf node. The backtracking module is used to backtrack the initial search tree based on the initial batch parts replacement scheme until the optimal batch parts replacement scheme for the S engines to be replaced is obtained.

16. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 14.

17. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any one of claims 1 to 14.

18. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 14.

Images