Spare part optimal configuration method and device based on bee colony algorithm

Abstract

The embodiment of the invention provides a spare part optimal configuration method and device based on a bee colony algorithm, and the method comprises the steps: constructing a fault tree logic structure according to the influence relation between internal fault events of equipment, and constructing a multi-agent simulation evaluation model based on the fault tree logic structure; simulating an operation state and a guarantee work flow of the equipment based on the multi-agent simulation evaluation model, and performing guarantee efficiency evaluation on the implemented spare part configuration scheme; and by taking the multi-agent simulation evaluation model as an evaluation function, carrying out optimization solution on the spare part configuration scheme based on a bee colony algorithm to obtain an optimal spare part configuration scheme.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of spare parts configuration optimization, and in particular, to a spare parts optimization configuration method and device based on a bee colony algorithm. Background Art

[0002] In the fields of military, aviation, aerospace, etc., the support efficiency of equipment directly affects the success of the mission. The design of spare parts configuration scheme is crucial to the reliability of equipment, especially in the maintenance activities of ships, spacecraft and other equipment. The rational configuration of spare parts and the optimization of the number of spare parts carried are the key to ensure the smooth progress of equipment maintenance and missions. Due to the limited equipment space, there is a restrictive relationship between the spare parts satisfaction rate and the spare parts utilization rate. How to optimize the spare parts configuration scheme under the premise of ensuring the spare parts satisfaction rate to improve the spare parts utilization rate and thus improve the equipment support efficiency has become an urgent problem to be solved.

[0003] At present, spare parts configuration schemes usually rely on empirical rules, statistical models or simple optimization algorithms to adjust the number and type of spare parts, which cannot fully consider actual usage and complex failure modes. In addition, due to the limited data samples, it is difficult to use more complex machine learning methods, which cannot meet the dual requirements of high utilization and high satisfaction of modern ship equipment. Summary of the invention

[0004] The embodiments described herein provide a spare parts optimization configuration method and apparatus based on a bee colony algorithm, and a computer-readable storage medium storing a computer program.

[0005] According to a first aspect of the present disclosure, a method for optimizing spare parts configuration based on a bee swarm algorithm is provided, comprising: constructing a fault tree logic structure according to the influence relationship between internal fault events of equipment, and constructing a multi-agent simulation evaluation model based on the fault tree logic structure; simulating the operating status and support workflow of the equipment based on the multi-agent simulation evaluation model, and evaluating the support effectiveness of the implemented spare parts configuration scheme; and optimizing and solving the spare parts configuration scheme based on a bee swarm algorithm using the multi-agent simulation evaluation model as an evaluation function to obtain an optimal spare parts configuration scheme.

[0006] In some embodiments of the present disclosure, a fault tree logic structure is constructed according to the influence relationship between internal fault events of the equipment, and a multi-agent simulation evaluation model is constructed based on the fault tree logic structure, including: taking unit component failure events as bottom events and equipment failure events as top events, connecting each event using corresponding logic gates, and constructing a fault tree logic structure of the equipment; and according to the fault tree logic structure, setting unit component agents, logic gate agents and equipment agents respectively for simulating the operating status of unit components, logic gates and equipment and ensuring the workflow, to construct a multi-agent simulation evaluation model.

[0007] In some embodiments of the present disclosure, the operating status of the unit component includes normal and faulty. When the unit component functions intact and operates normally, the unit component is in a normal state; when a unit component fails, the unit component enters a faulty state and requests support resources from a maintenance support library; the faulty state of the unit component includes a waiting for maintenance state and a maintenance state. In the waiting for maintenance state, the unit component waits for the dispatch of support resources, and in the maintenance state, maintenance activities are performed until the maintenance is successful and the unit component returns to a normal state.

[0008] In some embodiments of the present disclosure, the logic gate includes an AND gate, an OR gate, and a voting gate, and the logic gate implements different judgment logics according to the number of input fault events.

[0009] In some embodiments of the present disclosure, the judgment logic of the logic gate is uniformly expressed as u>r? Fault: Normal, where u is the number of fault events in the input events of the logic gate, and r is a preset threshold.

[0010] In some embodiments of the present disclosure, the setting rule of the preset threshold is:

[0011]

[0012] Where e is the minimum number of input events required for the voting gate output to be normal, and j is the number of input events.

[0013] In some embodiments of the present disclosure, the operating status of the equipment includes normal and faulty, and the equipment agent is used to transfer between the normal state and the faulty state according to the operating status of each unit component of the equipment monitored in real time and the information transmitted by the fault tree logic structure.

[0014] In some embodiments of the present disclosure, the operating status and support workflow of the equipment are simulated based on a multi-agent simulation evaluation model, and the support effectiveness evaluation of the implemented spare parts configuration plan is performed, including: when the underlying unit component agent fails, the fault information is transmitted upward along the fault tree logic structure, and the working status of the equipment is judged in the equipment agent; the corresponding spare parts configuration plan is executed according to the operating status of the equipment, and the usage of the spare parts is recorded; and after the task is completed, the spare parts satisfaction rate and utilization rate of the executed spare parts configuration plan are evaluated to determine whether the spare parts configuration plan is feasible.

[0015] In some embodiments of the present disclosure, the multi-agent simulation evaluation model is an evaluation function, and the spare parts configuration scheme is optimized and solved based on the bee colony algorithm to obtain the optimal spare parts configuration scheme, including: step 1, determining the optimization goal and constraints: taking the specified spare parts satisfaction rate as the constraint condition, and taking the minimum number of spare parts carried as the optimization goal; step 2, initializing the spare parts configuration scheme: using the method of randomly generating feasible solutions to design a set of spare parts configuration schemes, using the multi-agent simulation evaluation model to calculate the fitness value, judging whether the scheme is feasible according to the fitness value, and if not feasible, repeating this step until a feasible solution is obtained. The initial spare parts configuration plan of the row; Step three, bee colony search and update: the leading bee searches for a new food source based on the initial spare parts configuration plan, calculates the fitness value and makes a greedy choice. The leading bee recruits follower bees according to the roulette method. The follower bees select the food source according to the selection probability and further search for the new food source location. If a scout bee is generated, the scout bee randomly searches for a new food source; and Step four, iterative optimization: determine whether the end condition is met. If the end condition is not met, return to Step three for the next search. If the end condition is met, end the optimization search and output the optimal spare parts configuration plan.

[0016] In some embodiments of the present disclosure, the formula for randomly generating feasible solutions is as follows:

[0017] x ij =x min,j +rand(0,1)(x max,j -x min,j );

[0018] Where i = 1, 2, 3, ..., m, j ∈ {1, 2, ..., n}, m is the number of feasible solutions, n is the number of optimization parameters, x min,j and x max,j is a feasible solution x ij The lower and upper limits of the value range. rand(0,1) is a random number between 0 and 1.

[0019] In some embodiments of the present disclosure, the formula for the leader bee to search for a new food source based on the initial spare parts configuration scheme is:

[0020] v ij =x ij +φ ik (x ij -x kj ), where j∈{1,2,…,n}, k∈{1,2,…,m}, and k is

[0021] Randomly generated and k≠i, φ ik is a random number between [-1,1], v ij Configure solutions for new spare parts.

[0022] In some embodiments of the present disclosure, the fitness value f i The calculation formula is:

[0023]

[0024] Among them, obj(x i ) is the objective function value, and the calculation formula of the objective function value is: Among them, x i is the optimization parameter, n is the number of spare parts, x ij is the number of spare parts of the jth type carried in the i-th feasible solution, and the objective function is Among them, F(S) is the number of spare parts carried, P(S) is the spare parts satisfaction rate, P0 is the specified satisfaction rate, S is the spare parts configuration plan, x i =S.

[0025] In some embodiments of the present disclosure, in step three, if the follower bee is not assigned a food source, a food source is selected for the follower bee, the follower bee searches for a new food source, calculates a fitness value and performs a greedy selection, and records the current optimal food source location.

[0026] In some embodiments of the present disclosure, the probability of a follower bee selecting a food source according to the fitness value is calculated by the following formula:

[0027]

[0028] Among them, p i is the selection probability, m is the number of feasible solutions, f i is the fitness value of the i-th feasible solution.

[0029] In some embodiments of the present disclosure, the number of scout bees is 5%-20% of the number of bee colonies.

[0030] In some embodiments of the present disclosure, the termination conditions include: the maximum number of iterations has been reached, the fitness value of the current spare parts configuration scheme has reached a preset target, and the change in the fitness value in several iterations is less than a preset value.

[0031] According to the second aspect of the present disclosure, a spare parts optimization configuration device based on a bee colony algorithm is provided. The device includes at least one processor; and at least one memory storing a computer program. When the computer program is executed by at least one processor, the device: constructs a fault tree logic structure according to the influence relationship between internal fault events of the equipment, and constructs a multi-agent simulation evaluation model based on the fault tree logic structure; simulates the operating status and guarantee workflow of the equipment based on the multi-agent simulation evaluation model, and evaluates the guarantee effectiveness of the implemented spare parts configuration scheme; and optimizes and solves the spare parts configuration scheme based on the bee colony algorithm using the multi-agent simulation evaluation model as an evaluation function to obtain the optimal spare parts configuration scheme.

[0032] According to a third aspect of the present disclosure, a computer-readable storage medium storing a computer program is provided, wherein the computer program, when executed by a processor, implements the steps of the spare parts optimization configuration method based on a bee colony algorithm according to the first aspect of the present disclosure.

[0033] According to the spare parts optimization configuration method and device based on the bee colony algorithm of the embodiment of the present disclosure, based on the failure rate of equipment unit components and using the fault tree as a framework, a multi-agent simulation evaluation model is established to simulate the operation and support workflow of the equipment, and to achieve the support efficiency evaluation of the spare parts configuration scheme, thus solving the problem that the support efficiency corresponding to the spare parts configuration scheme cannot be directly calculated. At the same time, using this as the evaluation function of the bee colony algorithm, with the prescribed spare parts satisfaction rate as a prerequisite and the goal of reducing the number of spare parts carried, the onboard spare parts configuration scheme is optimized. When the prescribed spare parts satisfaction rate requirements are met, the spare parts utilization rate is further improved, thereby improving the equipment support efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly described below. It should be noted that the drawings described below only relate to some embodiments of the present disclosure, but are not intended to limit the present disclosure, wherein:

[0035] Figure 1 An exemplary flow chart of a spare parts optimization configuration method 100 based on a bee colony algorithm according to an embodiment of the present disclosure is shown;

[0036] Figure 2 is a state transition logic diagram of a unit component intelligent body according to an embodiment of the present disclosure;

[0037] Figure 3 is a state transition logic diagram of a logic gate agent according to an embodiment of the present disclosure;

[0038] Figure 4 is a state transition logic diagram of an equipment agent according to an embodiment of the present disclosure;

[0039] Figure 5 It is a solution flow chart for optimizing spare parts configuration scheme by combining multi-agent simulation evaluation model and bee colony algorithm according to an embodiment of the present disclosure;

[0040] Figure 6 It is a schematic block diagram of a spare parts optimization configuration device 600 based on a bee colony algorithm according to an embodiment of the present disclosure.

[0041] It should be noted that the elements in the drawings are schematic and not drawn to scale. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work also fall within the scope of protection of the present disclosure.

[0043] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which the subject matter of the present disclosure belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the specification and the relevant art, and will not be interpreted in an idealized or overly formal form unless otherwise explicitly defined herein.

[0044] In view of the problem of limited resource storage space and high requirements for spare parts guarantee efficiency, the embodiments of the present disclosure propose a spare parts configuration optimization method and device based on a bee colony algorithm. The spare parts configuration scheme is optimized by combining multi-agent simulation technology and a bee colony algorithm. Through multi-agent simulation, the use of spare parts in different maintenance tasks, as well as the spare parts demand and usage efficiency under different circumstances can be simulated, and the rationality of different spare parts configuration schemes can be evaluated, which can overcome the assumptions and limitations in traditional mathematical models. On the basis of the simulation model, the bee colony algorithm is combined to simulate the "dance" behavior of bees in the spare parts configuration space, and find the optimal configuration scheme that can maximize the spare parts utilization rate and ensure a certain spare parts satisfaction rate. This can effectively solve the spare parts configuration problem in ship equipment maintenance, while ensuring the spare parts satisfaction rate, the number of spare parts carried can be reduced as much as possible, thereby improving space utilization and optimizing maintenance resource allocation.

[0045] Figure 1 An exemplary flow chart of a spare parts optimization configuration method 100 based on a bee colony algorithm according to an embodiment of the present disclosure is shown.

[0046] exist Figure 1In block S102, a fault tree logic structure is constructed according to the impact relationship between internal fault events of the equipment, and a multi-agent simulation evaluation model is constructed based on the fault tree logic structure.

[0047] Considering that there is little public data on spare parts configuration schemes and performance evaluation, the disclosed embodiment uses multi-agent simulation to evaluate the performance of spare parts configuration, and forms a complete workflow simulation model through collaboration and interaction between multiple agents. Multi-agent simulation can dynamically simulate the "normal-fault-repair-restore to normal" operating status of equipment and the use of support resources. Its core advantage is that it can simplify complex systems while retaining the interactive relationship between each unit, and then deeply restore its maintenance and support workflow.

[0048] In some embodiments of the present disclosure, complex equipment such as ships are usually composed of multiple levels of nested subsystems and components, and there are complex interactions between these subsystems and components. The combination of multi-agent simulation technology and fault tree model can effectively restore the equipment's workflow and fault recovery process, and evaluate the effectiveness of resource allocation solutions.

[0049] Among them, the fault tree is a tool for analyzing system failure modes. It connects the failure events of unit components (bottom events) and equipment failures (top events) through logic gates (AND, OR, etc.) to form a hierarchical relationship. The fault tree can effectively simulate the equipment fault propagation process and analyze the impact of each component on the entire system. In order to build a multi-agent simulation and evaluation model based on the fault tree, agents at different levels can be used to simulate the equipment's operating status, fault propagation, and support workflow.

[0050] In some embodiments of the present disclosure, the unit component failure event is used as the bottom event, the equipment failure event is used as the top event, and the corresponding logic gates are used to connect the events to construct the fault tree logic structure of the equipment. According to the fault tree logic structure, the unit component agent, the logic gate agent, and the equipment agent are respectively used to simulate the operating status of the unit components, logic gates, and equipment and ensure the workflow, and a multi-agent simulation evaluation model is constructed.

[0051] Among them, the unit component agent is used to simulate the operating status of unit components such as circuit boards and engine components. Each component agent is responsible for monitoring the operating status of its components and passing the fault information upward according to the fault tree logic when a fault occurs.

[0052] Figure 2 FIG. 1 is a state transition logic diagram of a unit component intelligent body according to an embodiment of the present disclosure. Figure 2As shown, the operating status of the unit component includes "normal" and "faulty". When the unit component functions well and operates normally, the component is in the "normal" state; when a unit component fails, the component enters the "faulty" state, indicating that the component cannot continue to work, and requests support resources from the maintenance support library, and enters the "waiting for maintenance" secondary state to wait for the scheduling of support resources; when the faulty component meets the maintenance conditions, the component enters the "maintenance" secondary state and starts maintenance activities until it returns to the "normal" state after the maintenance is successful.

[0053] Logic gate agents are used to construct complete fault tree models. Figure 3 is a state transition logic diagram of a logic gate agent according to an embodiment of the present disclosure. Figure 3 As shown, the operation status of the logic gate includes "normal" and "fault". When the logic gate functions well, it is in the "normal" state; when the logic gate fails, it enters the "fault" state.

[0054] The equipment's fault tree logic gates mainly include AND gates, OR gates, and voting gates. The logic gates implement different judgment logics based on the number of input fault events. Since the equipment structure is fixed and the number of input events for each logic gate is known, it is only necessary to calculate the number of fault events in the corresponding gate to achieve the effects of different logic gates.

[0055] In some embodiments of the present disclosure, by introducing the threshold r, the fault judgment logic of the AND gate, OR gate and voting gate can be uniformly expressed, avoiding setting different processing methods for each gate separately. The judgment logic of the three logic gates can be uniformly expressed as:

[0056] u>r? Fault: Normal;

[0057] Where u is the number of fault events in the input events of the logic gate, and r is the preset threshold. The setting rules of the threshold r in the three logic gates are:

[0058] Where e is the minimum number of input events required for the voting gate output to be normal, and j is the number of input events.

[0059] Equipment agents are used to simulate the operating status of complex equipment. Figure 4 is a state transition logic diagram of an equipment intelligent agent according to an embodiment of the present disclosure. Figure 4 As shown, the operating status of the equipment includes "normal" and "faulty". The equipment agent is used to transfer between the normal state and the faulty state according to the operating status of each unit component of the equipment monitored in real time and the information transmitted by the fault tree logical structure.

[0060] Using the above three types of agents, a multi-agent simulation evaluation model is built according to the equipment fault tree logical structure.

[0061] Reference Figure 1 As shown, then in box S104, the operating status and support workflow of the equipment are simulated based on the multi-agent simulation evaluation model, and the support effectiveness of the implemented spare parts configuration plan is evaluated.

[0062] In some embodiments of the present disclosure, when a fault occurs in the underlying unit component agent, the fault information is transmitted upward along the fault tree logic structure, and the working status of the equipment is judged in the equipment agent. For example, if a component or multiple components fail and the failure propagates to the key parts of the equipment, the status of the entire equipment will change from "normal" to "failure". If the equipment returns to normal by repairing or replacing components, the status will change from "failure" to "normal".

[0063] Execute the corresponding spare parts configuration plan according to the operating status of the equipment and record the use of spare parts. After the task is completed, evaluate the spare parts satisfaction rate and utilization rate of the executed spare parts configuration plan to determine whether the spare parts configuration plan is feasible.

[0064] For example, the equipment agent records maintenance resources (spare parts, replacement tools, maintenance personnel, etc.), energy resources (consumed electricity, fuel, etc.), time resources (time consumed by maintenance, recovery and other tasks), and evaluates the effectiveness of the resource allocation plan after the task is completed.

[0065] In order to achieve the requirements of high spare parts satisfaction rate and spare parts utilization rate, in some embodiments of the present disclosure, the spare parts configuration optimization model is expressed as:

[0066]

[0067] Among them, F(S) is the number of spare parts carried, P(S) is the spare parts satisfaction rate, P0 is the specified satisfaction rate, and S is the spare parts configuration plan. The optimization model describes how to maximize the total utilization of spare parts, that is, minimize the number of spare parts carried, under the premise that the spare parts satisfaction rate P(S) of the spare parts configuration plan S reaches the specified satisfaction rate P0.

[0068] The utilization rate of spare parts can be calculated by counting whether spare parts are dispatched in time and whether they meet the repair requirements when a fault occurs. If the utilization rate of spare parts is too low, it means that the resource allocation is unreasonable, and you may need to increase the spare parts inventory or adjust the guarantee plan. When a fault occurs, the spare parts satisfaction rate is calculated by judging whether there are enough spare parts to repair all damaged parts.

[0069] In block S106, the multi-agent simulation evaluation model is used as the evaluation function, and the spare parts configuration scheme is optimized and solved based on the bee colony algorithm to obtain the optimal spare parts configuration scheme.

[0070] Among them, the bee colony algorithm is a group intelligence optimization algorithm that imitates the behavior of bees. It does not need to understand the special information of the problem, but only needs to compare the pros and cons of the problem. Through the local optimization behavior of each individual worker bee, the global optimal value will eventually emerge in the group, and it has a relatively fast convergence speed.

[0071] The minimum search model for the bee colony to generate collective intelligence contains three basic components: food sources, employed bees (leading bees) and unemployed bees (following bees and scout bees). Leading bees correspond to the collected food sources one by one. Leading bees store relevant information about the food source (relative distance from the hive, direction and abundance of the food source (size of fitness value), etc.) and share this information with other bees with a certain probability. If the nectar source is updated multiple times without improvement, the nectar source is abandoned and the employed bees turn into scout bees to randomly search for new nectar sources.

[0072] The main task of scout bees and follower bees is to find and exploit food sources. Scout bees search for new food sources nearby; follower bees wait in the hive and find food sources by sharing relevant information with leader bees. Leader bees express the profitability of food sources through the duration of their swing dances, so follower bees can observe a large number of dances and choose which food source to collect honey from based on the profitability. The profitability is proportional to the probability of a food source being selected. Therefore, the probability of a bee being recruited to a food source is proportional to the profitability of the food source.

[0073] Based on the above algorithm rules, the spare parts configuration scheme is optimized and solved in combination with the equipment multi-agent simulation evaluation model. In some embodiments of the present disclosure, the optimal spare parts configuration scheme can be found through the following steps:

[0074] First, in step 1, the optimization goal and constraints are determined: the specified spare parts satisfaction rate is used as the constraint, and the minimum number of spare parts carried is used as the optimization goal.

[0075] In step 2, the spare parts configuration scheme is initialized: a set of spare parts configuration schemes is designed by randomly generating feasible solutions, and the fitness value is calculated using a multi-agent simulation evaluation model. The feasibility of the scheme is determined based on the fitness value. If not, the step is repeated until a feasible initial spare parts configuration scheme is obtained.

[0076] In order to avoid the same process every time the same problem is solved, a random number generator can be used to configure the number of spare parts of each type or the selection of spare parts types. In some embodiments of the present disclosure, the formula for randomly generating feasible solutions is as follows:

[0077] x ij =x min,j +rand(0,1)(x max,j -x min,j);

[0078] Where i = 1, 2, 3, ..., m, j ∈ {1, 2, ..., n}, m is the number of feasible solutions, n is the number of optimization parameters, x min,j and x max,j is a feasible solution x ij The lower and upper limits of the value range. rand(0,1) is a random number between 0 and 1.

[0079] In step three, the bee colony searches and updates: the leading bee searches for new food sources based on the initial spare parts configuration plan, calculates the fitness value and makes a greedy selection. The leading bee recruits follower bees according to the roulette method. The follower bees select food sources according to the selection probability and further search for new food source locations. If scout bees are generated, they randomly search for new food sources.

[0080] In some embodiments of the present disclosure, the formula for the leader bee to search for a new food source based on the initial spare parts configuration scheme is:

[0081]

[0082] Where j∈{1,2,…,n}, k∈{1,2,…,m}, k is randomly generated and k≠i, is a random number between [-1,1], v ij Configure solutions for new spare parts.

[0083] Generally, the fitness value is related to multiple factors such as the efficiency, cost, and response time of the configuration solution. Spare parts utilization and satisfaction rate can measure the efficiency of the configuration solution. Spare parts utilization rate = actual number of spare parts used / number of configured spare parts; spare parts satisfaction rate = actual number of spare parts that meet the demand / satisfaction rate of the number of required spare parts. The fitness function comprehensively considers factors such as utilization rate and satisfaction rate. According to the embodiments of the present disclosure, the fitness value f i The calculation formula is:

[0084]

[0085] Among them, obj(x i ) is the objective function value, and the calculation formula of the objective function value is: Among them, x i For the optimization parameters, n is the number of spare parts, x ij is the number of spare parts of the jth type carried in the i-th feasible solution, and the objective function is Among them, F(S) is the number of spare parts carried, P(S) is the spare parts satisfaction rate, P0 is the specified satisfaction rate, S is the spare parts configuration plan, x i =S.

[0086] According to the calculated fitness value, the leader bee will select the best food source. If the fitness value of the new food source (spare parts configuration scheme) found is higher, it will be used as the new food source, otherwise the current food source will be retained.

[0087] The leader bee recruits follower bees by roulette according to the fitness value. The food source with higher fitness value is more likely to attract more follower bees. The follower bee selects the food source according to the selection probability and further searches for the new food source location. If the follower bee assigns the food source, the current optimal food source location is recorded. If the follower bee does not assign the food source, a food source is selected for the follower bee, and the follower bee searches for the new food source, calculates the fitness value and makes a greedy selection, and records the current optimal food source location.

[0088] The probability p of the follower bee selecting a food source according to its fitness value can be calculated by the following formula: i :

[0089]

[0090] Where m is the number of feasible solutions, f i is the fitness value of the i-th feasible solution.

[0091] The role of the scout bee is to explore new areas, that is, to jump out of the local optimal solution and randomly select a new spare parts configuration plan for evaluation. The introduction of the scout bee avoids the dilemma of the local optimal solution. The number of scout bees is 5%-20% of the number of bee colonies.

[0092] In step 4, iterative optimization: determine whether the end condition is met. If the end condition is not met, return to step 3 for the next search. If the end condition is met, end the optimization search and output the optimal spare parts configuration plan. Among them, the end conditions include: the maximum number of iterations has been reached, the fitness value of the current spare parts configuration plan has reached the preset target, and the change in the fitness value in several iterations is less than the preset value, indicating that a stable solution has been found.

[0093] Figure 5 This is a flowchart for optimizing the spare parts configuration scheme by combining a multi-agent simulation evaluation model and a bee colony algorithm according to an embodiment of the present disclosure. Figure 5 As shown, first, a set of solutions is designed by random value selection method, and then a multi-agent simulation evaluation model is used to determine whether the solution is feasible. If not, the steps are repeated until a feasible solution is obtained, and the feasible solution is used as the initial spare parts configuration solution of the bee swarm algorithm.

[0094] In the bee colony algorithm, the evaluation function determines the fitness of the individual. The final evaluation function can be combined with the objective function and the constraint function. The fitness value of each individual is calculated using the evaluation function. The higher the fitness, the better the spare parts configuration scheme.

[0095] Leading bee stage (looking for new solutions): The leading bee searches in a neighborhood to update the food source based on the initial spare parts configuration plan. The updated individual needs to recalculate the fitness and compare it with the fitness of the original solution to select the solution with higher fitness. The leading bee recruits follower bees according to the roulette method. If the follower bee allocates the food source, the current optimal food source location is recorded. If the food source is updated multiple times without improvement, the food source is abandoned and the leading bee turns into a scout bee to randomly search for a new food source. If the follower bee is not allocated a food source, a food source is selected for the follower bee, the follower bee searches for a new food source, changes the food source mark, calculates the fitness value and selects the food source according to the fitness value, and records the current optimal food source location.

[0096] Scout bee stage (global search): Scout bees randomly search for new food sources in the entire space, replace marked food sources, and mark new food sources.

[0097] In each round of iteration, all swarm individuals are updated according to their fitness, and the optimal spare parts configuration scheme is selected according to the evaluation function. The swarm algorithm can terminate the algorithm by setting the maximum number of iterations or the convergence of the fitness value. The optimization stops when one of the following conditions is met: The maximum number of iterations is reached. The fitness change of the current solution is less than a certain threshold, indicating that the search has converged. Find a solution that meets all constraints and is good enough. When the algorithm terminates, the individual with the highest fitness in the swarm is output as the optimal solution, that is, the optimal spare parts configuration scheme.

[0098] Figure 6 is a schematic block diagram of a spare parts optimization configuration device 600 based on a bee colony algorithm according to an embodiment of the present disclosure. Figure 6 As shown, the device 600 may include a processor 610 and a memory 620 storing a computer program. When the computer program is executed by the processor 610, the device 600 may perform the following operations: Figure 1 The steps of the spare parts optimization configuration method 100 based on the bee colony algorithm are shown. In one example, the device 600 can be a computer device or a cloud computing node. The device 600 can: construct a fault tree logic structure according to the influence relationship between the internal fault events of the equipment, and construct a multi-agent simulation evaluation model based on the fault tree logic structure; simulate the operating status and guarantee workflow of the equipment based on the multi-agent simulation evaluation model, and evaluate the guarantee effectiveness of the implemented spare parts configuration plan; and optimize and solve the spare parts configuration plan based on the bee colony algorithm using the multi-agent simulation evaluation model as the evaluation function to obtain the optimal spare parts configuration plan.

[0099] In some embodiments of the present disclosure, the device 600 can take the unit component failure event as the bottom event and the equipment failure event as the top event, and use the corresponding logic gates to connect the events to construct the fault tree logic structure of the equipment; and according to the fault tree logic structure, set the unit component agent, logic gate agent and equipment agent respectively for simulating the operating status of the unit components, logic gates and equipment and ensuring the workflow, and construct a multi-agent simulation evaluation model.

[0100] In some embodiments of the present disclosure, the device 600 may obtain an optimal spare parts configuration solution through the following steps:

[0101] Step 1: Determine the optimization goal and constraints: Take the specified spare parts satisfaction rate as the constraint and the minimum number of spare parts to be carried as the optimization goal;

[0102] Step 2: Initialize spare parts configuration scheme: Design a set of spare parts configuration schemes by randomly generating feasible solutions, calculate the fitness value by multi-agent simulation evaluation model, and judge whether the scheme is feasible according to the fitness value. If not, repeat this step until a feasible initial spare parts configuration scheme is obtained.

[0103] Step 3: Swarm search and update: The leader bee searches for a new food source based on the initial spare parts configuration plan, calculates the fitness value and performs greedy selection. The leader bee recruits follower bees according to the roulette method. The follower bees select a food source according to the selection probability and further search for a new food source location. If a scout bee is generated, the scout bee randomly searches for a new food source; and

[0104] Step 4: Iterative optimization: Determine whether the end condition is met. If not, return to step 3 for the next search. If the end condition is met, end the optimization search and output the optimal spare parts configuration plan.

[0105] In an embodiment of the present disclosure, the processor 610 may be, for example, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a processor based on a multi-core processor architecture, etc. The memory 620 may be any type of memory implemented using data storage technology, including but not limited to random access memory, read-only memory, semiconductor-based memory, flash memory, disk storage, etc.

[0106] In addition, in the embodiment of the present disclosure, the apparatus 600 may also include an input device 630, such as a keyboard, a mouse, etc. In addition, the apparatus 600 may also include an output device 640, such as a display, etc., for outputting the optimal spare parts configuration solution.

[0107] In other embodiments of the present disclosure, a computer-readable storage medium storing a computer program is further provided, wherein the computer program can achieve the following when executed by a processor: Figure 1 The steps of the spare parts optimization configuration method 100 based on the bee colony algorithm are shown.

[0108] In summary, according to the spare parts optimization configuration method and device based on the bee colony algorithm of the embodiment of the present disclosure, based on the failure rate of equipment unit components and using the fault tree as a framework, a multi-agent simulation evaluation model is established to simulate the operation and support workflow of the equipment, and to achieve the support efficiency evaluation of the spare parts configuration scheme, thereby solving the problem that the support efficiency corresponding to the spare parts configuration scheme cannot be directly calculated. At the same time, using this as the evaluation function of the bee colony algorithm, with the prescribed spare parts satisfaction rate as a prerequisite and the goal of reducing the number of spare parts carried, the onboard spare parts configuration scheme is optimized. When the prescribed spare parts satisfaction rate requirements are met, the spare parts utilization rate is further improved, thereby improving the equipment support efficiency. This scheme is not only feasible in theory, but can also be flexibly adjusted according to actual conditions to meet the needs of different tasks.

[0109] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the device and method according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, a program segment or an instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0110] Unless the context clearly indicates otherwise, the singular form of the words used herein and in the appended claims includes the plural and vice versa. Thus, when referring to the singular, the plural form of the corresponding term is generally included. Similarly, the words "comprise" and "include" are to be interpreted as inclusive rather than exclusive. Likewise, the terms "include" and "or" should be interpreted as inclusive unless such interpretation is expressly prohibited herein. Where the term "example" is used herein, particularly when it is located after a group of terms, the "example" is merely exemplary and illustrative and should not be considered exclusive or comprehensive.

[0111] Further aspects and scopes of adaptability become apparent from the description provided herein. It should be understood that various aspects of the present application can be implemented individually or in combination with one or more other aspects. It should also be understood that the description and specific embodiments herein are intended for purposes of illustration only and are not intended to limit the scope of the present application.

[0112] Several embodiments of the present disclosure are described in detail above, but it is obvious that those skilled in the art can make various modifications and variations to the embodiments of the present disclosure without departing from the spirit and scope of the present disclosure. The protection scope of the present disclosure is defined by the attached claims.

Claims

1. A spare parts optimization configuration method based on bee colony algorithm, characterized in that: include: Constructing a fault tree logic structure according to the influence relationship between internal fault events of the equipment, and constructing a multi-agent simulation evaluation model based on the fault tree logic structure; Simulating the operation status and support workflow of the equipment based on the multi-agent simulation evaluation model, and evaluating the support effectiveness of the implemented spare parts configuration scheme; as well as The multi-agent simulation evaluation model is used as the evaluation function, and the spare parts configuration scheme is optimized and solved based on the bee colony algorithm to obtain the optimal spare parts configuration scheme.

2. The spare parts optimization configuration method based on bee colony algorithm according to claim 1 is characterized in that: The method of constructing a fault tree logic structure according to the influence relationship between internal fault events of the equipment, and constructing a multi-agent simulation evaluation model based on the fault tree logic structure includes: Taking the unit component failure event as the bottom event and the equipment failure event as the top event, the corresponding logic gates are used to connect the events to construct the equipment fault tree logic structure; and According to the fault tree logic structure, unit component agents, logic gate agents and equipment agents are set up to simulate the operating status of unit components, logic gates and equipment and ensure the workflow, and a multi-agent simulation evaluation model is constructed.

3. The spare parts optimization configuration method based on bee colony algorithm according to claim 2 is characterized in that: The operating status of the unit component includes normal and faulty. When the unit component functions intact and operates normally, the unit component is in a normal state; when a unit component fails, the unit component enters a faulty state and requests support resources from the maintenance and support library; the faulty state of the unit component includes a waiting for maintenance state and a maintenance state. In the waiting for maintenance state, the unit component waits for the scheduling of support resources, and in the maintenance state, maintenance activities are performed until the maintenance is successful and the unit component returns to the normal state.

4. The spare parts optimization configuration method based on bee colony algorithm according to claim 2 is characterized in that: The logic gates include AND gates, OR gates and voting gates, and the logic gates implement different judgment logics according to the number of input fault events.

5. The spare parts optimization configuration method based on bee colony algorithm according to claim 4 is characterized in that: The judgment logic of the logic gate is uniformly expressed as u>r? Fault: Normal, where u is the number of fault events in the input events of the logic gate, and r is the preset threshold.

6. The spare parts optimization configuration method based on bee colony algorithm according to claim 5 is characterized in that: The setting rule of the preset threshold is: Where e is the minimum number of input events required for the voting gate output to be normal, and j is the number of input events.

7. The spare parts optimization configuration method based on bee colony algorithm according to claim 2 is characterized in that: The operating status of the equipment includes normal and faulty. The equipment intelligent agent is used to transfer between the normal state and the faulty state according to the operating status of each unit component of the equipment monitored in real time and the information transmitted by the fault tree logical structure.

8. The spare parts optimization configuration method based on bee colony algorithm according to claim 2 is characterized in that: The simulating the operation status and guarantee workflow of the equipment based on the multi-agent simulation evaluation model and evaluating the guarantee effectiveness of the implemented spare parts configuration scheme includes: When a fault occurs in the bottom unit component agent, the fault information is transmitted upward along the fault tree logic structure, and the operating status of the equipment is judged in the equipment agent; Execute the corresponding spare parts configuration plan according to the operating status of the equipment and record the usage of spare parts; and After the task is completed, the spare parts satisfaction rate and utilization rate of the executed spare parts configuration plan are evaluated to determine whether the spare parts configuration plan is feasible.

9. The spare parts optimization configuration method based on bee colony algorithm according to claim 1, characterized in that: The multi-agent simulation evaluation model is used as an evaluation function, and the spare parts configuration scheme is optimized and solved based on the bee colony algorithm to obtain the optimal spare parts configuration scheme, which includes: Step 1: Determine the optimization goal and constraints: Take the specified spare parts satisfaction rate as the constraint and the minimum number of spare parts to be carried as the optimization goal; Step 2, initializing the spare parts configuration scheme: designing a set of spare parts configuration schemes by randomly generating feasible solutions, calculating the fitness value by using the multi-agent simulation evaluation model, judging whether the scheme is feasible according to the fitness value, and if not feasible, repeating this step until a feasible initial spare parts configuration scheme is obtained; Step 3, bee colony search and update: the leading bee searches for a new food source based on the initial spare parts configuration scheme, calculates the fitness value and performs greedy selection, the leading bee recruits the follower bee according to the roulette method, the follower bee selects the food source according to the selection probability, and further searches for the new food source location, if a scout bee is generated, the scout bee randomly searches for a new food source; and Step 4: Iterative optimization: Determine whether the end condition is met. If not, return to step 3 for the next search. If the end condition is met, end the optimization search and output the optimal spare parts configuration plan.

10. The spare parts optimization configuration method based on bee colony algorithm according to claim 9, wherein the formula for randomly generating feasible solutions is as follows: x ij =x min,j +rand(0,1)(x max,j -x min,j ); Where i = 1, 2, 3, ..., m, j ∈ {1, 2, ..., n}, m is the number of feasible solutions, n is the number of optimization parameters, x min,j and x max,j is a feasible solution x ij The lower and upper limits of the value range. rand(0,1) is a random number between 0 and 1.

11. The spare parts optimization configuration method based on bee colony algorithm according to claim 10, characterized in that: The formula for the leader bee to search for a new food source based on the initial spare parts configuration scheme is: Where j∈{1,2,…,n}, k∈{1,2,…,m}, k is randomly generated and k≠i, is a random number between [-1,1], v ij Configure solutions for new spare parts.

12. The spare parts optimization configuration method based on bee colony algorithm according to claim 10, characterized in that: The fitness value f i The calculation formula is: Among them, obj(x i ) is the objective function value, and the calculation formula of the objective function value is: Among them, x i is the optimization parameter, n is the number of spare parts, x ij is the number of spare parts of the jth type carried in the i-th feasible solution, and the objective function is Among them, F(S) is the number of spare parts carried, P(S) is the spare parts satisfaction rate, P0 is the specified satisfaction rate, S is the spare parts configuration plan, x i =S.

13. The spare parts optimization configuration method based on bee colony algorithm according to claim 12, characterized in that: In step three, if the follower bee has not been assigned a food source, a food source is selected for the follower bee, the follower bee searches for a new food source, calculates the fitness value and performs greedy selection, and records the current optimal food source location.

14. The spare parts optimization configuration method based on bee colony algorithm according to claim 13, characterized in that: The probability of the follower bee selecting a food source according to the fitness value is calculated by the following formula: Among them, p i is the selection probability, m is the number of feasible solutions, f i is the fitness value of the i-th feasible solution.

15. The spare parts optimization configuration method based on bee colony algorithm according to claim 9, characterized in that: The number of the scout bees is 5%-20% of the number of the bee colony.

16. The spare parts optimization configuration method based on bee colony algorithm according to claim 9, characterized in that: The termination conditions include: the maximum number of iterations has been reached, the fitness value of the current spare parts configuration scheme has reached a preset target, and the change in the fitness value in several iterations is less than a preset value.

17. A spare parts optimization configuration device based on bee colony algorithm, characterized in that: The device comprises: at least one processor; and at least one memory storing a computer program; Wherein, when the computer program is executed by the at least one processor, the device executes the steps of the spare parts optimization configuration method based on the bee colony algorithm according to any one of claims 1 to 16.

18. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the computer program implements the steps of the spare parts optimization configuration method based on the bee colony algorithm according to any one of claims 1 to 16.

Images