Resource collaborative scheduling method and device for ship pipe fitting workshop, electronic equipment and storage medium

Abstract

The invention provides a resource collaborative scheduling method and device for a ship pipe fitting workshop, electronic equipment and a storage medium. The method comprises the steps that production parameters and decision variables of controllable states are determined; determining at least three objective functions and feature constraints according to the production parameters and the decision variables; performing hybrid coding on the decision variables according to the feature constraints to obtain a target coding structure; clustering the plurality of initial resource scheduling schemes based on the grey comprehensive correlation degree according to the objective function to obtain N initial resource scheduling scheme sets; performing adaptive local search based on meta-Ramak learning and population information interaction based on a chaos migration method on each initial resource scheduling scheme set to obtain N final resource scheduling scheme sets; and determining a target resource scheduling scheme of the ship pipe fitting workshop in the N final resource scheduling scheme sets. According to the invention, the resource scheduling scheme can be quickly obtained, and resource collaborative scheduling under multiple targets is realized.

Description

Technical Field

[0001] This application relates to the field of workshop production operation technology, and in particular to a resource collaborative scheduling method, device, electronic equipment and storage medium for a ship pipe fitting workshop. Background Technology

[0002] Shipbuilding is an important standard reflecting a nation's industrial technology level and a crucial lever for "improving efficiency, expanding functions, and increasing momentum." Ships are composed of thousands of parts, and pipe fittings are among the key components. Unlike components in general industrial products, ship pipe fitting workshops involve a wide variety of materials, many large and heavy parts, and irregular shapes. Their production workload accounts for a significant proportion of the overall shipbuilding workload and directly affects the shipbuilding schedule.

[0003] The introduction of mobile equipment resources such as gantry cranes and overhead cranes has improved the material handling efficiency of the ship fitting workshop, but it has also increased the difficulty of workshop management and scheduling. Collaborative operations are required between mobile logistics equipment and fixed processing equipment, as well as between mobile logistics equipment itself. This necessitates integrated and coordinated scheduling of resources, significantly increasing the decision-making space and constraints for scheduling. Conflicts between targets such as completion time, production energy consumption, and carbon emissions intensify, making production scheduling modes and models more complex. This presents a multi-objective scheduling challenge in a distributed, heterogeneous, hybrid production line with flexible collaboration of multiple mobile resources, representing one of the current bottlenecks facing ship fitting workshop production.

[0004] Therefore, a new resource collaborative scheduling method for ship fitting workshops is urgently needed to solve the above problems. Summary of the Invention

[0005] In view of this, this application provides a resource collaborative scheduling method, device, electronic device and storage medium for a ship pipe fitting workshop, which can quickly obtain resource scheduling schemes and realize resource collaborative scheduling under multiple objectives.

[0006] A first aspect of this application provides a resource collaborative scheduling method for a ship pipe fitting workshop, comprising: determining production parameters of the ship pipe fitting workshop and decision variables for characterizing the controllable state of the ship pipe fitting workshop; determining at least three objective functions for evaluating the merits of resource scheduling in the ship pipe fitting workshop, and feature constraints for limiting the feasibility of the decision variables, based on the production parameters and the decision variables; performing hybrid encoding on the decision variables according to the feature constraints to obtain a target encoding structure; obtaining multiple initial resource scheduling schemes with the target encoding structure; performing clustering based on grey comprehensive correlation degree on the multiple initial resource scheduling schemes according to the objective functions to obtain N initial resource scheduling scheme sets, where N is an integer greater than 1; performing adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method on each initial resource scheduling scheme set to obtain N final resource scheduling scheme sets; and determining the target resource scheduling scheme for the ship pipe fitting workshop from the N final resource scheduling scheme sets.

[0007] In one possible implementation, the production parameters include: the number of pipe fittings to be processed, batch size, number of pipe fittings per batch, number of production lines, number of processes, total number of processing equipment, number of processing equipment per process on each production line, number of overhead cranes, number of gantry cranes, and number of gantry cranes on each production line; the decision variables include: mixed production line allocation, pipe fitting allocation sorting, processing equipment allocation, processing equipment speed selection, overhead crane or gantry crane allocation, and overhead crane or gantry crane speed selection; the determination of at least three objective functions for evaluating the resource scheduling of the ship pipe fitting workshop based on the production parameters and the decision variables includes: determining the final completion time target of the pipe fittings to be processed, the energy consumption target of all equipment used in processing the pipe fittings to be processed, and the carbon emission target of the welding area equipment based on the production parameters and the decision variables.

[0008] In one possible implementation, before performing clustering based on grey comprehensive correlation of the multiple initial resource scheduling schemes according to the objective function, the method further includes: determining a first final target value based on the final completion time target and the energy consumption target, wherein the first final target value characterizes the correlation between the final completion time target and the energy consumption target and a reference final completion time target and a reference energy consumption target; determining a second final target value based on the carbon emission target, wherein the second final target value is used to define the upper and lower bounds of carbon emissions of the ship fittings production workshop; performing clustering based on grey comprehensive correlation of the multiple initial resource scheduling schemes according to the objective function includes: performing clustering based on grey comprehensive correlation of the multiple initial resource scheduling schemes according to the first final target value; determining the target resource scheduling scheme for the ship fittings workshop from the set of N final resource scheduling schemes includes: determining the target resource scheduling scheme from the set of N final resource scheduling schemes according to the second final target value.

[0009] In one possible implementation, determining the first final target value based on the final completion time target and the energy consumption target includes: constructing reference points based on historical best final completion time and historical best energy consumption; constructing comparison points based on the final completion time target and the energy consumption target; calculating the absolute correlation degree and relative correlation degree between the comparison points and the reference points, and calculating the grey comprehensive correlation degree based on the absolute correlation degree and the relative correlation degree, wherein the grey comprehensive correlation degree is the first final target value.

[0010] In one possible implementation, obtaining multiple initial resource scheduling schemes with the target encoding structure includes: obtaining M initial resource scheduling schemes, where M is an integer multiple of N; performing clustering based on gray comprehensive correlation degree on the multiple initial resource scheduling schemes according to the first final target value includes: calculating the gray comprehensive correlation degree corresponding to each initial resource scheduling scheme; sorting each initial resource scheduling scheme according to the gray comprehensive correlation degree, and clustering according to the difference in gray comprehensive correlation degree between adjacent initial resource scheduling schemes to obtain N sets of preliminary resource scheduling schemes; detecting whether the number of individuals in each set of preliminary resource scheduling schemes is equal to M / N; for a target set of preliminary resource scheduling schemes where the number of individuals in the set of preliminary resource scheduling schemes is not equal to M / N, calculating the congestion distance of each individual in the target set of preliminary resource scheduling schemes; adjusting the number of individuals in the target set of preliminary resource scheduling schemes to M / N according to the congestion distance to obtain the initial resource scheduling scheme set.

[0011] In one possible implementation, the step of performing adaptive local search based on meta-Lamarckian learning and population information interaction based on chaotic migration methods on each of the initial resource scheduling scheme sets includes: constructing S local search policies, where S is an integer greater than 1; during the training phase of meta-Lamarckian learning, executing the S local search policies on individuals in each of the initial resource scheduling scheme sets, calculating the improvement rate of each individual in each of the initial resource scheduling scheme sets when executing each of the local search policies, and updating the cumulative improvement rate of each of the local search policies according to the improvement rate; updating each of the initial resource scheduling scheme sets according to the cumulative improvement rate and selecting each... The initial probability of the local search strategy; during the working phase of meta-Lamarck learning, based on the initial probability, the target local search strategy for each set of initial resource scheduling schemes is determined by roulette wheel in the local search strategy; for each set of initial resource scheduling schemes, the target improvement rate after the individual executes the target local search strategy is calculated, and the target cumulative improvement rate of the target local search strategy is updated based on the target improvement rate; the initial probability is updated based on the target cumulative improvement rate; population information interaction based on the chaotic migration method is performed on the initial resource scheduling scheme set after updating the initial probability to obtain N sets of final resource scheduling schemes.

[0012] In one possible implementation, determining the target resource scheduling scheme for the ship fitting workshop from the N sets of final resource scheduling schemes includes: performing non-dominated sorting on the individuals in the N sets of final resource scheduling schemes to obtain a non-dominated solution set for the target individual in each set of final resource scheduling schemes; and decoding the non-dominated solution set to obtain the target resource scheduling scheme.

[0013] Secondly, embodiments of this application also provide a resource collaborative scheduling device for a ship pipe fitting workshop, comprising: a first determining module, a second determining module, an encoding module, an acquisition module, a clustering module, a calculation module, and a third determining module; the first determining module is used to determine the production parameters of the ship pipe fitting workshop and decision variables for characterizing the controllable state of the ship pipe fitting workshop; the second determining module is used to determine at least three objective functions for evaluating the merits of resource scheduling in the ship pipe fitting workshop based on the production parameters and the decision variables, and feature constraints for limiting the feasibility of the decision variables; the encoding module is used to perform mixed encoding on the decision variables based on the feature constraints. The code is used to obtain the target coding structure; the acquisition module is used to acquire multiple initial resource scheduling schemes with the target coding structure; the clustering module is used to perform clustering based on grey comprehensive correlation degree on the multiple initial resource scheduling schemes according to the objective function to obtain N initial resource scheduling scheme sets, where N is an integer greater than 1; the calculation module is used to perform adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method on each initial resource scheduling scheme set to obtain N final resource scheduling scheme sets; the third determination module is used to determine the target resource scheduling scheme of the ship pipe fitting workshop from the N final resource scheduling scheme sets.

[0014] Thirdly, embodiments of this application also provide an electronic device, which includes a processor and a memory. The memory is used to store instructions, and the processor is used to call the instructions in the memory, causing the electronic device to execute the resource collaborative scheduling method for the ship fitting workshop as described in the first aspect.

[0015] Fourthly, embodiments of this application also provide a storage medium that stores computer instructions, which, when executed on an electronic device, cause the electronic device to perform the resource collaborative scheduling method for a ship fitting workshop as described in the first aspect.

[0016] Compared with related technologies, the embodiments of this application have at least the following advantages: By determining at least three objective functions for evaluating the merits of resource scheduling in the ship fitting workshop based on production parameters and decision variables, and feature constraints for limiting the feasibility of decision variables, the actual production characteristics of the ship fitting workshop can be comprehensively considered. By performing mixed encoding on decision variables according to feature constraints, a target encoding structure is obtained, which can characterize the core decision variables and their distribution and regional characteristics, ensuring the feasibility and completeness of the scheduling solution in subsequent steps, and providing structured support for the generation, evaluation and evolution of solutions in subsequent algorithms. By obtaining multiple initial resource scheduling schemes set as target encoding structures, and then clustering multiple initial resource scheduling schemes based on objective functions, since there are at least three objective functions, after adaptive local search and population information interaction on the set of initial resource scheduling schemes, the target resource scheduling scheme for the ship fitting workshop is determined from the N final resource scheduling scheme sets obtained. On the one hand, it can quickly obtain excellent distributed multi-zone production multi-objective scheduling schemes to meet the green and intelligent scheduling needs of the ship fitting workshop. On the other hand, the generated target resource scheduling scheme comprehensively considers at least three objective optimization problems, realizing multi-objective collaborative scheduling.

[0017] It is worth noting that, through the design of a distributed multi-zone hybrid coding method for multi-mobility resource collaboration, and by combining co-evolutionary theory and meta-Lamarck learning method, an adaptive learning multi-population co-evolutionary multi-objective scheduling algorithm is established. This algorithm can quickly obtain excellent distributed multi-zone production multi-objective scheduling schemes, meet the green and intelligent scheduling requirements of ship fitting workshops, and also enrich the basic theory of multi-objective optimization.

[0018] The technical effects achieved by the second, third, and fourth aspects mentioned above are similar to those achieved by the corresponding technical means in the first aspect, and will not be repeated here. Attached Figure Description

[0019] Figure 1 A flowchart illustrating the steps of a resource collaborative scheduling method for a ship fittings workshop provided in an embodiment of this application; Figure 2 This is a schematic diagram of the encoding process of the target encoding structure provided in an embodiment of this application; Figure 3 A schematic diagram illustrating the fitness evaluation process based on grey comprehensive correlation provided in an embodiment of this application; Figure 4 A flowchart of an adaptive learning multi-population co-evolutionary multi-objective intelligent scheduling optimization algorithm provided in an embodiment of this application; Figure 5A functional block diagram of a resource collaborative scheduling device for a ship fitting workshop provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0020] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0021] The following description sets forth many specific details to provide a full understanding of this application. The described embodiments are only some, not all, of the embodiments of this application.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0023] It should be further noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0024] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and drawings of this application are used to distinguish similar objects, not to describe a specific order or sequence.

[0025] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] For ease of understanding, some concepts related to the embodiments of this application are illustrated and explained by way of example for reference.

[0027] Heuristic rules are a decision-making technique based on experience-based judgment. They quickly find approximate solutions to problems through simplified rules or models and are widely used in computer science, artificial intelligence, and complex optimization. The core of heuristic rules lies in utilizing prior knowledge and empirical rules to overcome the limitations of traditional precise methods by identifying combinations of features or behavioral patterns.

[0028] Grey comprehensive correlation degree: It is an index in grey system theory used to measure the degree of correlation between the actual data of two sequences and their rate of change relative to the starting point. It combines the calculation methods of grey absolute correlation degree and grey relative correlation degree.

[0029] Meta-Lamarck learning is an intelligent regulatory strategy that optimizes the learning efficiency of individuals in evolutionary algorithms (such as genetic algorithms). Its core lies in dynamically and selectively applying the Lamarck learning mechanism to achieve a balance between algorithm convergence speed and global search capability. Meta-Lamarck learning refers to an intelligent algorithm design paradigm that integrates the Lamarck learning mechanism and meta-learning ideas. Its core lies in dynamically regulating the underlying Lamarckian gene rewriting operations through high-level meta-strategies to achieve adaptive evolution of the algorithm.

[0030] Chaotic migration method: The chaotic migration method of population interaction is an optimization strategy that combines chaos theory and population dynamics. It enhances the diversity of the population through the randomness and ergodicity of chaotic mapping, and at the same time uses the migration mechanism to achieve adaptive optimization in dynamic environments.

[0031] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating one embodiment of the resource collaborative scheduling method for a ship fittings workshop according to this application. The order of steps in this flowchart can be changed, and some steps can be omitted, depending on different requirements.

[0032] It should be noted that the resource collaborative scheduling method for ship pipe fitting workshops in this application embodiment can be applied to production operation scenarios in ship pipe fitting workshops. The executing entity can be a resource collaborative scheduling device in the ship pipe fitting workshop. For example, in a production operation scenario in a ship pipe fitting workshop, the resource collaborative scheduling device can perform real-time resource collaborative scheduling of ship pipe fittings. Of course, the resource collaborative scheduling method for ship pipe fitting workshops can also be applied to other scenarios requiring resource collaborative scheduling, and this application does not specifically limit its application in this regard.

[0033] The specific process of this embodiment is as follows: Figure 1 As shown, it includes the following steps: S101, determine the production parameters of the ship fittings workshop, as well as the decision variables used to characterize the controllable state of the ship fittings workshop.

[0034] In some embodiments, production parameters include: the number of pipe fittings to be processed, batch size, the number of pipe fittings in each batch, the number of production lines, the number of processes, the total number of processing equipment, the number of processing equipment for each process on each production line, the number of overhead cranes, the number of gantry cranes, and the number of gantry cranes on each production line; decision variables include: mixed production line allocation, pipe fitting allocation sorting, processing equipment allocation, processing equipment speed selection, overhead crane or gantry crane allocation, and overhead crane or gantry crane speed selection.

[0035] It is understood that this embodiment does not specifically limit the types of production parameters and decision variables corresponding to the ship fitting workshop, and can be set according to actual needs.

[0036] S102, determine at least three objective functions for evaluating the merits of resource scheduling in the ship fitting workshop based on production parameters and decision variables, as well as feature constraints for limiting the feasibility of decision variables.

[0037] In some embodiments, the objective function includes: the final completion time target of the pipe fitting to be processed, the energy consumption target of all equipment used in processing the pipe fitting to be processed, and the carbon emission target of the equipment in the welding area.

[0038] To facilitate understanding, the following example uses a ship pipe fittings workshop to illustrate how this embodiment determines the objective function and characteristic constraints based on production parameters and decision variables: Decision variables include: hybrid pipeline allocation Pipe fitting allocation and sorting Allocation of processing equipment Processing equipment speed selection ; Crane or gantry crane allocation Speed ​​selection for overhead cranes or gantry cranes (including moving trolleys and spreading devices) .

[0039] Production parameters include: quantity of pipe fittings to be processed N ,batch , No. b The quantity of pipe fittings in the batch is , Number of production lines Number of processes Number of all processing equipment The number of processing equipment in each process on each production line Number of overhead cranes Number of gantry cranes The number of gantry cranes on each production line Hybrid production line Medium cutting machine Cutting a single piece Time and power standby power Hybrid production line Medium pipe bending machine Bending a single piece Time and power standby power Hybrid production line Lieutenant Colonel's Aircraft Correcting a single piece Time and power standby power Hybrid production line Medium welding machine Welding a single piece Time ,power and unit carbon emissions standby power Hybrid production line medium grinding machine Polishing a single piece Time and power standby power Hybrid production line Medium pump press Inspect a single item Time and power standby power The overhead crane's operating speed under beam load and power Crane trolley load operating speed and power Overhead crane load operating speed and power The overhead crane's speed when running unloaded on the crossbeam. and power Crane trolley no-load running speed and power No-load running speed of overhead cranes and power ; Gantry crane operating speed under load on ground rail and power gantry crane trolley load running speed and power gantry crane lifting equipment load operating speed and power The gantry crane's unloaded running speed on the ground rails and power gantry crane trolley no-load running speed and power Gantry crane lifting equipment no-load running speed and power Standby power of overhead cranes and gantry cranes , .

[0040] The types and determination methods of feature constraints include: Single-piece production characteristic constraints of gantry cranes: ;in, , They are batch Middle pipe fittings , The start and end times of production on a certain processing machine. , and For gantry cranes and trolley lifting equipment respectively , , Three-way movement of single piece Time; Constraints on bulk transportation of gantry cranes: ;in, , These are batch pipe fittings , Start and end production times , and For the gantry crane and trolley respectively , , Three-way cross-regional batch movement Time; Constraints on bulk transport by overhead crane: ;in, , These are batch pipe fittings , Start and end production times , and respectively the overhead crane and the trolley , , Three-way cross-regional batch movement Time; Buffer constraints: ;in, , These are the current quantity and maximum capacity of pipe fittings within the buffer zone of production area A, respectively. Spatial constraints of the gantry crane: ;in, Indicates gantry crane Spatial location, These represent the upper and lower limits of the movement range of the gantry crane, its trolley, and its lifting gear, respectively. These respectively indicate that the gantry crane, its trolley, and its lifting gear complete their tasks within their respective production areas. The unique planning coordinates; Crane spatial constraints: ;in, Indicates overhead crane Spatial location, These represent the upper and lower limits of the movement range of the overhead crane, its trolley, and the lifting gear, respectively. These respectively indicate that the overhead crane, its trolley, and the lifting gear complete their tasks within their respective production areas. The unique planning coordinates; Two gantry cranes and Safety clearance constraints: Among them, when batches of pipe fittings are used... , When there are overlapping work periods, , , These represent batches of pipe fittings. , Assigned to gantry crane and Complete the transfer. For minimum safe distance, It is a large negative integer. , ; Overhead crane cross-operation line transportation conflict constraints: ;in, It is a 0-1 variable; when it is 1, it represents an overhead crane. First transfer to the mixed assembly line Batch of pipe fittings , , These are batch pipe fittings , End time of transit , They are respectively overhead cranes Transfer Mixed Assembly Line Batch of pipe fittings The no-load and load movement time, For overhead crane In the mixed production line and Travel time between; Constraints on the coordination and connection between overhead cranes and stationary machining equipment: ;in, It is an overhead crane bulk shipping The start time, It is a batch processing equipment that completes the process. The The time required for each process It is an overhead crane Move to process The time required for processing; Constraints on the coordination and connection between gantry cranes and fixed processing equipment: ;in, It's a gantry crane. single item transfer The start time, A single piece is completed by fixed processing equipment. The The time required for each process It's a gantry crane. Move to process The time required for the process.

[0041] The types and methods for determining the objective function include: Final completion time target for pipe fittings to be processed: ;in, Indicates the first The time it takes for a batch of pipe fittings to be transported away by overhead crane after the completion of the mixed production line; Energy consumption targets for all equipment used in pipe fitting processing: ;in, Energy consumption for processing and waiting of all fixed equipment. This refers to the energy consumption of all overhead cranes and their trolleys during idle / loaded movement and waiting within the workshop. Energy consumption for all gantry cranes and their trolleys during no-load / load movement and waiting time within the workshop; Carbon emission targets for welding area equipment: ;in, , They are respectively hybrid production lines Medium welding machine Welding single piece Time and unit carbon emissions. It is a 0-1 variable, and a value of 1 indicates a single item. Assigned to the hybrid production line Medium welding machine welding.

[0042] Therefore, the multi-objective scheduling optimization function for the ship fittings workshop is: .

[0043] S103, perform hybrid encoding on the decision variables according to the feature constraints to obtain the target encoding structure.

[0044] To facilitate understanding, the following will be combined with... Figure 2 This embodiment provides a detailed explanation of how the target encoding structure is obtained: Please refer to Figure 2 This is a schematic diagram illustrating the encoding process of the target encoding structure provided in this application embodiment. Using a random key integer encoding method, A batch of ship pipe fittings were allocated to A single production line enables the coding of pipe fitting production tasks across multiple mixed production lines; in any mixed production line... In each production area, the first A batch of pipe fittings is randomly assigned to a specific processing equipment capable of processing that batch; within the same batch, pipe fittings are sorted using a process coding method to obtain the processing sequence of pipe fittings on a single processing equipment within a specific production area; the processing speed of each processing equipment is obtained through random coding; for any mixed production line... Each batch of pipe fittings in each production area One gantry crane is assigned based on the operating capacity range code. Or overhead crane and assign a selection to its encoding. , , Movement speed in three directions.

[0045] It is worth noting that the above-mentioned distributed multi-region hybrid encoding is a direct mapping of the multi-objective scheduling optimization function in the preceding steps, characterizing the core decision variables in the function and their distribution and regional characteristics. This encoding structure not only ensures the feasibility and completeness of the scheduling solution in subsequent steps, but also provides structured support for the generation, evaluation and evolution of solutions in the algorithm.

[0046] S104, obtain multiple initial resource scheduling schemes with target encoding structures.

[0047] In some embodiments, a certain number of initial resource scheduling schemes are generated based on heuristic rules (e.g., h1: maximum / minimum speed of gantry crane, h2: maximum / minimum speed of overhead crane, h3: priority of earliest idle equipment, h4: priority of shortest queue, h5: equipment capacity matching, h6: energy consumption balance rule, h7: carbon emission sensitive selection, etc.), and the remaining initial resource scheduling schemes are generated in a random manner.

[0048] In some embodiments, M initial resource scheduling schemes are obtained, where M is an integer multiple of N.

[0049] S105, determine a first final target value based on the final completion time target and the energy consumption target, wherein the first final target value characterizes the correlation between the final completion time target and the energy consumption target and the reference final completion time target and the reference energy consumption target.

[0050] In some embodiments, a reference point is constructed based on the historical best final completion time and historical best energy consumption; a comparison point is constructed based on the final completion time target and energy consumption target; the absolute correlation degree and relative correlation degree between the comparison point and the reference point are calculated, and the gray comprehensive correlation degree is calculated based on the absolute correlation degree and relative correlation degree, wherein the gray comprehensive correlation degree is the first final target value.

[0051] S106, determine a second final target value based on the carbon emission target, wherein the second final target value is used to define the upper and lower limits of carbon emissions from the ship fittings production workshop.

[0052] Specifically, the three-objective scheduling function suffers from theoretical complexity, algorithmic difficulties, and unfavorable result analysis, necessitating appropriate dimensionality reduction. Therefore, all objective functions are divided into two layers, with the final completion time objective being... and all equipment energy consumption targets For the overall common objective, the carbon emission target of the welding area equipment As sub-objectives for partitioning, the two dimensionality-reduced objectives together form the basis for the fitness evaluation of the algorithm in subsequent steps.

[0053] To facilitate understanding, the following will be combined with... Figure 3 This embodiment provides a detailed explanation of how to determine the overall layer common objective and the partitioned sub-objectives: For the overall common objective, an initial reference point is constructed by referring to the optimal objective data from the actual production of the ship fittings workshop. And store it in the reference point library; define the multi-objective function value of the population evolution individual as the comparison point. =[ , ], For the population number One solution; based on the grey comprehensive relational analysis method, comparison points are identified. With reference point The degree of correlation between them was determined by calculating the absolute and relative correlation between the two. and Finally, the grey comprehensive correlation degree was calculated. A new multi-objective optimization population individual fitness assessment coordination criterion is defined, based on grey comprehensive correlation degree. The larger the value, the higher the quality of the population solution; obtaining =[ , The optimal value in each target direction is used to form a new reference point; the new reference point is compared with the old reference points in the reference point library to achieve dynamic updating of the historical reference point library.

[0054] For each sub-objective, based on the operating time and power of the welding area machines and the unit carbon emissions during historical scheduling cycles, the maximum carbon emissions that the area may generate are estimated. At the same time, referring to the green production indicators of the workshop, the maximum carbon emissions are adjusted downwards and set as the upper bound of the model's carbon emissions. The minimum carbon emission value is determined by using a scheduling scheme that selects only the welding machine with the lowest unit carbon emission, and this minimum value is set as the lower bound of the carbon emission target. The upper and lower bounds constructed above The carbon emissions in the welding zone are applied as an "interval constraint" and used as a boundary constraint in subsequent steps. In the population evolution, if the carbon emissions in the welding zone of an individual solution exceed the upper bound, it is determined to be infeasible, and feasible solutions with carbon emission values ​​falling within this interval are selected first.

[0055] S107. Based on the first final target value, cluster the multiple initial resource scheduling schemes according to the gray comprehensive correlation degree to obtain a set of N initial resource scheduling schemes.

[0056] In some embodiments, a threshold for clustering is defined based on the individual fitness assessment coordination mechanism of grey comprehensive association. , , They are populations The maximum and minimum grey comprehensive correlation of individuals in the middle, The number of subpopulations after clustering is determined, and the difference in grey comprehensive correlation degree and crowding distance between two adjacent individuals are used to determine whether two individuals are clustered into one group.

[0057] Specifically, based on the grey comprehensive correlation degree of each initial resource scheduling scheme, multiple initial resource scheduling schemes are sorted. Then, the difference in grey comprehensive correlation degree between two adjacent initial resource scheduling schemes is compared. If the difference is less than or equal to a preset threshold, the two initial resource scheduling schemes are clustered into one group. After clustering in the above manner, there may be cases where the number of initial resource scheduling schemes in some sets exceeds the preset number. For example, if 90 initial resource scheduling schemes need to be clustered into three initial resource scheduling scheme sets, the above clustering method based on grey comprehensive correlation degree generates set A with 32 initial resource scheduling schemes, set B with 27 initial resource scheduling schemes, and set C with 31 initial resource scheduling schemes. By calculating the congestion distance of each initial resource scheduling scheme in set A and set C respectively, two individuals are selected from set A and placed in set B, and one individual is selected from set C and placed in set B.

[0058] S108. For each initial resource scheduling scheme set, perform adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method to obtain N final resource scheduling scheme sets.

[0059] In some embodiments, adaptive local search based on meta-Lamarckian learning and population information interaction based on chaotic transfer methods are performed on each initial resource scheduling scheme set, including: constructing S local search policies, where S is an integer greater than 1; during the training phase of meta-Lamarckian learning, executing the S local search policies on individuals in each initial resource scheduling scheme set, calculating the improvement rate of individuals in each initial resource scheduling scheme set when executing each local search policy, and updating the cumulative improvement rate of each local search policy according to the improvement rate; updating the initial probability of each initial resource scheduling scheme set selecting each local search policy according to the cumulative improvement rate; during the working phase of meta-Lamarckian learning, determining the target local search policy of each initial resource scheduling scheme set through roulette wheel selection based on the initial probability; for each initial resource scheduling scheme set, calculating the target improvement rate after individuals execute the target local search policy, and updating the target cumulative improvement rate of the target local search policy according to the target improvement rate; updating the initial probability according to the target cumulative improvement rate; and performing population information interaction based on chaotic transfer methods on the initial resource scheduling scheme sets after updating the initial probabilities to obtain N final resource scheduling scheme sets.

[0060] To facilitate understanding, the following example uses a ship fittings production workshop. Figure 4 This embodiment provides a detailed explanation of how adaptive local search and population information exchange are performed on each initial set of resource scheduling schemes: Please refer to Figure 4The flowchart below shows the adaptive learning multi-population co-evolution multi-objective intelligent scheduling optimization algorithm provided in the embodiments of this application.

[0061] Based on the distributed multi-zone production characteristics of the ship pipe fitting workshop, S local search strategies are constructed (e.g., N1: switching the speed gear of the overhead crane or gantry crane, N2: changing the processing order between different batches of pipe fittings, N3: changing the processing order of pipe fittings within the same batch, N4: adjusting the operation allocation of fixed processing equipment for a certain process, N5: switching the speed gear of fixed processing equipment, etc.), forming a local search strategy library. The meta-Lamarck learning method is introduced, and in each generation of the initial resource scheduling scheme set (hereinafter referred to as the population), a suitable local search strategy is dynamically and adaptively selected from the local search strategy library for each subpopulation based on the cumulative improvement rate.

[0062] The adaptive local search based on meta-Lamarck learning is as follows: 1. Training Phase: (1) Obtain the subpopulations after dividing the first generation population. And assess the fitness of each individual.

[0063] (2) For each subpopulation Perform the following operations: The cumulative improvement rate of each local search strategy and choice probability Set to 0; In each individual Execute all local search strategies. Calculate individual In executing the strategy Improvement rate at time Update cumulative improvement rate ; According to the formula Update subpopulation The probability of choosing each strategy.

[0064] 2. Work Phase: For each subpopulation Perform the following operations in a loop: (1) Select the local search strategy to be executed for each subpopulation by biased roulette wheel selection; (2) Hypothetical subpopulation The chosen local search strategy is Then for each individual in this subpopulation All execute strategies And calculate the corresponding improvement rate. Update cumulative improvement rate ; (3) According to the formula Update subpopulation The probability of choosing each strategy.

[0065] After completing the adaptive local search based on meta-Lamarck learning, a chaotic migration method is employed to determine the chaotic sequence number of the individuals involved in the migration interaction between each pair of subpopulations. Every two subpopulations achieve The interactions of a number of individuals are as follows: For any two subpopulations , Perform the following operations: According to the Logistic chaos equation Iterate to obtain a definite chaotic sequence. ,in As a chaotic factor, initial value , No more than 20% of the subpopulation size, i.e. , , Sub-populations and The scale; make According to the formula Obtaining an integer chaotic sequence ,in It is a rounding function in the negative direction; With sequence Each integer in the index serves as an individual index, which will be used to index the subpopulation. and Individuals within the system exchange [the information].

[0066] Perform non-dominated sorting on all individuals and calculate crowding distance, then select offspring individuals; determine whether the algorithm termination condition is met: the termination condition is whether the maximum number of iterations has been reached. If the maximum number of iterations has not been reached, re-cluster the population and return to step 1; if the maximum number of iterations has been reached, then determine the final set of resource scheduling schemes as the final set of resource scheduling schemes.

[0067] S109, determine the target resource scheduling scheme for the ship fitting workshop from the set of N final resource scheduling schemes.

[0068] In some embodiments, individuals in the N final resource scheduling scheme sets are sorted by non-dominated order to obtain the non-dominated solution set of the target individual in each final resource scheduling scheme set; the non-dominated solution set is decoded to obtain the target resource scheduling scheme.

[0069] Compared with related technologies, the embodiments of this application have at least the following advantages: By determining at least three objective functions for evaluating the merits of resource scheduling in the ship fitting workshop based on production parameters and decision variables, and feature constraints for limiting the feasibility of decision variables, the actual production characteristics of the ship fitting workshop can be comprehensively considered. By performing mixed encoding on decision variables according to feature constraints, a target encoding structure is obtained, which can characterize the core decision variables and their distribution and regional characteristics, ensuring the feasibility and completeness of the scheduling solution in subsequent steps, and providing structured support for the generation, evaluation and evolution of solutions in subsequent algorithms. By obtaining multiple initial resource scheduling schemes set as target encoding structures, and then clustering multiple initial resource scheduling schemes based on objective functions, since there are at least three objective functions, after adaptive local search and population information interaction on the set of initial resource scheduling schemes, the target resource scheduling scheme for the ship fitting workshop is determined from the N final resource scheduling scheme sets obtained. On the one hand, it can quickly obtain excellent distributed multi-zone production multi-objective scheduling schemes to meet the green and intelligent scheduling needs of the ship fitting workshop. On the other hand, the generated target resource scheduling scheme comprehensively considers at least three objective optimization problems, realizing multi-objective collaborative scheduling.

[0070] Based on the same idea as the resource collaborative scheduling method for a ship pipe fitting workshop in the above embodiments, this application also provides a resource collaborative scheduling device for a ship pipe fitting workshop. This device can be used to execute the resource collaborative scheduling method for the ship pipe fitting workshop described above. For ease of explanation, the structural schematic diagram of the resource collaborative scheduling device embodiment for a ship pipe fitting workshop only shows the parts related to the embodiments of this application. Those skilled in the art will understand that the illustrated structure does not constitute a limitation on the device, and it may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0071] like Figure 5 As shown, the resource collaborative scheduling device 50 for a ship fittings workshop includes a first determining module 501, a second determining module 502, an encoding module 503, an acquisition module 504, a clustering module 505, a calculation module 505, and a third determining module 507. In some embodiments, the above modules can be programmable software instructions stored in memory and executable by a processor. It is understood that in other embodiments, the above modules can also be program instructions or firmware embedded in a processor.

[0072] The first determining module 501 is used to determine the production parameters of the ship fitting workshop and the decision variables used to characterize the controllable state of the ship fitting workshop. The second determining module 502 is used to determine at least three objective functions for evaluating the merits of resource scheduling in the ship fitting workshop based on the production parameters and the decision variables, as well as feature constraints for limiting the feasibility of the decision variables; Encoding module 503 is used to perform hybrid encoding on the decision variables according to the feature constraints to obtain the target encoding structure; The acquisition module 504 is used to acquire multiple initial resource scheduling schemes with target encoding structures; Clustering module 505 is used to cluster multiple initial resource scheduling schemes according to the objective function to obtain a set of N initial resource scheduling schemes, where N is an integer greater than 1; The calculation module 506 is used to perform adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method on each of the initial resource scheduling scheme sets to obtain N final resource scheduling scheme sets. The third determining module 507 is used to determine the target resource scheduling scheme for the ship fitting workshop from the N sets of final resource scheduling schemes.

[0073] The resource collaborative scheduling device 50 for the ship pipe fitting workshop provided in the above embodiments can realize the technical solutions described in the above embodiments of the resource collaborative scheduling method for the ship pipe fitting workshop. The specific implementation principles of each module or unit can be found in the corresponding content in the above embodiments of the resource collaborative scheduling method for the ship pipe fitting workshop, and will not be repeated here.

[0074] Please refer to Figure 6 , Figure 6 This is a schematic diagram of an embodiment of the electronic device of this application.

[0075] In some embodiments, processor 601 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 602 or process data, such as the resource collaborative scheduling method for ship pipe fitting workshop in this invention.

[0076] In some embodiments, processor 601 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 601 may be local or remote. In some embodiments, processor 601 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, intranet, multi-cloud, etc., or any combination thereof.

[0077] In some embodiments, memory 602 may be an internal storage unit of electronic device 600, such as a hard disk or memory of electronic device 600. In other embodiments, memory 602 may also be an external storage device of electronic device 600, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 600.

[0078] Furthermore, the memory 602 may include both internal storage units of the electronic device 600 and external storage devices. The memory 602 is used to store application software and various types of data installed on the electronic device 600.

[0079] In some embodiments, display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 603 is used to display information from electronic device 600 and to display visual user applications. Components 601-603 of electronic device 600 communicate with each other via a system bus.

[0080] In one embodiment, when processor 601 executes the resource coordination scheduler in memory 602, the following steps can be implemented: Determine the production parameters of the ship fittings workshop, as well as the decision variables used to characterize the controllable state of the ship fittings workshop; Based on the production parameters and the decision variables, at least three objective functions are determined to evaluate the merits of resource scheduling in the ship fitting workshop, as well as feature constraints to limit the feasibility of the decision variables; The decision variables are hybrid-encoded according to the feature constraints to obtain the target encoding structure; Multiple initial resource scheduling schemes are obtained, wherein each of the initial resource scheduling schemes is set as the target encoding structure; Clustering the multiple initial resource scheduling schemes according to the objective function yields a set of N initial resource scheduling schemes, where N is an integer greater than 1; For each of the initial resource scheduling scheme sets, an adaptive local search and population information interaction are performed to obtain N final resource scheduling scheme sets; The target resource scheduling scheme for the ship fitting workshop is determined from the set of N final resource scheduling schemes.

[0081] It should be understood that when the processor 601 executes the resource coordination scheduling program in the memory 602, in addition to the functions mentioned above, it can also perform other functions, as detailed in the description of the corresponding method embodiments above.

[0082] Furthermore, this embodiment of the invention does not specifically limit the type of electronic device 600 mentioned. Electronic device 600 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the invention, electronic device 600 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0083] Accordingly, this application also provides a storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions of the resource collaborative scheduling method for the ship pipe fitting workshop provided in the above-described method embodiments.

[0084] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0085] The resource collaborative scheduling method, device, electronic equipment, and storage medium for ship pipe fitting workshops provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for coordinating and scheduling resources in a ship pipe workshop, characterized in that, The method comprises the following steps: determining production parameters of a ship pipe fitting workshop, and decision variables for characterizing controllable states of the ship pipe fitting workshop; determining at least three objective functions for evaluating the pros and cons of resource scheduling of the ship pipe fitting workshop according to the production parameters and the decision variables, and characteristic constraints for limiting the feasibility of the decision variables; mixing coding of the decision variables according to the characteristic constraints to obtain a target coding structure; obtaining a plurality of initial resource scheduling schemes with the target coding structure; clustering the plurality of initial resource scheduling schemes based on grey comprehensive correlation degree according to the target functions to obtain N initial resource scheduling scheme sets, N being an integer greater than 1; performing adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method on each of the initial resource scheduling scheme sets to obtain N final resource scheduling scheme sets; determining a target resource scheduling scheme of the ship pipe fitting workshop in the N final resource scheduling scheme sets.

2. The method of claim 1, wherein, The production parameters include: the number of pipe fittings to be processed, the batch quantity, the number of pipe fittings in each batch, the number of flow lines, the number of processes, the number of all processing devices, the number of processing devices on each process of each flow line, the number of overhead cranes, the number of gantry cranes, and the number of gantry cranes on each flow line; The decision variables include: mixed flow line allocation, pipe fitting allocation and sorting, processing device allocation, processing device speed selection, overhead crane or gantry crane allocation, and overhead crane or gantry crane speed selection; The determination of at least three objective functions for evaluating the pros and cons of resource scheduling of the ship pipe fitting workshop according to the production parameters and the decision variables comprises: determination of the last completion time target of the pipe fittings to be processed, the energy consumption target of all devices used for processing the pipe fittings to be processed, and the carbon emission target of the welding area device according to the production parameters and the decision variables.

3. The method of coordinating resources of a ship pipe plant according to claim 2, wherein, Before the clustering of the plurality of initial resource scheduling schemes based on grey comprehensive correlation degree according to the target functions, the method further comprises: determination of a first final target value according to the last completion time target and the energy consumption target, wherein the first final target value represents the correlation degree between the last completion time target and the energy consumption target and a reference last completion time target and a reference energy consumption target; determination of a second final target value according to the carbon emission target, wherein the second final target value is used to limit the upper and lower bounds of the carbon emission of the ship pipe fitting production workshop; The clustering of the plurality of initial resource scheduling schemes based on grey comprehensive correlation degree according to the target functions comprises: clustering of the plurality of initial resource scheduling schemes based on grey comprehensive correlation degree according to the first final target value; The determination of a target resource scheduling scheme of the ship pipe fitting workshop in the N final resource scheduling scheme sets comprises: determination of the target resource scheduling scheme in the N final resource scheduling scheme sets according to the second final target value.

4. The method of coordinating resources of a ship pipe plant according to claim 3, wherein, The determination of a first final target value according to the last completion time target and the energy consumption target comprises: construction of a reference point according to a historical optimal last completion time and a historical optimal energy consumption; construct a comparison point according to the last completion time target and the energy consumption target; calculate an absolute correlation degree and a relative correlation degree between the comparison point and the reference point, and calculate a grey comprehensive correlation degree according to the absolute correlation degree and the relative correlation degree, wherein the grey comprehensive correlation degree is the first final target value.

5. The method of coordinating resources of a ship pipe plant according to claim 4, wherein, The obtaining a plurality of initial resource scheduling schemes with the target coding structure comprises: obtaining M initial resource scheduling schemes, wherein M is an integer multiple of N; The grey comprehensive correlation degree-based clustering of the plurality of initial resource scheduling schemes according to the first final target value comprises: calculating the grey comprehensive correlation degree corresponding to each initial resource scheduling scheme; sorting each initial resource scheduling scheme according to the grey comprehensive correlation degree, and clustering according to the difference between the grey comprehensive correlation degrees of adjacent initial resource scheduling schemes to obtain N preliminary resource scheduling scheme sets; detecting whether the number of individuals in each preliminary resource scheduling scheme set is equal to M / N; for a target preliminary resource scheduling scheme set in which the number of individuals is not equal to M / N, calculating the crowding distance of each individual in the target preliminary resource scheduling scheme set; adjusting the number of individuals in the target preliminary resource scheduling scheme set to M / N according to the crowding distance to obtain the initial resource scheduling scheme set.

6. The method of claim 1, wherein, The meta-Lamarck learning-based adaptive local search and the population information interaction based on the chaotic migration method for each initial resource scheduling scheme set comprise: constructing S local search strategies, wherein S is an integer greater than 1; in the training stage of the meta-Lamarck learning, executing S local search strategies on individuals of each initial resource scheduling scheme set, calculating the improvement rate of each individual of each initial resource scheduling scheme set when each local search strategy is executed, and updating the cumulative improvement rate of each local search strategy according to the improvement rate; updating the initial probability of each local search strategy selected by each initial resource scheduling scheme set according to the cumulative improvement rate; in the working stage of the meta-Lamarck learning, determining the target local search strategy of each initial resource scheduling scheme set by roulette from the local search strategies according to the initial probability; for each initial resource scheduling scheme set, calculating the target improvement rate of the individual after the target local search strategy is executed, and updating the target cumulative improvement rate of the target local search strategy according to the target improvement rate; updating the initial probability according to the target cumulative improvement rate; performing the population information interaction based on the chaotic migration method for the initial resource scheduling scheme set after the initial probability is updated to obtain N final resource scheduling scheme sets.

7. The method of claim 1, wherein, The determining the target resource scheduling scheme of the ship pipe fitting workshop from the N final resource scheduling scheme sets comprises: performing non-dominated sorting on individuals in the N final resource scheduling scheme sets to obtain a non-dominated solution set of target individuals in each final resource scheduling scheme set; Decode the non-dominated solution set to obtain the target resource scheduling scheme.

8. A device for coordinating and scheduling resources of a ship pipe workshop, characterized in that, Comprise: The first determination module, the second determination module, the encoding module, the acquisition module, the clustering module, the calculation module and the third determination module; The first determination module is used for determining the production parameters of the ship pipe fitting workshop, and determining the decision variables for representing the controllable state of the ship pipe fitting workshop; The second determination module is used for determining at least three objective functions for evaluating the advantages and disadvantages of the ship pipe fitting workshop resource scheduling according to the production parameters and the decision variables, and determining the characteristic constraints for limiting the feasibility of the decision variables; The encoding module is used for mixing encoding the decision variables according to the characteristic constraints to obtain a target encoding structure; The acquisition module is used for acquiring a plurality of initial resource scheduling schemes with the target encoding structure; The clustering module is used for clustering a plurality of initial resource scheduling schemes based on gray comprehensive correlation according to the objective functions to obtain N initial resource scheduling scheme sets, N being an integer greater than 1; The calculation module is used for performing adaptive local search based on meta-Lamarck learning and population information interaction based on chaotic migration method on each initial resource scheduling scheme set to obtain N final resource scheduling scheme sets; The third determination module is used for determining the target resource scheduling scheme of the ship pipe fitting workshop in the N final resource scheduling scheme sets.

9. An electronic device comprising a processor and a memory, wherein The memory is used for storing instructions, and the processor is used for calling the instructions in the memory, so that the electronic device executes the resource collaborative scheduling method of the ship pipe fitting workshop as claimed in any one of claims 1 to 7.

10. A storage medium, characterized by The storage medium stores computer instructions, which, when executed on an electronic device, cause the electronic device to perform the resource collaborative scheduling method of the ship pipe fitting workshop as claimed in any one of claims 1 to 7.

Images