Manufacturing service combination optimization analysis method and system for cloud manufacturing platform

Abstract

The invention provides a manufacturing service combination optimization analysis method and system for a cloud manufacturing platform, and relates to the field of cloud manufacturing, and the method comprises the steps: obtaining a plurality of candidate services with similar functions in a large-scale service environment; classifying the plurality of candidate services through a K-means algorithm, and screening the service cluster to obtain a plurality of target services; solving the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to obtain a plurality of Pareto solutions; and according to a preference weight preset by the user, determining an optimal solution which most meets the requirement from the plurality of Pareto solutions, and determining a manufacturing service combination MSC corresponding to the optimal solution. According to the method, the fast non-dominated sorting genetic algorithm is adopted to carry out dual-objective optimization, multiple Pareto optimal solutions are provided while the QoS and sustainability are maximized, and the method is suitable for the cloud manufacturing platform needing efficient, reliable and sustainable service combination and especially has remarkable advantages in large-scale and complex manufacturing tasks.

Description

Technical Field

[0001] The present application relates to the field of cloud manufacturing technology, and in particular to a manufacturing service combination optimization analysis method and system for a cloud manufacturing platform. Background Art

[0002] Cloud manufacturing is a new type of manufacturing model. It integrates distributed manufacturing resources and capabilities to achieve resource sharing and collaboration to meet the diverse and personalized needs of users. It is crucial to promote the realization of intelligent, networked and digital manufacturing. In the cloud manufacturing environment, since simple services cannot meet the diverse needs of manufacturing tasks in complex business processes, manufacturing service composition (MSC) has emerged, and MSC has become the key to achieving complex manufacturing tasks. MSC refers to combining individual services in a certain way to form an integrated solution that can meet the needs of complex business processes. It provides a flexible and scalable way to build a large-scale industrial cloud service platform. There are some extensive methods and theories to solve the MSC problem.

[0003] In the field of cloud manufacturing, the existing service composition technology mainly focuses on the following solutions: (1) Using process description languages ​​such as BPEL to define and manage the service composition process, through clear process orchestration, different manufacturing services are combined according to predetermined rules to complete complex manufacturing tasks; this type of solution relies on a pre-set process model. (2) Using rule-based automatic planning technology and agent-based technology to realize the automatic discovery and combination of manufacturing services. Through preset rules and intelligent agents, appropriate services are automatically selected for combination according to task requirements, thereby improving the degree of automation of combination. (3) Using various optimization algorithms such as linear weighting, multi-objective optimization and hierarchical optimization to optimize the service composition process. These algorithms weigh different attributes of services and select the optimal service combination scheme. (4) Focusing on service quality QoS attributes such as time, cost, reliability, etc., taking service quality as the core standard of service composition; by evaluating the QoS of different candidate services, select the service that best meets user needs for combination. (5) Using constraints related to sustainability to filter out unfavorable services in the composition process.

[0004] However, the above-mentioned existing technical solutions have the following technical defects: (1) As the scale of cloud manufacturing platforms continues to expand, the number of services is growing exponentially, and traditional service composition methods face serious composition efficiency problems. (2) When optimizing service combinations, existing technologies usually only focus on the average QoS value of services, while ignoring the volatility and uncertainty of service quality. (3) In the context of today's manufacturing industry's increasing attention to environmental protection and resource conservation, if service composition ignores the sustainability requirements of the three aspects of economy, environment and society, it may lead to waste of platform resources and lack of sustainability considerations. (4) Due to the large number of candidate services in cloud manufacturing platforms, existing technologies do not have an effective screening mechanism in the service selection process, resulting in an overly large search space, which increases the complexity and computational cost of the combination algorithm. (5) In service composition optimization, although existing optimization algorithms can find a better solution to a certain extent, it is difficult to take all objectives into account when facing multi-dimensional objectives; and when dealing with large-scale, multi-constraint combination problems, they are prone to fall into local optimality and lack sufficient global search capabilities. In summary, current research mainly focuses on the selection of MSC and service quality QoS for manufacturing service composition, and rarely takes service quality QoS stability and service sustainability into consideration; existing technologies have significant defects in efficiency, reliability and sustainability in the face of large-scale service composition, and an efficient, reliable and sustainable service composition method is urgently needed to overcome the above bottlenecks. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a manufacturing service portfolio optimization analysis method and system for a cloud manufacturing platform, which solves the problems of low efficiency, unsatisfactory QoS stability and sustainability, and large limitations in the optimization process in the current manufacturing service portfolio.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0007] In the first aspect, an embodiment of the present application provides a manufacturing service combination optimization and analysis method for a cloud manufacturing platform, the method comprising: obtaining multiple candidate services with similar functions in a large-scale service environment; classifying the multiple candidate services through a K-means algorithm to form multiple service clusters to reduce the search space; screening the service clusters according to the stability of the service quality QoS corresponding to the candidate services through a preset service pruning mechanism to obtain multiple target services to eliminate services with large volatility; combining the multiple target services to obtain multiple manufacturing service combinations MSC, and constructing a multi-objective optimization model; solving the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize the multiple manufacturing service combinations MSC to obtain multiple Pareto solutions; determining the optimal solution that best meets the requirements from the multiple Pareto solutions according to the preference weights preset by the user to determine the manufacturing service combination MSC corresponding to the optimal solution.

[0008] According to the first aspect of the embodiment of the present application, the quality of service QoS includes time, cost and reliability, and each candidate service corresponds to a QoS attribute vector.

[0009] According to the first aspect of the embodiment of the present application, the aforementioned classification of multiple candidate services through the K-means algorithm to form multiple service clusters to reduce the search space may include the following steps: taking multiple candidate services as original data samples, inputting a parameter k to determine the number of divisions of the original data samples; based on the QoS attribute vectors of the two candidate services, determining the similarity between the two candidate services by calculating the corresponding Euclidean distance; based on the similarity between the two candidate services, calculating the average similarity value of each candidate service in the cluster to determine the cluster center point; according to the cluster center point, clustering the candidate services corresponding to each manufacturing subtask through k-means clustering to obtain k service clusters.

[0010] According to the first aspect of the embodiment of the present application, the calculation process of the similarity between two candidate services satisfies the expression:

[0011]

[0012] In the formula, Indicates candidate services and candidate services The similarity of the QoS attribute vectors between and Respectively and The attribute r value of .

[0013] According to the first aspect of the embodiment of the present application, the aforementioned screening and processing of the service cluster according to the stability of the service quality QoS corresponding to the candidate service through a preset service pruning mechanism to obtain multiple target services to eliminate services with large volatility may include the following steps: calculating the global service quality QoS value corresponding to the candidate service, the global service quality QoS value is used to characterize the correlation between the candidate service and the manufacturing service combination MSC; based on the target number of candidate services in each cluster, determining a global QoS threshold that is positively correlated with the target number; deleting candidate services whose global service quality QoS value is lower than the global QoS threshold from each cluster; calculating the QoS stability of the candidate services, and based on the preset QoS stability threshold, pruning the candidate services that are lower than the QoS stability threshold to obtain multiple target services.

[0014] According to the first aspect of the embodiment of the present application, the calculation process of the global quality of service QoS value satisfies the expression:

[0015]

[0016] in, is the global service quality QoS value, Represents a cluster Normalized QoS attribute value, n i Represents a cluster The number of services in the cluster, I m Indicates subtask ST m The number of candidate services.

[0017] According to the first aspect of the embodiment of the present application, the aforementioned method of solving the multi-objective optimization model based on the fast non-dominated sorting genetic algorithm to optimize multiple manufacturing service combinations MSCs to obtain multiple Pareto solutions may include the following steps:

[0018] S1: Randomly generate the initial population P g , for the initial population P g Perform non-dominated sorting and calculate the crowding degree of all individuals; among them, in the initial population P g In the example, g = 0, the initial population P g The size is N;

[0019] S2: Through binary competition, from the initial population P g Select some individuals from the population for crossover and mutation operations to obtain a descendant population Q with a population size of N g ;

[0020] S3: The initial population P g and the offspring population Q g Merge to form the target population R g , the target population Rg The size of is 2N;

[0021] S4: For the target population R g The individuals in the non-dominated sort are sorted to obtain n non-dominated solution sets F 1 ,F 2 , …, F n ;

[0022] S5: From the target population R g Select individuals in turn until the population size exceeds N, and obtain the target non-dominated solution set a = F i ;

[0023] S6: In F 1 ,F 2 , …, F n When the number of all individuals in exceeds N, calculate F i The crowding degree of all individuals in the game, and according to the preset elite strategy, select F 1 The best individuals and F 1 -F i-1 All individuals in form a new population P g+1 ;

[0024] S7: Let g=g+1, repeat steps S2-S6, and when the preset termination condition is met, terminate the algorithm iteration and obtain multiple Pareto solutions of multi-objective optimization.

[0025] According to the first aspect of the embodiment of the present application, the optimal solution is determined by the comprehensive evaluation value corresponding to each Pareto solution, and the calculation process of the comprehensive evaluation value satisfies the expression:

[0026] ce d =w 1 qos(SC d )+w 2 Sus(SC d )

[0027] Among them, ce d represents the comprehensive evaluation value of the dth Pareto solution; qos(SC d ) and Sus(SC d ) represent the QoS attribute value and sustainability value corresponding to the d-th Pareto solution respectively; w 1 and w 2 They represent the user's preference weights for service quality QoS and sustainability respectively.

[0028] In the second aspect, an embodiment of the present application provides a manufacturing service combination optimization and analysis system for a cloud manufacturing platform, the system comprising: an acquisition module, a classification module, a pruning module, a combination module, a solution module and a determination module; the acquisition module is used to acquire multiple candidate services with similar functions in a large-scale service environment; the classification module is used to classify multiple candidate services through a K-means algorithm to form multiple service clusters to reduce the search space; the pruning module is used to screen and process the service clusters according to the stability of the service quality QoS corresponding to the candidate services through a preset service pruning mechanism, to obtain multiple target services to eliminate services with large volatility; the combination module is used to combine multiple target services to obtain multiple manufacturing service combination MSCs, and to construct a multi-objective optimization model; the solution module is used to solve the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize multiple manufacturing service combination MSCs to obtain multiple Pareto solutions; the determination module is used to determine the optimal solution that best meets the requirements from multiple Pareto solutions according to the preference weights preset by the user, to determine the manufacturing service combination MSC corresponding to the optimal solution.

[0029] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the manufacturing service combination optimization and analysis method for a cloud manufacturing platform in the aforementioned first aspect is implemented.

[0030] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a program or instruction, which, when executed by a processor, implements the manufacturing service combination optimization analysis method for a cloud manufacturing platform in the aforementioned first aspect.

[0031] This application provides a manufacturing service portfolio optimization analysis method and system for a cloud manufacturing platform. Compared with the prior art, it has the following beneficial effects:

[0032] This application uses the K-means clustering algorithm to group and screen a large number of candidate services, reduce the search space, and speed up the service combination. By introducing a pruning mechanism based on QoS stability, unstable or poor quality services are eliminated, the risk of combination failure is reduced, and the reliability of services is improved. This application provides an optimized service combination solution, using a fast non-dominated sorting genetic algorithm for dual-objective optimization, while maximizing service quality QoS and sustainability, providing multiple Pareto optimal solutions, and selecting the optimal manufacturing service combination MSC corresponding to the optimal solution based on user preferences; this application is suitable for cloud manufacturing platforms that require efficient, reliable and sustainable service combinations, especially in large-scale, complex manufacturing tasks. It has significant advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 It is a flow chart of a manufacturing service combination optimization analysis method for a cloud manufacturing platform provided in an embodiment of the present application;

[0035] Figure 2 yes Figure 1 An exemplary flow chart of S120;

[0036] Figure 3 is an exemplary schematic diagram of the optimal configuration of the manufacturing service combination MSC provided in the embodiment of the present application;

[0037] Figure 4-a This is an exemplary schematic diagram of a candidate service cluster for the first subtask provided in an embodiment of the present application;

[0038] Figure 4-b is an exemplary schematic diagram of a candidate service cluster for the second subtask provided in an embodiment of the present application;

[0039] Figure 4-c is an exemplary schematic diagram of a candidate service cluster for the third subtask provided in an embodiment of the present application;

[0040] Figure 4-d is an exemplary schematic diagram of a candidate service cluster for the fourth subtask provided in an embodiment of the present application;

[0041] Figure 4-e is an exemplary schematic diagram of a candidate service cluster for the fifth subtask provided in an embodiment of the present application;

[0042] Figure 5-a is a schematic diagram of a multi-objective optimization Pareto solution with 50 generations provided in an embodiment of the present application;

[0043] Figure 5-b is a schematic diagram of a multi-objective optimization Pareto solution with 100 generations provided in an embodiment of the present application;

[0044] Figure 5-c is a schematic diagram of a multi-objective optimization Pareto solution with 150 generations provided in an embodiment of the present application;

[0045] Figure 5-d is a schematic diagram of a multi-objective optimization Pareto solution with 200 generations provided in an embodiment of the present application;

[0046] Figure 5-e is a schematic diagram of a multi-objective optimization Pareto solution with 250 generations provided in an embodiment of the present application;

[0047] Figure 5-f is a schematic diagram of a multi-objective optimization Pareto solution with 300 generations provided in an embodiment of the present application;

[0048] Figure 6 It is a structural diagram of a manufacturing service combination optimization and analysis system for a cloud manufacturing platform provided in an embodiment of the present application;

[0049] Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0052] The embodiments of the present application solve the problems of low efficiency, unsatisfactory QoS stability and sustainability, and large limitations in the optimization process in the current manufacturing service portfolio by providing a manufacturing service portfolio optimization analysis method and system for a cloud manufacturing platform.

[0053] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0054] The following first introduces a manufacturing service combination optimization analysis method for a cloud manufacturing platform provided in an embodiment of the present application.

[0055] The present application embodiment provides a flowchart of a manufacturing service combination optimization analysis method for a cloud manufacturing platform, such as Figure 1 As shown, the method may include the following steps S110-S160.

[0056] S110: Acquire multiple candidate services with similar functions in a large-scale service environment.

[0057] S120. Classify multiple candidate services using a K-means algorithm to form multiple service clusters to reduce the search space.

[0058] S130 , according to the stability of the quality of service QoS corresponding to the candidate services, the service cluster is screened and processed through a preset service pruning mechanism to obtain multiple target services to eliminate services with greater volatility.

[0059] S140, combining multiple target services to obtain multiple manufacturing service combination MSCs, and constructing a multi-objective optimization model.

[0060] S150. Solve the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize multiple manufacturing service combinations MSC and obtain multiple Pareto solutions.

[0061] S160 . According to the preference weights preset by the user, determine the optimal solution that best meets the requirements from multiple Pareto solutions to determine the manufacturing service combination MSC corresponding to the optimal solution.

[0062] The above is a specific implementation method of a manufacturing service combination optimization analysis method for a cloud manufacturing platform provided in an embodiment of the present application. The present application uses the K-means clustering algorithm to group and screen a large number of candidate services, reduce the search space, and speed up the service combination. By introducing a pruning mechanism based on QoS stability, unstable or poor quality services are eliminated, the risk of combination failure is reduced, and the reliability of the service is improved. The present application provides an optimized service combination solution, which uses a fast non-dominated sorting genetic algorithm for dual-objective optimization. While maximizing service quality QoS and sustainability, it provides a variety of Pareto optimal solutions, and selects the optimal manufacturing service combination MSC corresponding to the optimal solution according to user preferences. The present application is suitable for cloud manufacturing platforms that require efficient, reliable and sustainable service combinations, especially in large-scale and complex manufacturing tasks. It has significant advantages.

[0063] Please note that, please refer to Figure 3,The complex manufacturing tasks in the cloud manufacturing platform can be decomposed into several subtasks according to the execution process; ask ={ST 1 ,…,ST m ,…,ST M}, where ST m is the mth subtask, M is the total number of subtasks; services that meet the functional requirements of specific subtasks are aggregated to form a candidate service set: in, Indicates subtask ST m The i-th candidate service. Each candidate service has non-functional characteristics, and the non-functional attributes of the service are called quality of service QoS; definition is a vector describing the QoS of the service, Indicates candidate services The value of the QoS attribute r, R is the number of QoS attributes. QoS attributes can be divided into positive attributes and negative attributes. For example, time is a positive attribute and reliability is a negative attribute. Different QoS attribute values ​​have different value ranges and optimization directions.

[0064] It should also be noted that QoS performance is instantaneous. If a service is unstable and its QoS value fluctuates greatly, the MSC containing the service is likely to fail. Even if the QoS value of the service is very high; it is wiser to choose a service with more stable QoS. For example, suppose the subtask ST 1 There are two candidate services The platform collects historical QoS values ​​of the service. The QoS data set is {0.97, 0.88, 0.96, 0.88, 0.94}, The QoS data set is {0.92, 0.91, 0.92, 0.94, 0.93}. Although the average QoS values ​​of the two services are illustrate The average performance is the best, but The historical QoS values ​​fluctuate greatly; The QoS value of changes little and is more stable. Based on risk aversion considerations, Rather than To execute subtask ST 1 In order to eliminate the uncertainty caused by services with unstable QoS values ​​in MSC, QoS stability (Sta) needs to be considered when matching services for MSC. The more stable the service QoS is, the more reliable the service is. QoS stability can be measured by the standard deviation of the actual QoS value and the average QoS value.

[0065] It should be emphasized that in the multi-objective optimization model, the subtask ST m From the candidate service set CSS m Select candidate services To perform some routine tasks. Finally, the collection of multiple candidate services is mapped into a combined chain, and each chain corresponds to an MSC. The logical structure of MSC includes four types: sequential, parallel, selective and cyclic. The QoS aggregation of MSC under different structures has different calculation methods. Since parallel, selective and cyclic structures can be simplified or converted into sequential structures, this application focuses on the sequential structure of MSC.

[0066] It should also be emphasized that this application defines x im (i∈I,m∈M) is a binary decision variable, which is used to indicate whether to select a candidate service. To complete the subtask ST m If you select a candidate business Then x im =1; otherwise, x im = 0. The services in the candidate service set are mutually exclusive internally because they compete for subtasks and only one service can be selected to complete the subtask. Therefore, theoretically there are at most service paths satisfy the task T ask The requirements, including I m Indicates subtask ST m The number of candidate services to be searched in the candidate service set. The optimization problem is to select the best MSC under the consideration of multiple objectives and constraints.

[0067] In the MSC process, the service demander, as the task initiator, focuses on the high QoS performance of the composite service at the execution layer. At the same time, the platform operator, as the organizer and manager of resources, focuses on the sustainable development of the cloud manufacturing platform at the operational level. Based on this, this application proposes a multi-objective optimization model that considers QoS (service time t, cost c and reliability re) and sustainability (environmental performance Ep, economic benefit Eb and social impact Si) to allocate MSC. The main idea of ​​MSC allocation in this application is to select the optimal MSC to complete a specific manufacturing task, while maximizing the goals of QoS and sustainability Sus from the perspective of the platform operator and the service demander.

[0068] In some embodiments, the quality of service (QoS) includes time, cost and reliability, and each candidate service corresponds to a QoS attribute vector. Figure 2 As shown, the aforementioned classification of multiple candidate services by the K-means algorithm to form multiple service clusters to reduce the search space, that is, the aforementioned S120 may specifically include the following steps:

[0069] S210: Take multiple candidate services as original data samples and input a parameter k to determine the number of divisions of the original data samples.

[0070] S220 : Based on the QoS attribute vectors of the two candidate services, determine the similarity between the two candidate services by calculating the corresponding Euclidean distance.

[0071] S230 : Based on the similarity between the two candidate services, calculate the average similarity value of each candidate service in the cluster to determine the cluster center.

[0072] S240 , clustering the candidate services corresponding to each manufacturing subtask by k-means clustering according to the cluster center point to obtain k service clusters.

[0073] In the embodiment of the present application, considering the extensiveness of resource services in platforms with the same or similar functions, the present application adopts a clustering method to reduce the search space and improve the efficiency of the MSC. At this stage, the present application uses a k-means algorithm to divide the candidate services of each subtask into k service clusters. It can be understood that for the MSC, the primary task is to ensure the quality of the selected service. Therefore, the similarity between the two services is determined based on the QoS attribute vectors of the two services and calculated by the Euclidean distance.

[0074] In one example, the similarity calculation process between two candidate services satisfies the expression:

[0075]

[0076] In the formula, Indicates candidate services and candidate services The similarity of the QoS attribute vectors between and Respectively and The cluster center is determined by calculating the average similarity value of each service in the cluster.

[0077] In some embodiments, according to the stability of the quality of service QoS corresponding to the candidate services, the service cluster is screened and processed by a preset service pruning mechanism to obtain multiple target services to eliminate services with large volatility, that is, the aforementioned S130 may specifically include the following steps:

[0078] S310: Calculate a global service quality QoS value corresponding to the candidate service, where the global service quality QoS value is used to characterize the correlation between the candidate service and the manufacturing service combination MSC.

[0079] S320: Based on the target number of candidate services in each cluster, determine a global QoS threshold that is positively correlated with the target number.

[0080] S330: Delete candidate services whose global quality of service (QoS) values ​​are lower than the global QoS threshold from each cluster.

[0081] S340, calculating the QoS stability of the candidate services, and based on a preset QoS stability threshold, cutting the candidate services below the QoS stability threshold to obtain multiple target services.

[0082] In the embodiment of the present application, it can be understood that after obtaining k service clusters, some services with low QoS performance and stability are eliminated, and then only candidate services with excellent QoS are focused on, thereby shortening the service combination space. The service pruning mechanism can include two steps, first filtering services with poor global service quality QoS values, and then pruning services with unstable QoS, because services with unstable QoS may reduce the effectiveness of the MSC and cause the risk of interruption to the business workflow. After two steps of streamlining, unreliable candidate services are discarded and the service combination space is reduced.

[0083] In an example, the calculation process of the global quality of service QoS value satisfies the expression:

[0084]

[0085] in, is the global service quality QoS value, Represents a cluster Normalized QoS attribute value, n i Represents a cluster The number of services in the cluster, I m Indicates subtask ST m The number of candidate services.

[0086] In some embodiments, the aforementioned multi-objective optimization model is solved based on the fast non-dominated sorting genetic algorithm to optimize multiple manufacturing service combinations MSC to obtain multiple Pareto solutions, that is, the aforementioned S150 may specifically include the following steps:

[0087] S1: Randomly generate the initial population P g , for the initial population P g Perform non-dominated sorting and calculate the crowding degree of all individuals; among them, in the initial population P g In the example, g = 0, the initial population P g The size is N;

[0088] S2: Through binary competition, from the initial population P gSelect some individuals from the population for crossover and mutation operations to obtain a descendant population Q with a population size of N g ;

[0089] S3: The initial population P g and the offspring population Q g Merge to form the target population R g , the target population R g The size of is 2N;

[0090] S4: For the target population R g The individuals in the non-dominated sort are sorted to obtain n non-dominated solution sets F 1 ,F 2 , …, F n ;

[0091] S5: From the target population R g Select individuals in turn until the population size exceeds N, and obtain the target non-dominated solution set a = F i ;

[0092] S6: In F 1 ,F 2 , …, F n When the number of all individuals in exceeds N, calculate F i The crowding degree of all individuals in the game, and according to the preset elite strategy, select F 1 The best individuals and F 1 -F i-1 All individuals in form a new population P g+1 ;

[0093] S7: Let g=g+1, repeat steps S2-S6, and when the preset termination condition is met, terminate the algorithm iteration and obtain multiple Pareto solutions of multi-objective optimization.

[0094] In the embodiments of the present application, it can be understood that in order to ensure the quality of composite services and the long-term sustainable operation of the cloud service platform, the present application proposes a multi-objective optimization model that takes into account both QoS and sustainability. Compared with single-objective optimization, multi-objective optimization problems are solved due to the mutual constraints of multiple optimization objectives. Increasing the value of certain objective functions may weaken the remaining objective functions, so it is impossible to optimize all objectives at the same time. The solution to multi-objective optimization is usually a set of Pareto solutions, which is characterized by the fact that when improving any one objective function, it will inevitably lead to the degradation of at least another objective function.

[0095] In one example, the fast non-dominated sorting genetic algorithm is the NSGA-II algorithm. Compared with the NSGA algorithm, the NSGA-II algorithm has the following three improvements: (1) a new non-dominated sorting method is proposed to reduce the complexity of the algorithm; (2) an elite strategy is adopted, that is, the parent population and the offspring population compete together to select the best individuals to form the next generation population to ensure the overall quality of the population; (3) a crowding distance comparison operator is used to determine the crowding values ​​of different individuals at the same level, and it is used as a comparison standard for individuals at the same level to ensure the diversity of the population.

[0096] In one example, the optimal solution is determined by the comprehensive evaluation value corresponding to each Pareto solution, and the calculation process of the comprehensive evaluation value satisfies the expression:

[0097] ce d =w 1 qos(SC d )+w 2 Sus(SC d )

[0098] Among them, ce d represents the comprehensive evaluation value of the dth Pareto solution; qos(SC d ) and Sus(SC d ) represent the QoS attribute value and sustainability value corresponding to the d-th Pareto solution respectively; w 1 and w 2 They represent the user's preference weights for service quality QoS and sustainability respectively.

[0099] It is understandable that the Pareto solution is the result of finding the MSC from an objective perspective, while the final decision needs to consider the user's preference information from a subjective perspective. d The values ​​are sorted to select the best MSC solution for the user, and the remaining MSC solutions are used as alternatives.

[0100] In an example, according to the characteristics of the proposed multi-objective optimization model, the chromosome gene encoding, crossover and mutation processes of the NSGA-II algorithm are specifically expressed as follows:

[0101] S410, encoding: create a single chromosome gene corresponding to MSC in the integer array encoding; the gene length is equal to the number of subtasks K, and after pruning in the service clustering stage, the gene value can be within the number of services retained for each task. im Decide whether to select a candidate service Or do not select to complete the subtask ST m . Select the candidate service that completes the task, that is, the binary variable x of the candidate serviceim Equal to 1, its corresponding position is the code of the corresponding chromosome gene.

[0102] S420, crowding calculation: The crowding of the i-th individual is the sum of the differences between all objective function values ​​of the (i+1)-th individual and the (i-1)-th individual.

[0103] S430, crossover operation: use the two-point crossover method to perform the MSC crossover operation; randomly generate two integers in the range [1, M], where M is the number of subtasks; then intercept the chromosome segment between the two positions for crossover.

[0104] S440, mutation operation: According to the mutation probability, two position points are randomly selected on the chromosome code string for exchange.

[0105] In some embodiments, in order to verify the effectiveness of the manufacturing service combination optimization analysis method for cloud manufacturing platform proposed in this application, this application verifies the method through a simplified manufacturing example of customized automatic guided vehicle AGV products. The cloud manufacturing platform decomposes the received order and obtains 5 subtasks, which require 5 abstract services: AGV body service (ST 1 ), drive unit service (ST 2 ), Guide Device Service (ST 3 )、Battery Module Services (ST 4 ), Carrier Services (ST 5 ).

[0106] Through service discovery, multiple services that meet the functional requirements of the subtasks are generated and aggregated to form a candidate service set CSS m ,m=1,…,5. Without losing generality, 5 different candidate service numbers are set for these 5 subtasks, and the subtask numbers and their corresponding candidate service numbers are 1_30, 2_45, 3_60, 4_75 and 5_90 respectively. All experiments are conducted in MATLAB R2017b and run on a 3.0GHz Intel Core i7-9700CPU and 8GB RAM on a 64-bit Windows 10 platform.

[0107] Assume that the service demander is searching for an MSC with the following semantic and QoS attributes: (1) Task type: custom automated guided vehicle (AGV) manufacturing; (2) Manufacturing job: total completion time < 4 months, total task cost <10,000 RMB, reliability of various services> 0.9; (3) Service stability: high.

[0108] The aforementioned S120 and S130 can be specifically performed through the following detailed steps:

[0109] (1) CSS m K-means clustering of candidate services in : Cluster the candidate services of each subtask and filter out unreliable candidate services to reduce the combination space. The number of clusters k is related to the number of original data samples. Each sample data is a candidate service containing multi-dimensional attribute values. Set the number of clusters k to I stands for CSS m The number of candidate services in CSS m The similarity between two services in CSS m All candidate services in can be divided into k clusters; the clustering results of candidate services for the five subtasks are as follows: Figure 4-a to Figure 4-e shown.

[0110] (2) Pruning services of each cluster: Calculate the global QoS value of each candidate service: Set the global QoS threshold qos t = 0.1, remove candidate services with low service quality in each cluster. Then set the QoS stability threshold Sta t =0.92, and services with unreliable QoS are eliminated. After the pruning operation, the final qualified candidate services for each subtask are shown in Table 1.

[0111] Table 1 Results after pruning services

[0112]

[0113] The aforementioned S140, S150 and S160 can be specifically performed through the following detailed steps:

[0114] According to the order of process activities, the qualified candidate services of each subtask are combined to complete the entire manufacturing task. After performing the service clustering operation, the number of qualified candidate services obtained for the five subtasks is 4*5*4*4*6 respectively. Therefore, there are 1920 MSC schemes available for selection. Table 2 shows the relevant parameters of the qualified candidate services for each subtask. The goal of the second stage is to find the optimal MSC with maximum QoS and sustainability from the optional MSC schemes.

[0115] Table 2 Attribute information of qualified candidate services

[0116]

[0117] (3) MSC multi-objective optimization: QoS and sustainability are used as dual objectives for MSC optimization allocation. Three typical QoS attributes, time t, cost c, and reliability re, are used to evaluate the QoS performance of MSC, and the weights of each QoS attribute are set equal. Sustainability is evaluated based on economic, environmental, and social aspects. The NSGA-II algorithm is used to solve the multi-objective optimization problem and obtain the Pareto solution set. The initial population size is set to N = 50, the crossover probability PC = 0.9, and the mutation probability PM = 0.09.

[0118] Figure 5-a to Figure 5-f The Pareto solutions of the multi-objective MSC optimization for specific generations, namely, the 50th, 100th, 150th, 200th, 250th, and 300th generations are shown. The x-axis and y-axis represent the objective values ​​of QoS and sustainability, respectively. The results show that after 250 generations, the algorithm almost converges to the true Pareto optimal set. In other words, there is a major evolutionary process in the initial 250 generations. Since then, the Pareto frontier has been modified only a little in the subsequent generations, which indicates that the other processes in the evolution do not increase significantly. Therefore, in order to ensure the efficiency and effectiveness of the algorithm, the maximum number of generations allowed, G, is set to max Set to 250. The size of the Pareto solution obtained by the NSGA-II algorithm is 14. Each Pareto solution is non-dominant to other solutions. Table 3 lists the attribute values ​​of each Pareto solution.

[0119] Table 3 Attribute information of Pareto solution

[0120]

[0121] (4) Optimal MSC selection. Calculate the comprehensive evaluation value ce of each Pareto solution. The ce value and the ranking of each Pareto solution are shown in Table 4. The optimal MSC solution is This shows that for a binary decision variable x im (i∈I,m∈M), when the decision variable x 25,1 ,x 33,2 ,x 2,3 ,x 54,4 ,x 102,5 When is set to 1 and all other decision variables are set to 0, the optimal solution obtained by the MSC optimization model is better than any other decision. The highest comprehensive evaluation value of the MSC solution is 4.436, which is the optimal solution. 1 Subtask, select Because it has the best sustainable performance.

[0122] Table 4 Ranking of Pareto solutions

[0123]

[0124] It should be noted that the method proposed in this application includes service clustering, multi-objective optimization and optimal MSC selection steps for MSC allocation (method 1). Compared with the method that does not include the service clustering step (method 2), the number of Pareto solutions obtained by the two methods are 14 and 21 respectively. The average comprehensive evaluation value of the Pareto solution obtained by method 2 is 4.316, which is lower than the average comprehensive evaluation value of 4.402 obtained by method 1, indicating that service clustering can effectively improve the performance of the Pareto solution. Comparing the performance of the final optimal MSC obtained by the two methods, it can be inferred that the highest comprehensive evaluation value of the MSC of method 1 is 4.436, which is greater than 4.341 of method 2.

[0125] To further verify the effectiveness of the multi-objective optimization model, it is compared with the single-objective optimization model. The single-objective optimization model only includes two steps: service clustering and optimal MSC selection. That is, after service clustering, an optimal MSC is directly selected by weighting the QoS attributes and economic benefit-oriented sustainability. Compared with the performance of the single-objective optimization model (model 2), the optimal comprehensive evaluation value of the MSC obtained by the proposed model 1 is 4.436, which is greater than 4.177 of model 2, indicating that the MSC selected by model 1 has better performance. The qos value (3.282) obtained by model 1 is similar to that of model 2, but the sustainability (4.754) is higher than that of model 2, which also meets the requirements of long-term operation of the industrial platform.

[0126] The method of this application integrates service clustering and multi-objective optimization, takes into account QoS stability and sustainability, and achieves reliable and sustainable MSC allocation. In traditional multi-objective optimization methods, only economic benefit-oriented QoS and sustainability are considered as the two goals of MSC allocation. The latter leads to poor overall sustainability performance of the integrated service despite high economic benefits. The average QoS stability and the highest QoS stability of the MSC schemes in the Pareto solution set obtained by the two methods are further analyzed. In all cases, the average QoS stability obtained by this method is significantly higher than that of the traditional method.

[0127] It can be seen from these results that the manufacturing service portfolio optimization analysis method for cloud manufacturing platform proposed in this application is the most effective in both high-performance service selection and high-stability MSC allocation, which can both avoid risks in advance and constrain service behaviors in MSC.

[0128] In some embodiments, the present application provides a manufacturing service combination optimization analysis system 500 for a cloud manufacturing platform, please refer to Figure 6 The system 500 may include the following modules:

[0129] An acquisition module 510 is used to acquire multiple candidate services with similar functions in a large-scale service environment;

[0130] A classification module 520, configured to classify multiple candidate services by using a K-means algorithm to form multiple service clusters to reduce the search space;

[0131] The pruning module 530 is used to screen the service cluster according to the stability of the quality of service QoS corresponding to the candidate services through a preset service pruning mechanism to obtain multiple target services to eliminate services with greater volatility;

[0132] A combination module 540 is used to combine multiple target services to obtain multiple manufacturing service combination MSCs and construct a multi-objective optimization model;

[0133] A solution module 550 is used to solve the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize multiple manufacturing service combinations MSC to obtain multiple Pareto solutions;

[0134] The determination module 560 is used to determine the optimal solution that best meets the requirements from multiple Pareto solutions according to the preference weights preset by the user, so as to determine the manufacturing service combination MSC corresponding to the optimal solution.

[0135] According to an embodiment of the present application, any multiple modules among the acquisition module 510, the classification module 520, the pruning module 530, the combination module 540, the solution module 550 and the determination module 560 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module.

[0136] In some embodiments, the classification module 520 may be specifically used to:

[0137] Taking multiple candidate services as original data samples, inputting a parameter k to determine the number of divisions of the original data samples;

[0138] Based on the QoS attribute vectors of the two candidate services, the similarity between the two candidate services is determined by calculating the corresponding Euclidean distance;

[0139] Based on the similarity between two candidate services, the average similarity value of each candidate service in the cluster is calculated to determine the cluster center point;

[0140] According to the cluster center point, the candidate services corresponding to each manufacturing subtask are clustered through k-means clustering to obtain k service clusters.

[0141] In some embodiments, the pruning module 530 may be specifically configured to:

[0142] Calculate the global service quality QoS value corresponding to the candidate service, and the global service quality QoS value is used to characterize the correlation between the candidate service and the manufacturing service combination MSC;

[0143] Based on the target number of candidate services in each cluster, a global QoS threshold positively correlated with the target number is determined;

[0144] Remove candidate services from each cluster whose global service quality QoS value is lower than the global QoS threshold;

[0145] The QoS stability of the candidate services is calculated, and based on a preset QoS stability threshold, the candidate services below the QoS stability threshold are trimmed to obtain multiple target services.

[0146] In some embodiments, the solution module 550 may be specifically used to:

[0147] S1: Randomly generate the initial population P g , for the initial population P g Perform non-dominated sorting and calculate the crowding degree of all individuals; among them, in the initial population P g In the example, g = 0, the initial population P g The size is N;

[0148] S2: Through binary competition, from the initial population P g Select some individuals from the population for crossover and mutation operations to obtain a descendant population Q with a population size of N g ;

[0149] S3: The initial population P g and the offspring population Q g Merge to form the target population P g , the target population R g The size of is 2N;

[0150] S4: For the target population R g The individuals in the non-dominated sort are sorted to obtain n non-dominated solution sets F 1 ,F 2 , …, F n ;

[0151] S5: From the target population R g Select individuals in turn until the population size exceeds N, and obtain the target non-dominated solution set a = F i ;

[0152] S6: In F 1 ,F 2 , …, F nWhen the number of all individuals in exceeds N, calculate F i The crowding degree of all individuals in the game, and according to the preset elite strategy, select F 1 The best individuals and F 1 -F i-1 All individuals in form a new population P g+1 ;

[0153] S7: Let g=g+1, repeat steps S2-S6, and when the preset termination condition is met, terminate the algorithm iteration and obtain multiple Pareto solutions of multi-objective optimization.

[0154] Figure 6 Each module in the system shown has the function of implementing each step in the aforementioned manufacturing service combination optimization analysis method for the cloud manufacturing platform and can achieve its corresponding technical effect. For the sake of concise description, it will not be repeated here.

[0155] In some embodiments, the present application provides an electronic device, the structure diagram of the electronic device is as follows Figure 7 shown.

[0156] The electronic device may include a processor 610 and a memory 620 storing computer program instructions.

[0157] Specifically, the processor 610 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0158] The memory 620 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 620 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 620 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 620 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 620 is a non-volatile solid-state memory.

[0159] The memory 620 may include a read-only memory (ROM), a random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Therefore, generally, the memory 620 includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it can perform the operations described in any one of the manufacturing service portfolio optimization analysis methods for a cloud manufacturing platform in the above-mentioned embodiments.

[0160] The processor 610 reads and executes the computer program instructions stored in the memory 620 to implement any one of the manufacturing service combination optimization analysis methods for the cloud manufacturing platform in the above embodiments.

[0161] In one example, the electronic device may further include a communication interface 630 and a bus 600. Figure 7 As shown, the processor 610, the memory 620, and the communication interface 630 are connected via a bus 600 and communicate with each other.

[0162] The communication interface 630 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0163] Bus 600 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 600 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0164] In addition, in combination with the manufacturing service combination optimization analysis method for cloud manufacturing platform in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any one of the manufacturing service combination optimization analysis methods for cloud manufacturing platform in the above embodiment is implemented.

[0165] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0166] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0167] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0168] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0169] In summary, compared with the prior art, this application has the following beneficial effects:

[0170] 1. This application uses the K-means clustering algorithm to group and screen a large number of candidate services, reduce the search space, and speed up the service combination. By introducing a pruning mechanism based on QoS stability, unstable or poor quality services are eliminated, the risk of combination failure is reduced, and the reliability of services is improved.

[0171] 2. This application provides an optimized service combination solution, adopts a fast non-dominated sorting genetic algorithm for dual-objective optimization, provides multiple Pareto optimal solutions while maximizing service quality QoS and sustainability, and selects the optimal manufacturing service combination MSC corresponding to the optimal solution based on user preferences; This application is suitable for cloud manufacturing platforms that require efficient, reliable and sustainable service combinations, especially in large-scale and complex manufacturing tasks. It has significant advantages.

[0172] 3. This application takes into account the sustainability of services: In the process of service combination, economic, environmental and social sustainability are considered at the same time to ensure green manufacturing and long-term operation of the cloud manufacturing platform.

[0173] 4. This application has achieved a good balance in maximizing service quality QoS and sustainability. Compared with the single-objective optimization method, after comprehensively considering multiple objectives, the generated service portfolio not only meets the user's high requirements for time, cost and reliability, but also has higher environmental friendliness.

Claims

1. A manufacturing service combination optimization analysis method for a cloud manufacturing platform, characterized in that: include: Obtain multiple candidate services with similar functions in a large-scale service environment; Classifying the multiple candidate services by using a K-means algorithm to form multiple service clusters to reduce the search space; According to the stability of the quality of service QoS corresponding to the candidate service, the service cluster is screened and processed through a preset service pruning mechanism to obtain multiple target services to eliminate services with greater volatility; Combining the multiple target services to obtain multiple manufacturing service combinations MSCs, and constructing a multi-objective optimization model; Solving the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize the multiple manufacturing service combinations MSCs and obtain multiple Pareto solutions; According to the preference weights preset by the user, an optimal solution that best meets the requirements is determined from the multiple Pareto solutions, so as to determine the manufacturing service combination MSC corresponding to the optimal solution.

2. The manufacturing service combination optimization analysis method for a cloud manufacturing platform according to claim 1, characterized in that: The quality of service QoS includes time, cost and reliability, and each candidate service corresponds to a QoS attribute vector; The classifying the multiple candidate services by using the K-means algorithm to form multiple service clusters to reduce the search space includes: Taking the multiple candidate services as original data samples, inputting a parameter k to determine the number of divisions of the original data samples; Determining the similarity between the two candidate services by calculating the corresponding Euclidean distances based on the QoS attribute vectors of the two candidate services; Based on the similarity between the two candidate services, calculating the average similarity value of each candidate service in the cluster to determine the cluster center point; According to the cluster center point, the candidate services corresponding to each manufacturing subtask are clustered by k-means clustering to obtain k service clusters.

3. The manufacturing service combination optimization analysis method for a cloud manufacturing platform according to claim 2, characterized in that: The calculation process of the similarity between the two candidate services satisfies the expression: In the formula, Indicates candidate services and candidate services The similarity of the QoS attribute vectors between and Respectively and The attribute r value of .

4. The manufacturing service combination optimization analysis method for a cloud manufacturing platform according to any one of claims 1 to 3, characterized in that: The method of screening the service cluster according to the stability of the quality of service QoS corresponding to the candidate service through a preset service pruning mechanism to obtain multiple target services to eliminate services with greater volatility includes: Calculating a global quality of service QoS value corresponding to the candidate service, where the global quality of service QoS value is used to characterize the correlation between the candidate service and the manufacturing service combination MSC; Based on a target number of candidate services in each cluster, determining a global QoS threshold that is positively correlated with the target number; Deleting candidate services whose global quality of service (QoS) value is lower than the global QoS threshold from each cluster; The QoS stability of the candidate services is calculated, and based on a preset QoS stability threshold, the candidate services below the QoS stability threshold are trimmed to obtain a plurality of target services.

5. The manufacturing service combination optimization analysis method for cloud manufacturing platform according to claim 4, characterized in that: The calculation process of the global service quality QoS value satisfies the expression: in, is the global quality of service QoS value, Represents a cluster Normalized QoS attribute value, n i Represents a cluster The number of services in the cluster, I m Indicates subtask ST m The number of candidate services.

6. The manufacturing service combination optimization analysis method for a cloud manufacturing platform according to claim 1, characterized in that: The multi-objective optimization model is solved based on the fast non-dominated sorting genetic algorithm to optimize the multiple manufacturing service combinations MSC to obtain multiple Pareto solutions, including: S1: Randomly generate the initial population P g , for the initial population P g Perform non-dominated sorting and calculate the crowding degree of all individuals; where, in the initial population P g In the example, g = 0, the initial population P g The size is N; S2: Through binary competition, from the initial population P g Select some individuals from the population for crossover and mutation operations to obtain a descendant population Q with a population size of N g ; S3: The initial population P g and the offspring population Q g Merge to form the target population R g , the target population R g The size of is 2N; S4: for the target population R g The individuals in the non-dominated sort are sorted to obtain n non-dominated solution sets f1, F2, ..., F n ; S5: From the target population R g Select individuals in turn until the population size exceeds N, and obtain the target non-dominated solution set a = F i ; S6: In F1, F2, ..., F n When the number of all individuals in exceeds N, calculate F i The crowding degree of all individuals in F1 and the best individuals in F1-F are selected according to the preset elite strategy. i-1 All individuals in form a new population P g+1 ; S7: Let g=g+1, repeat steps S2-S6, and when the preset termination condition is met, terminate the algorithm iteration and obtain multiple Pareto solutions of multi-objective optimization.

7. The manufacturing service combination optimization analysis method for a cloud manufacturing platform according to claim 1 or 6, characterized in that: The optimal solution is determined by the comprehensive evaluation value corresponding to each of the Pareto solutions, and the calculation process of the comprehensive evaluation value satisfies the expression: What d =w1qos(SC d )+w2Up(SC d ) Among them, ce d represents the comprehensive evaluation value of the dth Pareto solution; qos(SC d ) and Sus(SC d ) represent the QoS attribute value and sustainability value corresponding to the d-th Pareto solution respectively; w1 and w2 represent the user's preference weights for QoS and sustainability respectively.

8. A manufacturing service combination optimization analysis system for a cloud manufacturing platform, characterized in that: include: An acquisition module is used to acquire multiple candidate services with similar functions in a large-scale service environment; A classification module, used for classifying the multiple candidate services by using a K-means algorithm to form multiple service clusters to reduce the search space; A pruning module, configured to screen the service cluster according to the stability of the quality of service QoS corresponding to the candidate service through a preset service pruning mechanism, to obtain multiple target services to eliminate services with greater volatility; A combination module, used for combining the multiple target services to obtain multiple manufacturing service combination MSCs and constructing a multi-objective optimization model; A solution module, used for solving the multi-objective optimization model based on a fast non-dominated sorting genetic algorithm to optimize the multiple manufacturing service combinations MSC to obtain multiple Pareto solutions; The determination module is used to determine the optimal solution that best meets the requirements from the multiple Pareto solutions according to the preference weights preset by the user, so as to determine the manufacturing service combination MSC corresponding to the optimal solution.

9. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the manufacturing service combination optimization analysis method for a cloud manufacturing platform as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores programs or instructions, and when the programs or instructions are executed by the processor, the manufacturing service combination optimization analysis method for a cloud manufacturing platform as described in any one of claims 1 to 7 is implemented.

Images