| As an emerging service-oriented manufacturing model,cloud manufacturing aims to achieve efficient sharing and collaborative manufacturing of manufacturing resources among enterprises through key technologies such as service matching and service optimization selection,and provide users with access to,on-demand configurable and high quality cheap manufacturing services.This article considers the heterogeneity of manufacturing resources,the complexity and diversity of service requirements in the context of cloud manufacturing,and studies the service-based packaging of manufacturing resources and the optimal configuration of manufacturing services in the cloud manufacturing platform.The main research contents are as follows:(1)Aiming at the diverse,heterogeneous and complex characteristics of manufacturing resources in the context of cloud manufacturing,a cloud manufacturing service packaging method based on extended OWL-S is proposed.According to the characteristics and model requirements of manufacturing resources,a formal description model of physical manufacturing resources is constructed.Abstractly describe manufacturing resources as manufacturing capabilities.Then,the extended OWL-S manufacturing service description language is used and combined with ontology modeling technology to carry out semantic description and service encapsulation,and finally achieve the purpose of eliminating semantic heterogeneity and shielding the underlying attribute details of manufacturing resources.Finally,taking a vertical CNC drilling machine as an example,the service packaging method of manufacturing resources is described.(2)In view of the characteristics of large number of manufacturing services,large scale of service search,and short service description texts under the background of cloud manufacturing,and the existing service discovery methods only consider the semantics of the service document itself,without considering the latent semantics,a topic model based on Biterm is adopted.Manufacturing service discovery method,using BTM to mine latent semantic information of service documents,combined with numerical interval matching method to determine the candidate service set with the highest similarity with user service requests.(3)In view of the multi-task concurrent manufacturing service composition requirements in the context of cloud manufacturing,a multi-task overall optimization strategy is adopted to build a service composition model that uses limited service resources to achieve balanced distribution among multiple tasks,and designs a service composition optimization model based on Improve the solution algorithm of artificial bee colony algorithm.The algorithm has the good exploration ability of the basic artificial bee colony algorithm.At the same time,the threshold acceptance strategy and the chaotic optimization strategy are introduced to improve the ability to traverse the solution space and avoid the premature convergence problem.In addition,the improved roulette strategy can improve the convergence speed of the algorithm and avoid falling into local optimum.Numerical simulation experiments show that,compared with the traditional first-come-first-served single-task optimization strategy,the multi-task overall optimization strategy can better balance the configuration of manufacturing services,and the improved artificial bee colony algorithm also has better convergence performance and search results.excellent performance.(4)Based on the above research content,a cloud manufacturing platform application prototype system is built,and the method proposed in this paper is verified. |
