Research And Application Of Adaptive Combination Optimization Method For Manufacturing Services In Cloud Manufacturing Environment

Cloud manufacturing is a new manufacturing mode that deeply integrates cuttingedge information technology and manufacturing technology.Combined optimization and on-demand configuration of manufacturing services under cloud manufacturing environment is the core business in the development process of cloud manufacturing mode.Manufacturing services are the core resources in the cloud manufacturing model.To improve the reliability and stability of cloud manufacturing service portfolio optimization is of great significance to support the landing application of the cloud manufacturing model.Since cloud manufacturing tasks and services are in a dynamic production environment with frequent disturbances for a long time,the dynamic adaptability of manufacturing service combination optimization method needs to be strengthened urgently.Therefore,based on the existing research results at home and abroad,this paper focuses on the adaptive combination method of manufacturing services in cloud manufacturing environment.Firstly,service mode and attribute information composition of manufacturing resources under cloud manufacturing environment are analyzed,and a cognitive model of manufacturing resources and a service-oriented encapsulation method are studied.On this basis,random disturbance factors in cloud manufacturing environment are analyzed,and an adaptive combinatorial optimization strategy of manufacturing services is proposed,which includes initial planning stage and execution stage.Secondly,from the perspective of initial planning stage of cloud manufacturing task,considering the factors of dynamic change of historical evaluation index with time,a static combination optimization method of manufacturing service based on multiobjective evolutionary algorithm was studied to realize initial planning of manufacturing task flow.From the perspective of cloud manufacturing task execution stage,considering the factors of service demand change and resource state change,a dynamic optimization method of cloud manufacturing service based on Reinforcement Learning(RL)is studied,and the adaptive optimization decision of cloud manufacturing service in dynamic production environment is realized.Finally,based on the above research results,combined with the cloud manufacturing service platform previously developed by the research group,the optimization combination and dynamic decision module of manufacturing service under cloud manufacturing environment is developed to verify the application of the above research contents.