Study On Edge-Cloud Collaborative Production Mode And Scheduling On The Internet Of Things

With the depth development and widespread application of edge intelligence technology based on the Internet of Things,the edge-side distributed intelligence features are more obvious,and the edge-side data increased dramaticlly.Therefore,cloud-side can not deal with the storage and processing for large amounts of real-time data.So the cloudside services therefore needs to be sunk and migrated to the edge,which has led to edgecloud collaboration and related research.At the same time,with intemsifying global economics and market competition,the management characteristics of enterprises with multiple manufacturing plants served for headquarters under super-city groups have become increasingly obvious.The problem of order dynamic fluctuations caused by personalized customization requirements has become more prominent,which makes it impossible to do global long-period prediction or real-time short-period response relied solely on the cloud or edge.Scheduling as an important mean of production operation optimization has an important role in dealing with the optimization in complex manufacturing environments.Therefore,based on the research background and problems,this thesis focuses on a edge-cloud collaborative production paredigm and scheduling issues of the Internet of Things,including:(1)The thesis analyzes and refines the characteristics and scheduling problems of the production system under the edge-cloud-collaborative paredigm,and builds a scheduling framework on this basis.Then the thesis defines the operation mechanism of the edge-cloud-collaborative production scheduling.(2)Based on edge-side value-added data,factory manufacturing capacity and historical order data,the thesis establishes an order completion time periodic forecast model of cloud headquarters,and uses BP neural network algorithm to solve it.After split the order according to product type,the manufacturing tasks are evaluated and ranked using the prediction results and weighted average method based on multiple manufacturing tasks.Then considering the remaining manufacturing capacity of the edge factory,with the goal of minimizing processing costs,the manufacturing tasks are assigned to edge factories by genetic algorithms.(3)According to the priority of the tasks assigned by the cloud-side in the headquarters and the usage of the edge-side factory resources,with the scheduling optimization goals and constraints,the genetic algorithms are used to dynamically match the tasks of edge-side factory and resources.Then upload the value-added data required by the cloud after the completed.(4)The thesis takes a group enterprise under the Shanghai super city group in China as the research object.The intelligent algorithm and discrete event simulation are used to analyze and verify the edge-cloud-collaborative scheduling paredigm,and to compare and analyze the prediction errors of two methods.Through the above research,we prove the rationality of the scheduling mechanism and the effectiveness of the scheduling model.The proposed method can provide a certain reference for task scheduling in the edge-cloud collaborative production paradigm.