Study On Optimal Scheduling Of Service-oriented Cyber-physical Production System

With the rapid development of technology and the constant changes in market demand,the manufacturing industry is facing profound transformation that the new round of industrial revolution represented by Germany's "Industry 4.0" is in full swing.In this reform,the deep integration of the new generation of information and communication technology represented by Cyber-Physical System(CPS),Internet of Things(Io T)and Artificial Intelligence(AI)and the new advanced manufacturing technology represented by 3D printing with production system to gradually promote the development of production system towards the direction of intellectualization and servitization,so as to realize intelligent manufacturing.Service-oriented Cyber-Physical Production System(So-CPPS),a new type of intelligent manufacturing model,has the characteristics of distribution and reconfiguration.Compared with the traditional production system centralized optimization scheduling,because the scheduling model of So-CPPS has obvious decentralized enhanced features,it can better respond to the flexible production needs of the personalized and customized market.In view of this,this thesis takes SoCPPS system as the research object and studies its distributed optimization scheduling problem.Firstly,based on the analysis of the characteristics of the So-CPPS system,the SoCPPS scheduling framework is constructed using multi-agent modeling methods,and the corresponding So-CPPS scheduling strategy is formulated;and according to CPS,service-oriented technology and Agent technology,Production Task Agent model and Equipment Resource Agent model are designed in the system.Secondly,the production tasks and machine resources are servitized in this thesis,and the production task ontology model and machine resource ontology model are respectively constructed.Then,the matching rules of production tasks and machine resources are designed based on the ontology semantic model,and the appropriate set of candidate machine resources is determined for the production task.Thirdly,the pre-scheduling algorithm of So-CPPS is designed,and the genetic algorithm is improved by using biological immunity mechanism.The execution process of the algorithm,such as coding,crossover,mutation,immunity and selection,is described in detail,and the effectiveness of the algorithm is tested with benchmark..Then,a real-time optimal scheduling algorithm for So-CPPS is designed.Based on the analysis of the traditional contract network negotiation mechanism(CNP),it is improved with the help of game theory(GT),which is described in detail from the aspects of negotiation protocol,negotiation process and negotiation strategy.Finally,the AnyLogic simulation platform is used to build a So-CPPS running simulation model based on the multi-agent model method,the improved genetic algorithm and game theory based negotiation mechanism were implemented,and the simulation experiments verify the feasibility and effectiveness of the above theory and method.