Modeling And Analysis Of Cyber-Physical System Based On Petri Net

With the cooperation and implementation of the "made in China 2025" strategy and Industry 4.0,manufacturing enterprises around the world have begun to expedite the pace of intelligent upgrading of manufacturing,and the core of which is Cyber-physical System(CPS).As an advanced embedded system with complex structures and functions,CPS modeling and analysis is becoming a hot research subject.However,in the existing models,only few can comprehensively model the logical structure,process,and interactions between components of CPS,and most of them cannot simultaneously meet the demand for qualitative,quantitative and energy consumption analysis.Besides,they rarely consider the uncertainties that are common in CPS,which leads to the inaccurate analysis results.To solve these problems,this paper extends the traditional Petri net to Object-oriented Generalized Stochastic Petri Net(OGSPN)and Object-oriented Energy Consumption Petri Net(OECPN),then proposes the fuzzy analysis methods.The innovations and main research work of this paper are as follows:1.Due to the limitations of analysis methods and the insufficient description ability of the existing formal CPS models,this paper combines the object-oriented methodology and Petri net.Then,the time and energy consumption attributes are attached to transitions to form OGSPN and OECPN,respectively.On this basis,the software Platform Independent Petri Net Editor(PIPE)is used to simulate them to verify their correctness,also the qualitative analysis of the modeled CPS is performed.Besides,the OGSPN Compression Algorithm(OCA)is proposed to transform the OGSPN into Generalized Stochastic Petri Net(GSPN),making it able to do enough quantitative analysis.2.In order to cope with the uncertainties in CPS,the triangular fuzzy number and its ?-cut are added in GSPN and OECPN.On this basis,for the quantitative CPS performance indicators,fuzzy analysis method based on the Reduced Embedded Markov Chain(REMC)which is isomorphic to the GSPN is proposed.And for the energy consumption of CPS,the CPS Component Energy Consumption Analysis Algorithm(CECAA)based on the reachability graph of OECPN is proposed.Moreover,the defuzzification methods of the two types of analysis are discussed.3.This paper applies the proposed methods to a real-world CPS case,i.e.,the thick metal plate production line of XIOLIFT(www.xiolift.com).Through the modeling and analysis,it is found that the difference between analysis results and the production data of XIOLIFT is within a reasonable range,which further validates the effectiveness of the methods proposed in this paper.On this basis,extensive experiments are performed by simulating changes in the system condition,then the optimization and energy management direction of the modeled CPS is discussed using the experimental results.The CPS modeling and analysis is one of the key challenges in the design of complex systems,and it is also an effective way in the intelligent upgrading of manufacturing and other companies.The research in this paper provides a basis for the initial design and long-term operation,as well as the structural adjustment and process optimization of CPS,and has theoretical and practical significance for the development of CPS.