Research On Modeling And Scheduling Method Of Intelligent Manufacturing System Based On Petri Net

Along with the rising of the customization of manufacturing gradually, intelligent manufacturing system has been put into practicing widely. Intelligent scheduling is the foundation of intelligent manufacturing in the Industry 4.0 which is made up of smart plant and intelligent logistics that is put forward. Most of the intelligent scheduling problem is combinatorial optimization problem which has been proved to be NP-hard problem on the calculation theory. The method of traditional studying combinatory can’t obtain the optimal result in polynomial time, people begin to study obtaining the near optimal solution to meet the needs of production scheduling in polynomial time. The paper studies three aspects as follows.Firstly, in the section of establishing the model of intelligent manufacturing system, the main content of this paper is how to use Petri Net to analysis the model of the intelligent system. Furthermore, we put forward to the idea of establishing the model of the entire manufacturing system from basic manufacturing to unit. We estimate the deadlock situation of the system through analysis of reachability graph of Petri Net.The next thing we have done is that we study on batch scheduling. The paper presents a hybrid algorithm combined with the annealing algorithm and genetic algorithm to apply to the field of intelligent scheduling. We design the hybrid algorithm for the reason that it not only makes use of the global search ability of simulated annealing algorithm, but also takes into the properties of fast convergence of genetic algorithms account. Through simulation comparison experiment with genetic algorithm, the hybrid algorithm shows that it can draw up the better scheduling results.Lastly, we divide the scheduling into determining pay period and fuzzy delivery according to the type of task in research on the scheduling, which makes the research more targeted. The fuzzy delivery problem is made into a real scheduling evaluation by establishing the model of itself, which makes the algorithm well-directed.Through the research of three aspects above, this paper presents different scheduling methods that are adopted respectively according to the size of scheduling. Scheduling is performed through the establishment of the Petri net model for single or multiply scheduling problem. Otherwise, using intelligent scheduling algorithm in production scheduling is a better method for the batch scheduling problem. There are two innovations in this paper as follows:The first innovation is that the paper makes the Petri Net model apply to intelligent scheduling, and the Petri Net model is established by solving the scheduling problem of single and multiple tasks.The simulated annealing algorithm and genetic algorithm are combined to apply to the field of intelligent scheduling, which proves that the hybrid algorithm has huge advantages by the experiments with genetic algorithm.