Research On Dynamic Flexible Workshop Scheduling Method For Intelligent Manufacturing

With the rapid development of 5G and other information and communication technologies and artificial intelligence technologies,the Internet of Things characterized by "all things perceive,interconnected,and smart" has already achieved a further realization in the production workshop.Intelligent manufacturing has become the development of modern manufacturing important direction.Workshop production scheduling optimization is an important and closely related topic.Studying the scheduling problem of intelligent manufacturing is helpful to promote the intelligent production of the manufacturing shop,quickly respond to the diversified needs of customers,and enhance the core competitiveness of the market for manufacturing enterprises.This thesis focuses on the production scheduling of flexible workshops,and the main research content and results are as follows:Firstly,designing the overall plan of the scheduling system for intelligent manufacturing.Based on the summary of the meaning and characteristics of intelligent manufacturing-oriented workshop scheduling,this thesis analyzes the needs of the intelligent workshop for the Io T environment,and builds the overall architecture of the5 G Io T-based scheduling system and 5G Io T network,laying a theoretical foundation for the workshop to realize interconnection perception and data acquisition.A multi-agent scheduling model with a federated structure is established,and aiming at the problem that the model is unstable during frequent interference during the scheduling process,a concept of "set of scheduled operations" is proposed,and the corresponding dynamic scheduling method is described based on this concept.Secondly,aiming at the characteristics of "multi-objective",proposes a comparison strategy between non-dominated solutions based on the actual needs of scheduling.This thsis briefly summarizes the Pareto-based fitness allocation strategy and its improvement methods.It is proved that this comparison strategy can theoretically enable the algorithm to obtain the optimal non-dominated solution,reasonably reduce the target space dimension,and use the weighted method to incorporate the decision maker’s preference information,so that the final solution can meet the actual needs.Then,aiming at two different types of flexible shop scheduling problems,the corresponding scheduling algorithm is studied based on the standard CS algorithm.First,a multi-objective mathematical model of FFSP and FJSP is established.Then,for FFSP,a permutation-based coding method is improved and a random rule decoding method is improved to improve the quality of the solution.According to the characteristics of FFSP encoding,the discrete lévy flight based on position crossover and individual distance is proposed,and the brood parasitic strategie based on optimal insertion and optimal exchange is designed.For FJSP,an operation-based coding method is adopted,and a decoding method based on the minimum completion time is designed.According to the characteristics of FJSP encoding,the discrete lévy flight based on random disruption operation is proposed,and the brood parasitic strategie based on insertion and disturbance operation is designed.Finally,based on the standard cuckoo search algorithm.Finally,based on the CS algorithm structure,two types of workshops scheduling algorithm processes are designed.Finally,designing scheduling simulation experiments for two types of workshops.Orthogonal experiments are designed to reasonably determine the parameter values of the two algorithms;algorithm performance experiments are designed to solve ten examples of FFSP and FJSP respectively,and compared with other algorithms to verify the effectiveness of the algorithms;finally The dynamic scheduling method is used to simulate the production cases of two types of workshops in three scenarios,and compared with the traditional multi-agent method,the effectiveness of the method in this thesis is verified.