Research On Scheduling Optimization Of Workshop Assembly Line Based On Digital Twin

With the in-depth integration and development of new generation information technologies such as cloud computing,Internet of Things,big data,artificial intelligence,and virtual reality with traditional manufacturing,countries around the world have respectively proposed national-level manufacturing transformation strategies.How to improve the intelligence of workshop production,achieve timely delivery,and improve customer satisfaction has become a key issue in enhancing the competitiveness of small and medium-sized manufacturing enterprises.The hybrid flow shop combines the traditional Flow shop scheduling problem and the multi-machine parallel allocation problem.It has been proved to be a typical NP-hard problem,which is widely used in automobile production,metallurgy,chemical industry and other fields.However,as product demand tends to be individualized and manufacturing processes become more diverse,manufacturing companies have put forward higher requirements in terms of autonomy,intelligence,and predictability for solutions to the mixed flow shop scheduling problem.Realizing the interconnection and intelligent operation of the physical space and the information space of manufacturing is the key to realizing intelligent manufacturing,as well as the key to intelligent scheduling.It is an effective way to improve the autonomy,intelligence,and predictability of scheduling.However,the existing scheduling mode is difficult to realize the real-time interaction between the information space and the physical space,and the data of the information space and the physical space lack integration.In order to solve the problems existing in the dynamic scheduling problem of the mixed flow shop,this paper combines the digital twin and takes a truck production shop as the background to study the scheduling optimization problem of the mixed flow shop.The research mainly includes the following parts:(1)A new model of workshop scheduling based on digital twin is proposed.In view of the lack of physical information space fusion and the difficulty of dealing with workshop disturbances in time in the existing scheduling mode,combined with digital twins,a workshop architecture system driven by digital twin is proposed,and the key technologies under this framework,such as digital twin modeling and intelligent real-time collection of scheduling data,are described.(2)An off-line model based on truck hybrid flow shop is established.Aiming at the production characteristics of the truck production workshop,an off-line scheduling model of the hybrid flow shop that meets the truck production characteristics is built.The goal is to minimize the maximum completion period,minimize the maximum number of products,minimize the maximum delay or penalize costs in advance,and adopt improvements The genetic algorithm solves multi-objectives,obtains the truck production sequence of the offline scheduling model.(3)An online model based on truck hybrid flow shop is established.Combining the production data of the truck hybrid production workshop obtained by the off-line model,build the online scheduling model of the truck hybrid production workshop,using Support Vector Machines(SVM)as the data mining tool,and using the binary particle swarm optimization algorithm(BPSO)to analyze the production attributes(feature Set)to optimize,obtain the dynamic scheduling strategy classification model based on SVM.(4)Design and implementation of visual simulation and information management system for truck hybrid flow shop.Through the rapid modeling of the truck production workshop in 2D and 3D environments,the production tasks of the truck production workshop are optimized for production scheduling,and the corresponding production scheduling results are imported into the actual production system to obtain relevant index data and build the dispatching information management system of the truck mixed flow shop is introduced to facilitate the operation of enterprises.