Automated Discovery Method For Typical Process Knowledge Of Discrete Manufacturing Systems

With the increasing application of intelligent manufacturing,enterprises must apply intelligent manufacturing technology to traditional manufacturing industries in order to continue to develop in an increasingly competitive environment.The process route in discrete manufacturing systems is characterized by flexibility and diversity,making the extraction and pushing of typical process routes extremely difficult.This paper focuses on the extraction and push of the typical process route of discrete manufacturing systems,and studies the clustering and pushing problems of one-way process routes and mesh process routes in discrete manufacturing systems.The clustering analysis and pushing of process routes in discrete manufacturing system are carried out by using algorithm clustering and knowledge-based pushing methods.Finally,the knowledge pushing software is constructed and applied in the field.The application has achieved good results.Specific research work is as follows:(1)Aiming at the typical process discovery and knowledge reuse in discrete manufacturing systems,a typical process knowledge discovery method based on the Gaussian Pigeon-inspired Optimization Algorithm(GPIO)is proposed.On the basis of unified coding of the process route,considering the same process information and the sorting information of the process,a new comprehensive index is proposed to describe the similarity between the process routes,and the dissimilarity matrix is constructed.At the same time,in order to optimize the cluster analysis process,the Pigeon-inspired Optimization Algorithm(PIO)is introduced and the Gaussian Pigeon-inspired Optimization Algorithm is proposed to improve the clustering effect.Intelligent clustering of the routing is achieved.(2)Because the mesh process route has the advantages of easy system reconfiguration and reduced conversion time,and can cope with some unexpected situations in the actual production process of an enterprise,a new comprehensive index is put forward,considering the characteristics of the mesh process route,the sequence information,the same process information and the volume information of the product.This paper describes the similarity between the mesh process routes.At the same time,the clustering analysis is carried out by using the Gaussian Pigeon-inspired Optimization Algorithm.Finally,the effectiveness of the method is verified by comparison and simulation.(3)Aiming at the problem of automatic pushing process knowledge and as process knowledge plays a more and more important role,a push system based on process knowledge is constructed.On the basis of acquiring typical process route knowledge,knowledge description methods and knowledge index methods of process knowledge push system are briefly described and the overall frame of process knowledge push is constructed.It mainly includes job requirement analysis module,knowledge automatic matching module,active knowledge acquisition module,personalized knowledge pushing module,knowledge base rules,database maintenance and knowledge template building module,etc.Subsequently,each module is analyzed in detail and the key technology and implementation method of automatic pushing process route are briefly described.