Research On Active Perception And Collaborative Decision Of Manufacturing State Based On Edge Computing

With the rapid development of Cyber-Physical Systems,Big Data and Internet of Things technologies,smart manufacturing has become the future development direction of the manufacturing industry.It has brought the higher requirements of reaction speed of the machining operations and accuracy of management decisions.Under this background,the smart manufacturing models represented by cloud manufacturing and digital twins,have aroused the widespread concern in academia and industry.The research on methods and technologies to realize the active perception,calculation of manufacturing process status and human-machine collaborative decision-making is a difficult problem.Therefore,the manufacturing process state of discrete workshops was taken as the research object,and the active perception and collaborative management of manufacturing events supported by edge computing were studied in this paper.First,as for the problem of manufacturing data collection and active perception calculation,a real-time manufacturing data collection model was built based on the analysis of state awareness needs;the state data was transformed into original events,and related rules were extracted through adopting Apriori algorithm to implement rule-based reasoning of manufacturing key events.To achieve the workshop logistics data collection,an improved flower pollination algorithm was used to solve the logistics model.Secondly,as for the problem of modeling and computing service configuration of manufacturing CPS nodes based on edge computing,the Protégé was used to establish the ontology of manufacturing resources to realize the standardized description,reasoning and inquery;a CPS-based computing service configuration model was put forward to reduce the information processing delay.Taking the load balancing as the goal,the flower pollination algorithm was used to solve the problem of computing resource allocation at edge nodes.Then,as for the problem of Human-Machine collaborative scheduling decision based on real-time manufacturing information,a framework model of human-machine collaborative scheduling was established form the CPS demand.Based on the basis of the flexible scheduling modeling,the interactive,imperative and warning man-machine collaborative decision-making schemes were designed from the disturbance analysis in manufacturing systems.Finally,a genetic algorithm was used to solve the man-machine collaborative scheduling decisionmaking scheme.Finally,using the MATLAB software,a prototype system with three major functions including the active perception of manufacturing states,manufacturing resource allocation and Human-Machine collaborative scheduling decision was developed.After that,the research case in this article was used to demonstrate the operation process of the prototype system developed,and the feasibility of the research content in this article was verified.