Robust Control Of Automatic Manufacturing Systems With Uncontrollable And Unobservable Behaviors

Robustness refers to the fact that the system can still maintain a stable state even when the ideal conditions are violated or deviated.It is also an important problem to be solved in the field of automated manufacturing.With the increasing level of automation,the scale of the system is expanding day by day,in the process of manufacturing,with imperfect monitoring,automated manufacturing system is prone to undesirable state,blocking processing process and bringing huge economic losses and safety accidents.Therefore,it is a practical problem to improve the safety and robustness of the system in the field of automated manufacturing.The control strategy of discrete event systems is a key issue that must be considered in the design of systems in various fields.For observable and controllable manufacturing systems,a variety of monitoring strategies have emerged.However,in the actual manufacturing process,there are not only problems of sudden failure of resources,but also inevitable problems of devices,circuit aging and sensor failure,which lead to uncontrollable and unobservable behaviors.If there is no clear and accurate judgment of unsafe factors in the processing process,it is difficult for the system to choose an optimal strategy to avoid risks,resulting in the disorder of processing path.In order to ensure the safe operation of the automated manufacturing system and good economic benefits,it is necessary to monitor the manufacturing system with appropriate supervisory controllers.In this thesis,modeling and analysis of automated manufacturing systems are carried out by using Petri nets.In view of that unsafety factor in the automated manufacturing system,reasonable path planning,deadlock avoidance algorithm,robust control strategy and concurrency optimization scheme are proposed for the automated manufacturing system with unobservable and uncontrollable behaviors and unreliable resources.The main work and achievements are as follows:(1)An automated manufacturing system deadlock avoidance strategy with uncontrollable and unobservable behavior is proposed.On the basis of the key place,control respectively according to the characteristics of uncontrollable and unobservable events;Aiming at the uncontrollable behavior,a dynamic control step size N is proposed,which can check the number of resources online and control them in real time within the range of step size N.For unobservable behavior,control,judgment and recontrol criteria are used to prevent system from undesirable behavior.(2)A robust control algorithm for automated manufacturing systems with uncontrollable behavior and unreliable resources is proposed.According to the location of unreliable resources,processes are divided into processes with unreliable resources and other processes.Then,according to the transition properties of unreliable resources processes,the control strategies are classified and discussed.Finally,according to the transition attributes and relative positions in the system,the process priority is divided,so as to obtain a set of robust transitions.(3)A robust supervisory control strategy with unobservable behavior is proposed.According to the transition attribute and the position,the system resources are reviewed through classified changes to obtain the minimum number of resources.The number of resources is used to implement the criteria of control,judgment and recontrol.At the same time,in order to improve productivity,priority is divided according to the relative position of enabling transitions and unreliable resources,so as to obtain the robust control strategy.(4)The optimal percentage algorithm is proposed to optimize the concurrency.In view of that improvement of the imperfect points in the present solution,the optimal percentage of algorithms are proposed for the systems with the different step required for the parts to be processed,and the processing completion degree is described by using the accounting ratio,so that the relatively backward processing of the token is selected so as to bring the progress between the processes nearly synchronous.(5)Extend the control strategy.Aiming at the uncontrollable and unobservable behavior of automated manufacturing system,robust supervisory control strategy is extended to S~3PR with flexible path selection.According to the structural characteristics of S~3PR and the safety and performance criteria,a path selection mechanism is proposed to expand the linear Petri net control strategy.The research contents and achievements of this paper have some theoretical and realistic significance for the robust development of automated manufacturing system.