Research On Fault Diagnosis And Scheduling Of Flexible Manufacturing Systems Via Petri Nets

The manufacturing industry is the foundation of national economy.As the global market grows,the demands of customers become more and more diversified,and this market becomes more competitive.Flexible manufacturing systems have been extensively used in the manufacturing industry due to its flexibility and high efficiency.However,there are so many resources in a flexible manufacturing system that how to schedule these resources to maximize its profit is a challenging problem.Furthermore,once a fault occurs,we have to deal with the problem that is how to adjust an original production plan to reduce the loss.Therefore,the research on modeling,fault diagnosis,and optimal schedule of a flexible manufacturing system have great scientific significance and economic value.A computerized-numerical-control(CNC)machine is one of the most important resources in a flexible manufacturing system.In this work,a fault diagnosis method of CNC machines,and a scheduling method of flexible manufacturing system are presented.Its main contents are summarized as follows:1.As for a CNC machine tool control system,each of its components is modeled as a Petri-net unit according to its dynamic characteristics,and,then,these units are connected according to the operation logic.As a result,a labeled Petri net model of CNC machine tool control system is obtained,and can be further utilized to perform system analysis and schedule.2.According to fault behavior of CNC machines tool control system,a method is presented to model fault events by fault transitions connected with an original labeled Petri net.Consequently,the Petri net can model a CNC machine with fault behavior.In turn,a method is obtained to diagnose fault events of a CNC machine tool control system by this Petri net model according to the detected signal sequence.3.A Petri net is presented to model a flexible manufacturing system according to its tasks and CNC machines.In order to improve its production efficiency,a heuristic search method is obtained via its Petri net model.It can be used to compute aschedule policy to make the span shortest.4.Furthermore,fault transitions are added to the Petri net model of a flexible manufacturing system.The occurrence of this kind of fault transition can be detected by the above method.Then,the heuristic search method is extended so that a new scheduling strategy can be computed once fault transitions occur.A typical CNC machine tool control system and flexible manufacturing system are used to illustrate the above methods,respectively.