Resource Optimization Algorithm For Automatic Manufacturing System Using Petri Nets

Resource optimization of automated manufacturing system(AMS)refers to the premise of meeting the preset production task,using the least resource,in order to achieve the purpose of saving production cost and improving economic benefits.As AMS is strongly discrete and event-driven,it is often abstracted as a discrete event system for research at the logic level.Petri net is widely used in the research of discrete event systems due to its dual representation of mathematics and graphics as well as abundant system analysis means.This paper studied the AMS resource optimization problem under different working conditions based on the labeled Petri net and the Petri net model with unobstructible transitions.In order to avoid calculating the state space,the resource optimization problem was transformed into an integer linear programming problem based on the analysis of the structural characteristics and related properties of Petri nets,and the problem was solved by LINGO.The case analysis showed that the proposed method can reduce the computational complexity,improve the computational efficiency,and provide a theoretical method for solving the practical resource optimization problems of AMS.The main research work and achievements are summarized as follows:(1)An initial resource optimization algorithm was proposed for the Petri net model with all observable transitions.Based on the structural characteristics of Petri net and the equation of state method,the constraint relationship between the task to be executed and the initial resource was obtained,and the specific rules and algorithms for transforming the resource optimization problem into an integer linear programming problem were proposed.Finally,the algorithm was extended to the resource optimization problem of labeled Petri net model,and the initial resource optimization scheme of labeled Petri net was obtained by solving multiple integer linear programming problems.(2)Due to technical or cost constraints,sensors cannot be deployed at all parts of AMS,so it is difficult to monitor the occurrence of some events,which are usually modeled as unobservable transitions.An initial resource optimization algorithm was proposed for Petri net model with unobservable transitions.Given a predetermined production task,according to the structural characteristics of Petri net and transition firing rules,the linear constraint relations between observable transition sequence,unobservable transition sequence and initial resource were given.The resource optimization problem of the model was analyzed by integer linear programming method,and the initial resource optimization algorithm was proposed.Finally,the algorithm was extended to the resource optimization problem of labeled Petri net model with unobservable transition.(3)S~3PR net is a resource allocation Petri net,which is often used for modeling AMS.In this paper,the proposed resource optimization algorithm of labeled Petri Net and the resource optimization algorithm of labeled Petri Net with unobservable transition were applied to S~3PR net for example analysis to verify the effectiveness of the proposed algorithm.