Optimal Scheduling Method For Flexible Machining System Based On Petri Net Model

Flexible machining systems widely exist in discrete industrial processes,and their energy consumption issues are particularly serious.Thus,the issue of efficient and energy-saving manufacturing has received widespread attention from academic circles at home and abroad.There is a complex and dynamic correlation between the energy consumption of flexible machining system and productive process.The total energy consumption of the flexible processing system is not only coupled with the operating state of the equipment and its nonlinear process parameters,but also closely related to the complex and changeable process route for product and scheduling plan,which increases the complexity of modeling and optimization in energy consumption.Arranging productive resources,processing equipment and process routes reasonably is a critical technical path to reduce energy consumption and improve efficiency.The problem of selecting the process route of each workpiece,processing equipment for each process,transportation equipment for transferring workpieces between operations and determining the processing sequence on the equipment,and the handling sequence on the transportation equipment needs to be considered and resolved in the schedule of the flexible machining system.In order to these problem,this thesis focus on three aspects:dynamic scheduling,multi-objective scheduling,and batch scheduling under multi-process routes.The main research is presented as follows:(1)Aiming at the optimal scheduling problem of flexible machining system under dynamic disturbance,a dynamic energy consumption optimal scheduling method based on timed place Petri net(TPPN)is proposed.In this method,two kinds of dynamic events,new task insertion and machine fault and repair,are considered.At the moment when dynamic event occurs,the Petri net model is re-established from the moment to the makespan in this method.Then the model is re-optimized based on modified dynamic programming method,and the scheduling plan of the system after the disturbance occurs is solved.The reliability of the model in the scheduling process,and the feasibility of this method under dynamic disturbance are verified by the example simulation.(2)Aiming at the multi-objective optimal scheduling problem of flexible machining system,a multi-objective collaborative energy-saving scheduling method based on timed transition Petri net(TTPN)model was proposed.This method aims at energy consumption,completion time and machine utilization balance.The two-level integer encoding and decoding rule is designed to implemented one-to-one correspondence between chromosome encoding in algorithm and feasible transition sequences in the reachability graph for the TTPN model based on the NSGA-Ⅱ(Non-dominated Sorting Genetic Algorithms-Ⅱ)algorithm.Meanwhile,a hybrid population initialization strategy and simulated annealing local search mechanism are adopted to improve the algorithm’s performance.The feasibility of this method and the superiority of the obtained solution set are verified by numerical examples.(3)A heuristic search algorithm based on the reachability graph of TTPN model is proposed for the batch scheduling problem of flexible machining system under complex working conditions of workpiece transportation and product multi-process routes.Taking the states in the algorithm search process as input,a heuristic prediction function for the actual cost function and the remaining cost is established,which realizes the filtering of many states in each stage,and only searches some states in the next stage,thereby reducing the reachability graph search space.The reliability of the TTPN model and the feasibility of the search algorithm are verified by case analysis and algorithm comparison.