Intelligent Scheduling Method Of Mixed Operation Production Line Based On Improved Genetic Algorithm

The steel industry is a key industry supporting the development of the national economy,and roll grinding is an important component of steel rolling production.As a typical mixed operation production line,the scheduling scheme of the roll grinding production line directly affects the efficiency of steel rolling production.Different from the single flexible job shop or crane scheduling problem,the mixed operation production line needs to consider multiple workpieces,multiple processes,multiple equipment and multiple cranes at the same time.Although the traditional expert system based on human experience can obtain a feasible scheduling scheme,it cannot guarantee the optimal performance of the scheme.Scholars at home and abroad have conducted extensive and in-depth research on the scheduling problem of a single flexible job shop or crane.At the same time,some scholars have studied the flexible job shop scheduling problem that considers the transportation time and includes a single crane,but the single crane system avoids the interference problem between multiple cranes.How to achieve efficient and intelligent scheduling under the constraints of "workpiece-machine-crown" space-time coupling is a key issue facing the current hybrid production line.This paper takes the roll grinding production line,a typical mixed operation production line,as the object,and conducts in-depth research on the intelligent scheduling method to solve the problem of efficient scheduling of the mixed operation production line.Firstly,the mathematical model of the scheduling problem of mixed operation production line is established,and then an intelligent scheduling method based on improved genetic algorithm is proposed and verified by simulation.Finally,the proposed method is verified by the Siemens Plant Simulation software.The details are as follows:（1）The production process of the roll grinding production line is analyzed.From the perspective of problem optimization,the objectives,variables and constraints of the scheduling problem are clarified,and the minimization of the maximum completion time is selected as the objective function.The spatial and temporal constraints among the workpiece process,machine and crane are given,and finally a hybrid production line scheduling model is established,and the optimization and convergence process of the model is given by taking the genetic algorithm as an example,which provides a basis for algorithm improvement.（2）An improved genetic algorithm is proposed to improve chromosome coding and initial solution generation,selection,crossover,mutation,etc.,to generate a feasible solution to the mixed operation production line scheduling problem,and to enhance its ability to seek optimization and jump out of local optimum.In the decoding part,a series of operators are designed according to the action of the crane to realize the scheduling of the crane based on the action decomposition,and finally a hybrid operation production line scheduling method under the constraints of space-time coupling is formed.Comparing the improved genetic algorithm with the original genetic algorithm and the scheduling algorithm based on human experience,the results show that the proposed improved genetic algorithm is obviously better.（3）The improved genetic algorithm is applied to Siemens Plant Simulation software for scene verification.Two typical scenarios were designed,and the model settings,crane and material control settings of the scenarios were completed.The scene verification data is collected and analyzed.The results show that,consistent with the simulation verification,the proposed improved genetic algorithm shows better effectiveness and superiority than the other two methods in the scene verification.Provides core algorithm support.