Research On Hot Rolling Scheduling Problem Considering Preventive Maintenance

Hot rolling is the last step in steel production.It determines whether the final shape of the slab meets the expected requirements.Hot rolling dispatching is the core content and key link in the process of hot rolling production management,which directly determines product quality,output,delivery date and production benefit of enterprises.Reasonable scheduling technology can effectively use time,materials,and other resources,which can reduce energy consumption and cost,improve production efficiency to obtain the maximum benefit for enterprises.By aiming at the hot rolling scheduling problem,this thesis takes the hot rolling production process of small and medium-sized iron and steel enterprises as the research background.Specific research contents are as follows.(1)The scheduling problem of hot rolling is taken as the research object.It considers the attributes and machining requirements of each workpiece order.Minimizing economic cost,time cost,and transition was set as the optimization objective to build a multi-objective mathematical model,including the decision of the order processing sequence of the workpiece,the start and end times of the workpiece processing,the position of the roll change and machine maintenance,and the start and end times of the roll change and machine maintenance.(2)The artificial bee colony algorithm is designed to solve the problem.Firstly,the encoding and decoding methods are designed.Subsequently,the main processes of the artificial bee colony algorithm are designed.It includes initial solution generation,reconstruction design,machine collapse effect design,solution target selection and dominant nectarines selection,deep search design and optimal solution retention strategy.Among them,the refactoring part is improved according to the sorting method and the three basic refactoring methods are combined to form four reconstructions.The experimental results show that the up-down reconstruction can improve the efficiency of the algorithm.It uses Pareto high quality solution set and tournament mode to select the target and superior nectar source.Different designs for the number of artifacts are used to explore depth.The method of obtaining the Pareto optimal solution set several times is used to provide the optimal solution and to preserve the data explosion phenomenon during the process.(3)Artificial bee colony algorithm has been improved,including reconstruction optimization and tabu search optimization.In order to find a more efficient way of reconstruction,which combines the basic ascending reconstruction,descending reconstruction and maximum reconstruction and completes the reconstruction optimization through the experimental test of various reconstruction performance.The same solution is searched repeatedly in the search process.So,the tabu table,the solutions in the tabu table are not repeated searches and are set up to combine the artificial bee colony algorithm with the tabu search,which completes the optimization.(4)The designed algorithm is simulated.The simulation experiment consists of three groups.It carries out comparative experiments for standard genetic algorithm and standard artificial bee colony algorithm,seven reconstruction methods of artificial bee colony algorithm,standard artificial bee colony algorithm and improved artificial bee colony algorithm.In three groups of comparative experiments,the target values of standard artificial bee colony algorithm,rise and fall reconstruction and improved artificial bee colony algorithm are smaller in economic cost,time cost and jump cost.The standard artificial bee colony algorithm is superior to the standard genetic algorithm,the ascending and descending reconstruction mode is the optimal reconstruction mode,and the improved artificial bee colony algorithm is more efficient than the standard artificial bee colony algorithm.