Research On Multi-objective Optimization Problem Of Metallurgical Rolling Scheduling Considering Energy Consumption

Rolling is an important process in the metal smelting production.In the metallurgical rolling production process,planning and scheduling are directly related to the product quality,energy consumption,and continuity of the process.Optimization can help enterprises reducing energy consumption and improving production efficiency.So there are both theoretical and practical significance to study the planning and scheduling for metallurgical rolling.This thesis mainly studies the production scheduling for continuous casting and continuous rolling of aluminum and the planning for hot rolling of steel.Mathematical modeling and intelligent optimization algorithms are proposed for the scheduling problems in the process of aluminum smelting,continuous casting and continuous rolling.Mathematical model and the heuristic algorithm based on association rules are proposed for the hot rolling planning of steel.Research and development of the steel hot rolling production planning system is also been introduced.The main research contents are as follows:1)Aiming at the production scheduling problem of aluminum continuous casting and continuous rolling,a multi-objective optimization model considering energy consumption was proposed to reduce energy consumption while ensuring the minimum manufacturing period.A single-objective mathematical model is proposed,in which the energy consumption target is limited to a certain constraint condition.The GAMS software is used to solve the problem,and calculation experiment shows that the optimal solution can be quickly find in small-scale problem.2)An improved NSGA-? algorithm based on matrix coding is proposed to solve the multi-objective production scheduling model.The dominant strength and adaptive individual difference metrics are introduced to select the dominant population,which improves the population quality and accelerates the convergence of the algorithm.The model is solved and the result is compared with the classical NSGA-? and the GAMS.The calculation experiment shows that the improved algorithm solves is faster and can obtain solution that considering both time and energy consumption.3)Aiming at the hot rolling production planning problem,the mathematical model is established,and the feasible solution is obtained by using the heuristic algorithm.The Apriori algorithm based on the data analytics is introduced to mine the regular pattern in slab data.After finding the association rules between the adjacent slab of the main material and bottleneck slab,we introduce it into the heuristic algorithm to solve the problem,in which the bottleneck slab is difficult to join the hot-rolling plan in the steel plate enterprise.4)A hot rolling production planning system is introduced.The system based on the hot rolling production process of a steel company can realize electronical management to data and automatically arrange the existing slab data,which allows manually adjustment.The data can be analyzed and displayed.The aluminum continuous casting and continuous rolling production scheduling model proposed in this thesis can be applied to practical process in production enterprises.The improved NSGA-? algorithm can also help solving the process industry multi-objective scheduling problem.The heuristic algorithm based on association rules provides a new idea in data analytics for the hot rolling planning of steel.The developed hot rolling production planning system for steel can be applied to the actual hot rolling production to optimize the production process.