Research On Modeling And Optimization For Scheduling In Aluminum Electrolysis-Casting Processes

Electrolysis and casting is the main process of aluminum smelting.The optimal scheduling of electrolytic casting is mainly about the batch scheduling of aluminum liquid in the process of casting aluminum from the electrolytic cell to the alloy furnace.A reasonable scheduling scheme for aluminum liquid can effectively improve the production efficiency,ensure the quality of products and reduce the energy consumption.The electrolytic casting batch scheduling need to consider multi equipment production environment,the problem has the characteristics of large-scale,nonlinear and complex structure,which makes it is difficult for manual processing by enterprise technical personnel,and modeling and optimization techniques is needed for aluminum-batch scheduling problem.This thesis mainly focuses on the research of mathematical modeling and intelligent optimization algorithm for the aluminum production process from the aluminum electrolysis,the vacuum package transportation,the alloy smelting furnace,and casting on the foundry unit.1)For the aluminum electrolytic-casting process two stage batch scheduling problem with minimize makespan as the objective,considering the capacity constraint of vacuum package,batch melting capacity constraints and melting batch adjustment time,a discrete time model and a unit-specific event based continuous time model are formulated.Standard optimization software CPLEX is used to solve and analysis the performance the two proposed models.2)From the point view of solving practical industrial problems,a batch scheduling model of aluminum electrolysis-casting,which considers more practical constraints and objective functions,is established.The objective function of the model is the liquid aluminum purity difference to minimize the makespan and the two group in the process of cost,considering the constraints including electrolysis aluminum liquid,the arrival time difference,the purity of liquid aluminum electrolyzer position,vacuum package capacity,transport distance,furnace capacity,washing furnace operation etc..This is a mixed integer nonlinear programming model.For the nonlinear constraints,the equivalent constraint transformation is carried out,and the model is transformed into a mixed integer linear programming model,and is solved by CPLEX software.3)An improved differential evolution algorithm based on integer encoding is proposed to solve the problem of the two-stage batch scheduling of aluminum electrolysis casting for industrial practice.A new evolutionary strategy is proposed to improve the problem that the classical difference is easy to fall into the local optimal.Computation results indicate that the optimal algorithm can be obtained by small and medium scale problems quickly;in the large-scale problem,also can be in a reasonable time for fast approximate solution can meet the actual needs of production.The scheduling model proposed in this thesis can be applied to the actual aluminum production enterprises as a decision support for the production scheduling of enterprises.The improved differential evolution algorithm can also be applied to the two stages of solving the batch production scheduling problem of process industry.