| In this paper, a new optimization program is developed to search out the optimum processing parameters for beam blank continuous casting. The parameters optimizing method is set up combined the multi-objective genetic algorithms (based on MATLAB) with finite element method (based on ANSYS). The finite element method is used to calculate the thermo-mechanical process for beam blank continuous casting. The multi-objective genetic algorithm is used to search out the optimum processing parameters for beam blank continuous casting. Those optimum parameters can meet all specified requirements but with a minimum expense of the operational and the design constraints and can make it possible to run the caster at its maximum productivity, minimum cost and to cast defect free products. Now, online verifying of this optimization project has been put in practice, which can prove that it is very helpful to control the real production. Compared with the zero order method, the result of this optimizing method is better. This developed optimizing method, taking full advantage of the feature of FEM that computes precisely and the feature of MOGA that search the optimal solution globally and rapidly, provides a method for solving the complex and large engineering problem, especially to the multi-objective optimizing problems. |
full paper (pdf, 720 Kb)