OPTIMAL-CONTROL OF CONTINUOUS-CASTING BY NONDIFFERENTIABLE MULTIOBJECTIVE OPTIMIZATION

A new version of an interactive NIMBUS method for nondifferentiable mu ltiobjective optimization is described. It is based on the reference p oint idea and the classification of the objective functions. The origi nal problem is transformed into a single objective form according to t he classification information. NIMBUS has been designed especially to be able to handle complicated real-life problems in a user-friendly wa y. The NIMBUS method is used for solving an optimal control problem re lated to the continuous casting of steel. The main goal is to minimize the defects in the final product. Conflicting objective functions are constructed according to certain metallurgical criteria and some tech nological constraints. Due to the phase changes during the cooling pro cess there exist discontinuities in the derivative of the temperature distribution. Thus, the problem is nondifferentiable. Like many real-l ife problems, the casting model is large and complicated and numerical ly demanding. NIMBUS provides an efficient way of handling the difficu lties and, at the same time, aids the user in finding a satisficing so lution. In the end, some numerical experiments are reported and compar ed with earlier results.