Abstract
The work was inspired by the fact that commonly used isothermal dilatometric tests may give erroneous results. Application of the optimization techniques to the interpretation of more complex constant cooling rate dilatometric tests is proposed in the paper. These tests are more reliable experimentally but more difficult in interpretation. Thus, the inverse algorithm, which uses non-gradient optimization methods, was developed and applied to solve this problem. The identification of phase transformation models for various carbon-manganese and eutectoid steels is performed on the basis of constant cooling rate tests. The models, after identification, are implemented in the finite element code, which simulate controlled cooling of products after hot forming. The program was tested for cooling of angles, rods and rails.