The article was published in the Russian Physics Journal. The research was conducted with the financial support of the Ministry of Education and Science of Russia as part of the national project "Science and Universities".
Thermomechanical processing involves a series of operations including deformation, heating, holding, and cooling of metallic alloys to achieve the desired structure and properties. During this processing, various mechanisms of structural transformation are activated, with one of the key mechanisms being recrystallization, during which low-defect grains are formed. Throughout the processing, these grains grow by absorbing neighboring, more defective grains. As a result, a structure with new properties is established.
Previously, scientists from PNIPU developed a multi-level model to describe these processes. The model includes a set of relationships and parameters that allow for a detailed consideration of the mechanisms affecting structural changes. However, to obtain reliable numerical results, it is essential to accurately determine the model parameters in comparison with experimental data.
Researchers at Perm Polytechnic University conducted parameter identification for the developed model using numerical optimization methods. The identification process was carried out in two stages. In the first stage, approximate estimates of the parameter values were obtained. In the second stage, these values were refined by solving an optimization problem aimed at minimizing the deviation between experimental and calculated data.
“The optimization procedure enables the acquisition of physically justified parameter values for the model. Following its implementation, the correspondence between numerical and experimental data approached one percent, confirming the accuracy and validity of the procedure. The advantage of the proposed method lies in the ability to conduct identification using only experimental deformation diagrams in the absence of additional structural data,” comments Nikita Kondratiev, PhD in Physics and Mathematics, head of the laboratory for multi-level modeling of structural and functional materials at PNIPU.
The research conducted by PNIPU scientists will allow for more accurate predictions of changes in the structure and properties of materials resulting from thermomechanical processing. This enables the design of functional materials and components with optimal operational characteristics and reduces the risks of producing substandard products.