Complementary transient thermal models and metaheuristics to simultaneously identify linearly temperature-dependent thermal properties of austenitic stainless steels

Abstract

Abstract This paper presents an experimental approach for simultaneously identifying the temperature-dependent thermal conductivity (k) and specific heat (c p ) of 304 austenitic stainless steel (ASS) using complementary transient experiments and metaheuristics. Inverse thermal analysis was based on two heat conducting solids with different geometries. In estimation problems in general, one seeks to obtain as much sensitive data as possible using as few sensors as possible. Single thermocouple data were collected for each thermal model. An objective function fitting these complementary measurements to the corresponding numerical temperatures was minimized using the Lichtenberg algorithm. This metaheuristic algorithm takes advantage of more sensitive information provided by using complementary data, enabling for an accurate inverse solution, even when dealing with wide search ranges. The proposed technique provides a cost-effective and robust property estimation from tests conducted at room temperature. Single-step estimation occurred throughout the whole temperature domain to determine the parameters for linear functions representing the temperature dependence of k and c p . The obtained lines agreed well with curves from the literature. The 95% confidence bounds for the parameters of interest indicated deviations below ± 8.5%. Error analysis considering numerical and experimental processes showed an uncertainty close to ± 3%, applied to all estimated parameters.