Using a linear inverse heat conduction model to estimate the boundary heat flux with a material undergoing phase transformation

Modeling heat conduction during steel-cooling processes is challenging because solid–solid phase transformations can take place, introducing highly non-linear phenomena to the equation. Using a linear model for inverse problem be helpful estimating dissipated flux, especially complex models that use empirical parameters are not always available. However, performance and applicability of such simplified strongly process still unclear. This study presents several numerical simulations evaluate accuracy limitations solving (i.e. constant thermophysical properties no internal source) estimate boundary flux cooling material undergoing transformations. Preliminary with stable materials but temperature-dependent showed performs well except when has specific heat, like pure iron near its Curie temperature. Then, we performed 42CrMo4 steel, which undergoes and, hence, phase- latent as an source. The very small effect on estimation fast conditions; however, it lead interpretation bias in medium slow conditions due temporary underestimates reaching errors up 100%. Even estimated fluxes seem accurate (average smaller than 10%), further estimates temperature evolution or kinetics inaccurate (range uncertainties large 200 °C 20%, respectively) phase-dependent properties. Hence, using acceptable only (h>1500 Wm?2K?1) exclusively without simulations.