Parameter estimation for heat transfer analysis during casting processes based on ensemble Kalman filter

Abstract

It is very important for production of casts with high quality to predict and control the solidification processes of the alloy. Heat transfer analysis has been utilized for understanding solidification processes. However, it is often difficult to obtain values of all input parameters such as thermal conductivity and heat transfer coefficient precisely. In this study, a parameter estimation method in heat transfer analysis is developed based on data assimilation. In the authors’ previous study, the particle filter, a method of data assimilation, was applied to estimation of thermal conductivity and heat transfer coefficient in heat transfer analysis for mold casting, and its applicability was systematically investigated. It was shown that the particle filter is very effective in estimating these parameters. However, the particle filter suffers from a shortcoming called sample degeneracy which often prevents accurate estimation of parameters in phenomena of interest. The present study focuses on a different method of data assimilation called the ensemble Kalman filter and its applicability to the estimation of heat transfer coefficient and thermal conductivity is investigated based on twin experiments. It is shown that thermal conductivity and constant or time-dependent heat transfer coefficient can be accurately estimated independently with three and two cooling curves, respectively. Furthermore, the thermal conductivity and time-dependent heat transfer coefficient can be estimated simultaneously with high accuracy.