To develop highly formable aluminum alloys, predicting microstructure evolutions during a static recrystallization is crucial. Multi-phase-field (MPF) method has been used to simulate the static recrystallization. Precise prediction of the microstructure evolution using MPF simulation requires an accurate distribution of stored energy. However, experimental measurements of the distribution of stored energy have not been established. Therefore, this study focuses on a data assimilation (DA) based on the Bayes’ theorem for the estimation of the stored energy distribution. In this study, we developed a novel hybrid DA method to estimate the distribution of stored energy by integrating the sequential and non-sequential DA algorithms. The hybrid DA method proposed in this study enables the estimation of the initial distribution of stored energy and the subsequent states at time points where the observation data are available, including their associated uncertainties. Using the hybrid DA method, the microstructure evolution and the distribution of stored energy were estimated. The result demonstrates that the hybrid DA method is an effective way to achieve the precise prediction of the microstructure evolutions during the static recrystallization.
UMEZAWA et al. (Wed,) studied this question.