学 术 报 告
报告题目： An adaptive model order reduction method for parametrized evolution equations
报告人： 张永金 博士
摘要：Model order reduction (MOR) has emerged as an important tool in reducing the computational burden of large-scale systems, particularly in real-time or many-query contexts, e.g., optimization, control, and uncertainty quantification. In this talk, we present a brief introduction of projection-based MOR methods and show some applications of MOR in chemical engineering. In particular, efficient output error estimates are derived to adaptively construct reduced-order models with desired accuracy. Applications to Burgers Equations and chromatographic models demonstrate that the reduced-order models are very efficient in reducing the computational cost when applied to accelerate PDE constrained optimization and uncertainty quantification.