Reduction and predictive control design for a computational fluid dynamics model

Many models of commercial industrial applications are based on computational fluid dynamics (CFD) models. The models are usually of high order that it becomes infeasible to control or to optimize them. In the paper, it is shown that CFD models can be reduced very effectively by applying proper orthogonal decomposition. The resulting reduced CFD model has a state space structure and therefore enables application of many well-known control designs, including model predictive controllers.