Finite dimensional modeling and control of distributed parameter systems

Developing low-order models of high fidelity is important if the objective is an accurate control of the distributed parameter system (DPS). This work presents a novel method to develop a low-order models when there is no available exact model of the system. The foundations for this method, SVD-KL, are singular value decomposition (SVD) theory and the Karhunen-Love (KL) expansion. It is shown that satisfactory closed-loop performance of the nonlinear DPS can be obtained using a dynamic matrix controller designed using the finite order model.