Kodra, Kliti. Reduced-order modeling and control of PEM fuel cells via balancing transformation and singular perturbations. Retrieved from https://doi.org/doi:10.7282/T3SX6C0C
DescriptionFuel cell systems are considered important clean energy sources with great potential for the future. The mathematical models of fuel cell – fuel processing systems (FC – FPS) are quite complex therefore it is important to simplify them for efficient study. In this thesis we apply order-reduction techniques to replace a large scale model with a much smaller one while still retaining the original behavior. Two order-reduction techniques, namely, the balancing transformation and balancing residualization applied to an 18th-order FC – FPS model are investigated in the first part of the study. The results show that the reduced system even down to 5th-order still retains the original dynamics. In the second part of the thesis we demonstrate how to put linear system in singularly perturbed form and then investigate the approximate gramians and balancing transformation of the reduced-order system. For the linear singularly perturbed system in explicit form we provide a method to evaluate the exact gramians in terms of pure slow and pure fast reduced-order Lyapunov algebraic equations and improve the approximate method available in the control literature for order-reduction of singularly perturbed systems via balancing transformation.