HOME

CURRICULUM VITÆ

RESEARCH

LIVE VIDEO

PUBLICATIONS

TEACHING

PROJECTS

HOBBIES

WEB DESIGN

CALENDAR

Abstract
Discrete optimal control methods are applied to a PEM hybrid fuel cell vehicle (HFCV) for vehicle to grid (V2G) applications. The HFCV model is developed from past control-oriented models. Grid power demand is modeled by adapting a 24-hour power forecast created by CAISO. The control problem formulation includes regulating both battery state of charge and fuel cell system efficiency. Linearization is performed about a specified operating point and the resulting transfer functions are discretized for a sampling frequency four times faster than the fastest plant dynamics. Linear systems analysis is performed on the discrete pulse transfer functions to extract physical interpretations about system dynamics. A linear quadratic regulator (LQR) and Luenberger estimator are synthesized in both continuous and discrete time to achieve the desired performance specifications. A thorough discussion on LQ weight selection is provided in addition to a presentation of several analysis tools. Observer pole placement and state estimation are analyzed with respect to disturbance rejection and error convergence. Conclusions are drawn about the effect of sampling time for the dynamic model, considering a 24-hour process disturbance cycle. To close, a summary of the salient lessons learned from the project is presented, along with recommendations for further extending this study.

REPORT | PRESENTATION