A method of estimating a state vector of a fuel cell system, a method of controlling a state quantity of a fuel cell system, and a fuel cell system

摘要

The invention relates to a method for estimating a state vector of a fuel cell system by means of an observer algorithm, in particular a nonlinear observer algorithm, such as by means of a Kalman filter. An observation matrix of the algorithm connects a plurality of measured values of the fuel cell system with a plurality of state variables of the fuel cell system. A system matrix of the algorithm is derived from balance equations of at least one operating medium of the fuel cell system. The balance equations preferably describe the entire fuel cell system and are preferably first set up for subsystems of the fuel cell system and then connected to one another via coupling terms. According to the invention, the observation matrix is based on submodels which preferably describe subsystems or system components of the fuel cell system arranged therein. A plurality of state variables of the fuel cell system are combined into a state vector of the system and developed over time by means of the observation matrix and the system matrix. The invention likewise relates to a method for regulating or controlling a state variable of the fuel cell system. The regulation or control is based on the estimated or temporally developed state vector of the fuel cell system. Likewise provided by the invention is a fuel cell system designed to carry out the method.