Sensitivity analysis of stack power uncertainty in a PEMFC-based powertrain for aircraft application

Experimental data concerning the reliability of fuel cell systems (FCS) in aviation are still unavailable to technical community, while assessing reliability of a component or a whole system represents a fundamental aspect that allows a new technology to be introduced in a high safety system such as an aircraft. The main aim of this paper is to show a method to estimate the reliability of an aircraft power system based on a hydrogen fuel cell, mainly for design purposes. The method is based on a high-order adaptive response surface technique, coupled with a dynamic model of the aircraft power system, and it is applied to the failure event represented by an incorrect power supply due to the failure of sensors of the control system of the powertrain. The most important advantage of the proposed method is the low computational effort it requires. The result is a ranking of the most critical sensors to be considered in the design phase of the power system and demonstrate that accurate temperature sensors and sensor calibration are of dramatic importance for the control of the stack power, in case of powertrain based on PEM fuel cell systems.