Characteristic blower map modeling for volume flow rate estimation in solid oxide fuel cell systems

Abstract

In this paper characteristic blower map modeling approaches are implemented in a solid oxide fuel cell system as a replacement to flow rate sensors, reducing pressure drop and investment costs. Blower laws of similitude and different regression modeling approaches based upon detailed manufacturer's data are applied to express the volume flow rate as a function of the measured blower speed, pressure and temperature. The indirect model based volume flow rate calculation is compared to a direct volume flow rate measurement in a test bench using a side channel blower operated with ambient air. For blower operational modes with low to medium pressure differences, high matching between the indirect and direct measurement of the volume flow rate is achieved, showing the high accuracy of the model. Furthermore, the superiority of the modeling approach is underlined by reduced occurring absolute error span and average error span of 55% and 66%, respectively. Replacing the air flow rate sensor with a characteristic blower map modeling for the determination of the air flow rate in a 10 kW solid oxide fuel cell system leads to potential increases of the electrical efficiency between 0.5%- and 0.7%-points due to the reduced pressure drop.