journal article·2009-05

Prevention of fuel cell starvation by model predictive control of pressure, excess ratio, and current

Michael A. Danzer, Simon J. Wittmann, Eberhard P. Hofer

Journal of Power Sources· Volume 190· Issue 1
Publisher page

Abstract

Abstract Starvation of polymer electrolyte fuel cells (PEFC) takes place, especially during transients, if reactants are consumed in the fuel cell faster than they can be supplied. It is one of the main causes of aging and degeneration of fuel cells. To prevent oxidant starvation and to allow for a dynamic operation of the fuel cell, the excess ratio of oxygen needs to be adjusted rapidly by increasing the mass flow into the cathode. This increase is limited by the inertia of the actuators. Especially at fast load changes the risk of starvation is high. This problem can be faced by limiting the dynamics of load changes or by decoupling the desired load from the effective load. This work presents a decoupled approach, where the effective fuel cell current becomes a controllable variable. Consequently, the control variables are the oxygen excess ratio, the fuel cell pressure, and the effective current. To prevent starvation the control design has to guarantee that the oxygen excess ratio does not fall beyond a minimum value. This goal is achieved by model predictive control which is an optimal control scheme that incorporates actuator limitations and state constraints in the control design.
Source: openaire · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Fuel Cells and Related Materials · Electrocatalysts for Energy Conversion · Advanced Control Systems Optimization

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (16)

Show source-provided references
1. Schmittinger · J. Power Sources · 2008 · 180 · 110.1016/j.jpowsour.2008.01.070 ↗
2. Meyers · J. Electrochem. Soc. · 2006 · 153 · A143210.1149/1.2203811 ↗
3. Franco · J. Electrochem. Soc. · 2008 · 155 · B36710.1149/1.2838165 ↗
4. Vahidi · 2004 · 834
5. Golbert · 2005
6. Sun · 2004 · 828
7. Töpfer · 1990
8. Schmidt · 1991
9. Danzer · J. Power Sources · 2008 · 176 · 51510.1016/j.jpowsour.2007.08.049 ↗
10. Danzer · 2008
11. Maciejowski · 2002
12. M. Rau, Nichtlineare modellbasierte prädiktive Regelung auf Basis lernfähiger Zustandsraummodelle (in German), Ph.D. Thesis, TU München, 2003.
13. Nocedal · 2006
14. Fletcher · 2006
15. Lizarralde · 1999 · 4263
16. Jung · ASME J. Dyn Syst., Measure. Control · 2004 · 126 · 66610.1115/1.1789541 ↗