We present an adaptive sliding-mode extremum seeker that minimizes an unknown function that is subject to an unknown static constraint. The same algorithm can be applied when the static constraint is replaced by a dynamic one, provided that the dynamics possess strong stability properties. The application and feasibility study is focused on hydrogen consumption minimization in PEM fuel cell based systems.
Kunusch, C.; Castaños, F. Extremum seeking algorithms for minimal hydrogen consumption in PEM fuel cells. A: American Control Conference. "Proceedings of the 2013 American Control Conference". Washington: 2013, p. 1144-1149.