PEM fuel cell air-feed system observer design for automotive applications: An adaptive numerical differentiation approach

• Adaptive first and second order sliding mode differentiators.
• Hardware-In-Loop simulation based on a commercial twin screw compressor and a real time fuel cell emulation system.
• Algebraic observability.

In this paper, an adaptive algebraic observer is proposed for a Polymer Electrolyte Membrane Fuel Cell (PEMFC) air-feed system, based on Higher Order Sliding Mode (HOSM) differentiators. The goal is to estimate oxygen and nitrogen partial pressures in the fuel cell cathode side, using measurements of supply manifold pressure and compressor mass flow rate. As the proposed technique requires the time derivatives of the states, Lyapunov-based adaptive HOSM differentiators are synthesized and implemented for estimating these derivatives without a-priori knowledge of the upper bounds of their higher order time derivatives. The performance of the proposed method is validated through experiments on a Hardware-In-Loop (HIL) test bench. These results illustrate the feasibility and effectiveness of the proposed approach.