This paper presents an observer-based fault reconstruction method for PEM fuel cells. This method extends the results of a class of nonlinear uncertain systems with Lipschitz nonlinearities. An adaptive-gain second-order sliding mode (SOSM) observer is developed for observing the system states, where the adaptive law estimates the uncertain parameters. The inherent equivalent output error injection feature of SOSM algorithm is then used to reconstruct the fault signal. The performance of the proposed observer is validated through a hardware-in-loop emulator. The experimental results illustrate the feasibility and effectiveness of the proposed approach for application to fuel cell air-feed systems.

In this brief, we present Lyapunov-based robust 1 and adaptive higher order sliding mode (HOSM) controllers 2 for the air-feed system of polymer electrolyte membrane fuel 3 cells, which is a nonlinear single-input, single-output system 4 with bounded uncertainty. The system consists of a motorized 5 compressor, which is driven at its optimal point in order to 6 minimize the internal energy consumption of the system. This 7 brief provides an experimental demonstration of the applicabil-8 ity of the recently developed fixed-gain robust controller and 9 adaptive controller for this problem. Third-order controllers 10 are developed in order to obtain a continuous profile for 11 the input current of the compressor motor. In this regard, a 12 complete adaptive arbitrary HOSM control has been presented 13 for the first time, with Lyapunov-based proof. A performance 14 comparison between the two controllers is presented in the 15 end. 16 Index Terms-Adaptive control, finite time stabilization, 17 higher order sliding mode (HOSM), Lyapunov analysis, polymer 18 electrolyte membrane fuel cell (PEMFC), robust control. 19 I. INTRODUCTION 20 F UEL cells and their auxiliary systems pose challenging 21 control problems, as they are typically nonlinear and 22 difficult to characterize. They require robust or adaptive 23 control methods, as their physical parameters are uncertain, 24 varying with operating conditions and environmental effects.

No abstract