Search

• Data driven method is used for Fuel Cell prognostics under a stationary load profile. • The parameters of the Echo State Network are optimized via an optimization algorithm. • Cycle repetetion and markov chains are used to generate the stationary load profile. • The remaining useful lifetime is estimated using the developed approach. • The prognostics performance is validated using experimental data. Improving Proton Exchange Membrane Fuel Cell durability is a key that paves the way to its large scale industrial deployment. During the last five years, the prognostics discipline emerged as an interesting field for Proton Exchange Membrane Fuel Cell state of health prediction and lifetime estimation. The information provided by the prognostic module is crucial for optimizing the control strategy to extend the fuel cell lifetime. In this paper, an approach based on Echo State Network for fuel cell prognostics under a variable load is developed. The novelty of this paper is to perform prognostics under a variable load profile without prior knowledge of this latter. Two solutions are developed in this work. The first one consists of evaluating the remaining useful lifetime under a repeated load cycle. The second one is based on using Markov chains to generate estimations of the future load profile, allowing thus to overcome the need of real future load profile prior knowledge. Both proposed solutions give accurate prediction results of proton exchange membrane fuel cell remaining useful lifetime, with low uncertainties. [ABSTRACT FROM AUTHOR]

The goal of increasing efficiency and durability of fuel cells can be achieved through optimal control of their operating conditions. In order to implement such controllers, accurate and computationally efficient fuel cell models must be developed. This work presents a hybrid (physics-based and data-driven), control-oriented model for approximating the output voltage of proton exchange membrane fuel cells (PEMFCs) while operating under dynamical conditions. First, a physics-based model, built from simplified electrochemical, membrane dynamics and mass conservation equations, is developed and validated through experimental data. Second, a data-driven, neural network (echo state network) is trained, fitted and tested with the same dataset. Then, the hybrid model is formed as a parallel structure, where the simplified physics-based model and the trained data-driven model are merged through an algorithm based on Gaussian radial basis functions. The merging algorithm compares the output of both single models and assigns weights for computing the prediction of the hybrid result. The proposed hybrid model structure is successfully trained, validated and tested with an experimental dataset originating from fuel cells within an automotive PEMFC stack. The hybrid model is assessed through the mean square error index, with the result of a low tracking error. [ABSTRACT FROM AUTHOR]

In this work, a nonlinear model predictive control (NMPC) strategy is proposed to improve the efficiency and enhance the durability of a proton exchange membrane fuel cell (PEMFC) power system. The PEMFC controller is based on a distributed parameters model that describes the nonlinear dynamics of the system, considering spatial variations along the gas channels. Parasitic power from different system auxiliaries is considered, including the main parasitic losses which are those of the compressor. A nonlinear observer is implemented, based on the discretised model of the PEMFC, to estimate the internal states. This information is included in the cost function of the controller to enhance the durability of the system by means of avoiding local starvation and inappropriate water vapour concentrations. Simulation results are presented to show the performance of the proposed controller over a given case study in an automotive application (New European Driving Cycle). With the aim of representing the most relevant phenomena that affects the PEMFC voltage, the simulation model includes a two-phase water model and the effects of liquid water on the catalyst active area. The control model is a simplified version that does not consider two-phase water dynamics. [ABSTRACT FROM AUTHOR]

We apply a 2 5−1 fractional factorial Design of Experiments (DoE) test plan in order to discriminate the direct effects and interactions of five factors on the water management of a 500 We PEMFC stack. The stack is submitted to current steps between different operating levels and several responses are extracted for the DoE analysis. A strong ageing effect on stack and cell performances is observed. Therefore, in order to perform the DoE analysis, responses which values are too strongly affected by ageing are “corrected” prior to the analysis. A “virtual” stack, considered as “healthy”, is also “reconstructed” by “putting in series” the cells exhibiting very low performance drop. The results show that stacks and cells' resistivities are mostly impacted by direct effects of both temperature and cathodic inlet relative humidity and by compensating interaction between temperature and anodic overstoichiometric ratio. It also appears that two responses are able to distinguish a “healthy” stack from a degraded stack: heterogeneities in cell voltages and cell resistivities distributions. They are differently impacted by considered effects and interactions. Thus, a customised water management strategy could be developed, depending on the stack's state of health to maintain it in the best possible operating conditions. [ABSTRACT FROM AUTHOR]

The present study focuses on developing a reliable fault identification and localisation tool for Proton Exchange Membrane Fuel Cell (PEMFC). This diagnosis tool relies on 2 steps: first, a 3D model for a single PEMFC, that allows estimating local parameters (temperature, current and voltage distributions within the single cell). An experimental set-up allows reliable monitoring of different parameters used by the model to estimate the current density distributions over the 3D space zones of the cell. Furthermore, this 3D fault-sensitive model allows to simulate faults linked to poor water management (drying out and flooding) with 2 severity levels (“flooding”, “drying”, “high flooding” and “high drying”). Then, the second step consists of developing a two-layer feed-forward ANN that has been with an adapted database to localize each fault within the different segments of the single cell. The results show good recognition and localization rate for the considered fault, which allows to conclude that this approach is very promising for fault localization, which is one of the key points regarding diagnosis. [ABSTRACT FROM AUTHOR]

In this paper, a Fault Tolerant Control Strategy (FTCS) dedicated to PEMFC (Polymer Electrolyte Membrane Fuel Cell) water management is implemented and validated online on a real PEMFC system. Thanks to coupling a Fault Detection and Isolation (FDI), an adjustable controller and a reconfiguration mechanism, FTCS allows addressing the important challenge of Fuel Cell (FC) reliability improvement. Only few works have already been conducted on FTCS applied to FC actuators faults, and none of them on FC water management faults. In this work, a neural-based diagnosis tool is computed online as FDI component and is coupled to a self-tuning PID controller. This diagnosis tool shows low computational time and high detection performance. The self-tuning PID controller shows robustness against noise measurements and model uncertainties. Its low computational cost makes it a suitable control method for real-time FTCS. Performed on a PEMFC system, the FTCS shows promising results on fault diagnosis and performance recovery. [ABSTRACT FROM AUTHOR]