The durability and the reliability of fuel cells are key points to promote hydrogen. To increase the fuel cell system lifetime, diagnostic, prognostic and smart control approaches are developed by researchers. They allow detecting failures, forecasting the remaining useful lifetime and act on the system through sensors to face degradation. Nevertheless, as the ageing of the FC stack is slow, some long-term tests are needed. The aim is to develop accelerated stress tests to reduce expensive and time-consuming experiments to validate durability approaches. This paper is dedicated to studying three different experimental tests on proton exchange membrane fuel cells (PEMFC) and the degradation impact of two stress factors: the frequency and the magnitude

This article presents the electrochemical aspect of a multiphysics model for a 1 kW proton exchange membrane (PEM) water electrolyzer. The electrochemical sub-model is based on established equations, incorporating corrections to the standard electrochemical formula to enhance generalization across different temperatures. The model parameters affecting electrochemical performance are identified to better represent real physical processes. Validation is conducted using experimental data obtained under various inlet temperature conditions. The optimized model accurately predicts electrochemical behavior in the temperature range of $\mathbf{4 0}^{\circ} \mathrm{C}$ to $70^{\circ} \mathrm{C}$. These results demonstrate the model’s capability to reflect temperature-dependent electrochemical processes.

International audience

Model predictive control (MPC) based energy management strategies (EMS) are promising to achieve high-efficiency power conversion for different hybrid electric vehicles. In this work, we investigate the impact of velocity prediction on the performance of EMS. For this, MPC controllers are designed respectively for fuel cell hybrid electric vehicles (FCHEV) using four prediction settings: Prescient MPC, Frozen time MPC, exponentially decreasing MPC, and MPC with Markov chain model. The comparison of the results using different driving cycles is performed to study the effects of prediction horizon and prediction accuracy on the performance of EMS, in terms of hydrogen consumption and battery charge sustainability. Simulation results show that the performance of MPC-based EMS is highly dependent on the prediction accuracy and the control horizon length. With proper velocity prediction methods and horizon length configurations, low hydrogen consumption and sustainable battery charge can be achieved. Moreover, the necessity of co-designing the prediction model and the horizon length by specifying the driving condition is highlighted.

In this paper, a Multi-Reservoir Echo State Network is used to estimate the Fuel Cell degradation, and its remaining useful lifespan. It proposes a methodology for predicting the fuel cell output voltage evolution with time. Echo State Network is a powerful Artificial Intelligence tool for time series predicting which main characteristics is the use of a reservoir of neurons, randomly created, instead of hidden layers such as for Artificial Neural Networks. Only the output layer is optimized by a multi-linear regression, resulting in a time reduced training phase. This leads to a possible increase of the reservoir size to preserve, even improve, its accuracy. However, the bottleneck linked to the use of this tool lies in its architecture optimization. This paper proposes a way to overcome the echo state network parameters optimization process by using a Multi-Reservoir Echo State Network. Then a comparison between an Echo State Network optimized algorithm and a Multi-Reservoir Echo State Network for fuel cell RUL prediction is proposed. In order to have a good prediction of the FC lifetime, an innovative approach based on the Multi-Reservoir Echo State Network is developed and validated using experimental data.

Proton Exchange Membrane Fuel cells (PEMFC) are one of the most promising fuel cell technologies, which qualify for variety of applications as power generation source. The Prognostics & Health Management of fuel cell is an emerging field, which is paving the way for large scale industrial deployment of PEMFC technology. More precisely, prognostics of PEMFC become a major area of focus nowadays that enables predicting the behavior of PEMFC to produce actionable information to extend its life span. This paper contributes the first application on data-driven prognostics of PEMFC stack under variable load for combined heat and power generation (μCHP). In brief, an ensemble structure of Summation Wavelet-Extreme Learning Machine models is proposed with a new incremental learning scheme, to achieve long-term predictions on stack state of health (SOH) and to give confidence for better decisions. The proposed prognostics model is validated on data from PEMFC stack used for a μCHP application under variable load profile for a complete year. A thorough comparison on SOH predictions results clearly shows the significance of proposed prognostics model, which can predict with few learning data for a long-term prognostics horizon around 650 hours with high accuracy and low uncertainty.

Being a very efficient and clean energy converter, a proton exchange membrane fuel cell may be utilized to power an electrical vehicle efficiently. Nevertheless, degradation mechanisms affect the lifespan of this electrochemical converter. Consequently, the estimation of the State of Health and Remaining Useful Life have been the subject of numerous researches in the past years. However, most of the methods available considering fuel cell prognostic do not allow the uncertainty quantification of the estimation that can be implemented online considering the calculation cost. As a novelty, the present article depicts a prognostic algorithm based on an Extended Kalman Filter. This observer estimates the State of Health, the speed of the degradation and also provides the estimation uncertainty. Then, an Inverse First Order Reliability Method computes the Remaining Useful Life with a 90% confidence interval based on the estimation of the observer. This method is applied on a 175 hours data set subsequent of an experiment on an 8-cells fuel cell stack that was subjected to an automotive power profile.

Switch Short-Circuit Fault (SSCF) is one of the most harmful failure mode in a DC-DC Converter. As a consequence, the earliest identification should be ensured in order to avoid the shutdown of the whole system. In this paper, a fast and simple method is introduced for detecting and identifying the faulty leg caused by a SSCF in a six-phase Interleaved Boost Converter (IBC) for Fuel Cell Vehicle application. The proposed detection technique is based on a comparison between the power MOSFET Drain-to-Source Voltage VDSin ON state with an adjustable threshold voltage VDS-ON-THby using only control pulses and driver information. After the fault detection, and its isolation, remedial actions by control reconfiguration methodology are applied to allow the DC-DC converter to operate in pre-fault conditions. To confirm the effectiveness of the suggested approach, simulation tests are reported using Matlab/Simulink©and ANSYS/Simplorer©. Also, comparative study between the developed voltage-based method and current-based method is provided by numerical simulations. The results obtained by the proposed method indicate that the SSCF can be detected not only in switch ON state but also in OFF state by adding some modifications to the proposed algorithm. In switch OFF state, the results prove similar behavior detection time for both the methods which detects fault occurrence in less than one switching period. By contrast, for the ON state, the proposed method detects the failure for a 100 kHz frequency using the voltage information provided by the driver. The advantage of this method is that the detections are valid in both states (ON and OFF), the suggested algorithm is simple, and easily configurable.

Proton Exchange Membrane Fuel Cells (PEMFC) are promising energy converters, but still suffer from a short life duration. Applying Prognostics and Health Management seems to be a great solution to overcome that issue. But developing prognostics to anticipate and try to avoid failures is a critical challenge. To tackle this problem, a hybrid prognostics approach is proposed. It aims at predicting the power aging of a PEMFC stack working at a constant operating condition and a constant current solicitation. The main difficulties to overcome are the lack of adapted modeling of the aging for prognostics, and the occurrence of disturbances creating recovery phenomena through aging. Consequently, this work proposes a new empirical model for power aging that takes into account these recoveries based on different features extracted from the data. These models are used in a joint particle filter framework directly initialized by an automatic parameter estimate process. When sufficient data are available, the prognostics can give accurate behavior predictions compared to experimentation. Remaining useful life estimates can be given with an error smaller than 5% for a horizon of 500 hours on a life duration of 1750 hours, which is clearly long enough for decision making.

Fuel cells, especially the proton exchange membrane fuel cell, are promising energy converters that can be used for various applications (transportation, mobile and stationary applications). Although, they have a limited lifespan due to degradation mechanisms (chemical and mechanical stresses) that are not completely understood. Consequently, researches have been conducted to estimate the Remaining Useful Life of such devices, in order to take mitigation actions, extending the life of this electrochemical converter. The aim of the presented work is to propose an observer based method for prognostic of Proton Exchange Membrane Fuel Cell. An Extended Kalman Filter estimates the State of Health and its derivative using a single empirical degradation model, in any operating conditions. Those estimations are used to predict the future State of Health. Once the prediction equalizes a defined threshold, the fuel cell is considered out of use and the Remaining Useful Life is given. This method is validated by simulation and then used on a set of experimental data resulting from a long term test on a 5-cell stack operated under a constant current solicitation.

This paper deals with the integration of the electric power-conditioning interface to the Fuel Cell (FC) module for embedded electric power generation. Specifically, the current study aims to propose an adequate control strategy to achieve a high efficiency of a compact 20kW DC/DC PEMFC output multiphase converter based on new wide-bandgap technology devices. The proposed control strategy is based on the part-load approach fulfilling demanding specifications such as low input current ripple, high performances of the converter over the load power range and small losses enabling the simplification of the cooling circuit.

In order to perform an energy management strategy in hybrid electrical vehicles containing fuel cells, based on a power supply linking ultra-capacitors, batteries and fuel cells, time series prediction based on wavelet transform and auto-regressive integrated moving average is proposed in this paper. By wavelet denoising, the noise is removed from a part of the signal, by the auto-regressive integrated moving average method; a modeling and a prediction are done and thanks to the wavelet transform, the different frequency bands existing in the signal are attributed to the different power sources on board. The low frequency signal is attributed to the fuel cell and/or the batteries and the high frequency signal to the UC. Simulation results show the efficiency of the proposed method.

One remaining technological bottleneck to develop industrial Fuel Cell (FC) applications resides in the system limited useful lifetime. Consequently, it is important to develop failure diagnostic and prognostic tools enabling the optimization of the FC. Among all the existing prognostics approaches, datamining methods such as artificial neural networks aim at estimating the process' behavior without huge knowledge about the underlying physical phenomena. Nevertheless, this kind of approach needs huge learning dataset. Also, the deployment of such an approach can be long (trial and error method), which represents a real problem for industrial applications where real-time complying algorithms must be developed. According to this, the aim of this paper is to study the application of a reservoir computing tool (the Echo State Network) as a prognostics system enabling the estimation of the Remaining Useful Life of a Proton Exchange Membrane Fuel Cell. Developments emphasize on the prediction of the mean voltage cells of a degrading FC. Accuracy and time consumption of the approach are studied, as well as sensitivity of several parameters of the ESN. Results appear to be very promising.

To realize the commercialization of proton exchange membrane (PEM) fuel cells, durability and reliability remain big challenges. This paper aims to develop a fault detection, identification and analysis methodology based on a commercial fuel cell system. Effect of air stoichiometry is studied using electrochemical impedance spectroscopy (EIS). Relevant faults are: oxygen starvation, water flooding and drying. Based on the EIS measurements, a non-model based methodology is proposed consisting of four parts: feature extraction based on the spectra, feature selection, fuzzy clustering and fault analysis. Validity of the proposed diagnostic methodology is verified experimentally.

This paper describes a pattern recognition approach aiming to estimate fuel cell duration time from electrochemical impedance spectroscopy measurements. It consists in first extracting features from both real and imaginary parts of the impedance spectrum. A parametric model is considered in the case of the real part, whereas regression model with latent variables is used in the latter case. Then, a linear regression model using different subsets of extracted features is used for the estimation of fuel cell time duration. The performances of the proposed approach are evaluated on experimental data set to show its feasibility. This could lead to interesting perspectives for predictive maintenance policy of fuel cell.

This paper proposes an overview about the integration of fuel cell power systems into electrical vehicles. At the present time, many car manufacturers around the world are presenting no emission vehicles build around fuel cells. But what about the general architecture of these vehicles ? What type of fuel cell is preferably used ? How can the global efficiency of the whole powertrain be optimized and using what kind of simulation model ? This paper presents some elements to answer these interesting questions.