Pore network modeling of cathode catalyst layer of proton exchange membrane fuel cell

In this paper, a pore network model is developed to investigate the coupled transport and reaction processes in the cathode catalyst layer (CCL) of proton exchange membrane fuel cell (PEMFC). The developed model is validated by comparing the predicted polarization curve with the experimental data, and the parametric studies are carried out to elucidate the effects of CCL design parameters. With the decrease of the CCL thickness and the Nafion content, the cell voltage reduces at the low current density but increases when the current density is higher. The cell performance is also improved by increasing the proton conductivity of the Nafion film in the CCL. As compared to the CCL of uniformly distributed Nafion, the CCL with the Nafion volume decreasing along the thickness direction exhibits better performance at the high current density.     

Maximum power point tracking supported proton exchange membrane fuel cell based intelligent dynamic voltage restorer

In modern era, electrical power utilities are more concerned about power quality. In this endeavour, dynamic voltage restorer (DVR) provides adequate support to the system. Accordingly, the present work illustrates intelligent hybrid control mechanism for DVR. Artificial neural network (ANN) is incorporated to obtain real-time optimal gains under distinct voltage situations. Closed loop type 2 fuzzy logic (CLT2-FL) is also realized in order to assist the ANN supported unit concomitantly. In order to enhance the potential of the present DVR, a CLT2-FL controlled maximum power point tracking (MPPT) based proton exchange membrane fuel cell arrangement is also explored in the present study. CLT2-FL module is adopted in DC-DC converter topology to provide simultaneous supply to the loads at different and regulated voltage levels. Consequently, the results are evaluated and compared to the state-of-the-arts which unveil the efficacy of the implemented controller against the oddity seen in the voltage waveform, thereby, exhibiting better voltage regulation and less harmonics. The effectiveness of the implemented MPPT unit from the viewpoints of convergence speed and oscillations is also established.     

Prediction and analysis of the cathode catalyst layer performance of proton exchange membrane fuel cells using artificial neural network and statistical methods

A mathematical model was developed to investigate the cathode catalyst layer (CL) performance of a proton exchange membrane fuel cell (PEMFC). A numerous parameters influencing the cathode CL performance are implemented into the CL agglomerate model, namely, saturation and eight structural parameters, i.e., ionomer film thickness covering the agglomerate, agglomerate radius, platinum and carbon loading, membrane content, gas diffusion layer penetration content and CL thickness. For the first time, an artificial neural network (ANN) approach along with statistical methods were employed for modeling, prediction, and analysis of the CL performance, which is denoted by activation overpotential. The ANN was constructed to build the relationship between the named parameters and activation overpotential. Statistical analysis, namely, analysis of means (ANOM) and analysis of variance (ANOVA) were done on the data obtained by the trained neural network and resulted in the sensitivity factors of structural parameters and their mutual combinations as well as the best performance.     

Experimental study on the dynamic performance of an air-cooled proton exchange membrane fuel cell stack with ultra-thin metal bipolar plate

In this paper, a compact 3 kW air-cooled fuel cell stack consists of 95 single cells with metallic bipolar plate is designed. Compared with graphite bipolar plates, metal stamping bipolar plates are lighter in weight, smaller in size and faster in heat conduction, therefore the transient behaviors of the voltage and temperature of each cell are analyzed. The results show that the heat distribution of the air-cooled fuel cell is very uniform, and the temperature difference between the inlet and outlet of cathode air of the fuel cell is lower than 15 °C. The individual cell voltage uniformity percentage variation value reaches 7% when the drop in the loading current is over 25 A. Moreover, the voltage uniformity variation value is higher than 4% when the loading current output exceeds 35A. Thus, a large drop in loading and a high loading current easily increase the voltage uniformity variation value. Long-term continuous operation has a negative influence on the performance of the stack, especially the last fuel cell near the anode outlet. Anode purging can effectively alleviate the uniformity percentage variation in the voltages. The designed air-cooled fuel cell exhibits good performance and strong environmental adaptability.     

Experimental study and mitigation of abnormal behavior of cell voltage in a proton exchange membrane fuel cell stack

The aim of this study is to investigate the abnormal behavior of cell voltage in a proton exchange membrane fuel cell stack and a mitigation strategy. The proposed strategy is simple and requires only a three‐way solenoid valve to replace the direct way solenoid valve of the original system. It is applied to a proton exchange membrane fuel cell stack with a dead‐ended anode to verify its validity. The behavior of the cell voltages in the stack is discussed in detail, especially the cell reversal process. The results show that the proposed strategy can significantly reduce the severity of hydrogen starvation. And the maximum power of the stack is increased by 10.67%. It is a sudden increase related to cell reversal mitigation. Uneven hydrogen distribution is the cause of low cell voltage and cell reversal. This strategy increases the cell voltage by increasing the hydrogen content in the anode flow channel downstream. It also significantly reduces the fluctuations in cell voltage and improves the uniformity of the cell voltage. This experimental study contributes to mitigate hydrogen starvation in cells of proton exchange membrane fuel cell stacks in application.     

Improvement of the three-dimensional fine-mesh flow field of proton exchange membrane fuel cell (PEMFC) using CFD modeling,artificial neural network and genetic algorithm

This study proposes a systematic methodology for improving PEMFC's performance combining computational fluid dynamic (CFD), artificial neural network (ANN), and intelligent optimization algorithms. Firstly, a three-dimensional (3-D) multiphase PEMFC CFD model with 3-D fine-mesh flow field is developed. Then the key structural features of the fine-mesh flow field are extracted as optimization decision variables, and the sampling points are selected by using the Latin hypercube sampling (LHS) experimental method. The power density and oxygen uniformity index of sampling points are calculated by CFD modeling to form the database, which is used to train the artificial neural network (ANN) surrogate model. Finally, the single-objective optimization (SOO) and multi-objective optimization (MOO) are implemented by using genetic algorithm (GA) and non-dominated sorting genetic algorithm (NSGA-II), respectively. It was found that using trained ANN surrogate models can get a high prediction precision. The maximum power density of SOO is increased by 7.546% than that of base case and is 0.562% larger than that of MOO case. However, the overall pressure drop in cathode flow field of SOO case is greater than that of MOO case and the base case. Furthermore, the oxygen concentration, the oxygen uniformity index and the water removal capacity of MOO case are better than that of SOO case. It is recommended that the improved flow field structure optimized by MOO is more beneficial to improve the overall performance of PEMFC.     

Coupling flow channel optimization and Bagging neural network to achieve performance prediction for proton exchange membrane fuel cells with varying imitated water-drop block channel

The proton exchange membrane fuel cell (PEMFC) flow channel structure obviously affects the reaction gas distribution and electrochemical reactions. In this study, the imitated water-drop block heights and widths within the channel are optimized for better PEMFC performance. A machine learning-based Bagging neural network is applied for the first time to predict PEMFC output performance based on different block structure parameters. First, the proposed imitated water-drop block height and width are optimized by changing parameters. Then, a database is established. Finally, after the Bagging model is validated, the performance is compared with the back-propagation (BP) neural network. Results indicate that the mass transfer and the electrochemical reaction are improved under the optimal width and height of imitated water-drop block for PEMFC. The Bagging prediction model uses less training data to obtain high-precision prediction results in 10 s. The performance prediction model can effectively improve the efficiency of channel optimization.     

Numerical studies on an air-breathing proton exchange membrane (PEM) fuel cell stack

Air-breathing proton exchange membrane (PEM) fuel cells provide for fully or partially passive operation and have gained much interest in the past decade, as part of the efforts to reduce the system complexity. This paper presents a detailed physics-based numerical analysis of the transport and electrochemical phenomena involved in the operation of a stack consisting of an array of vertically oriented air-breathing fuel cells. A comprehensive two-dimensional, nonisothermal, multi-component numerical model with pressurized hydrogen supply at the anode and natural convection air supply at the cathode is developed and validated with experimental data. Systematic parametric studies are performed to investigate the effects of cell dimensions, inter-cell spacing and the gap between the array and the substrate on the performance of the stack. Temperature and species distributions and flow patterns are presented to elucidate the coupled multiphysics phenomena. The analysis is used to determine optimum stack designs based on constraints on desired performance and overall stack size.     

Development of a proton exchange membrane fuel cell cogeneration system

A proton exchange membrane fuel cell (PEMFC) cogeneration system that provides high-quality electricity and hot water has been developed. A specially designed thermal management system together with a microcontroller embedded with appropriate control algorithm is integrated into a PEM fuel cell system. The thermal management system does not only control the fuel cell operation temperature but also recover the heat dissipated by FC stack. The dynamic behaviors of thermal and electrical characteristics are presented to verify the stability of the fuel cell cogeneration system. In addition, the reliability of the fuel cell cogeneration system is proved by one-day demonstration that deals with the daily power demand in a typical family. Finally, the effects of external loads on the efficiencies of the fuel cell cogeneration system are examined. Results reveal that the maximum system efficiency was as high as 81% when combining heat and power.     

A hybrid neural network model for PEM fuel cells

The goal of this paper is to discuss a neural network modeling approach for developing a quantitatively good model for proton exchange membrane (PEM) fuel cells. Various ANN approaches have been tested; the back-propagation feed-forward networks and radial basis function networks show satisfactory performance with regard to cell voltage prediction. The effects of Pt loading on the performance of the PEM fuel cell have been specifically studied. The results show that the ANN model is capable of simulating these effects for which there are currently no valid fundamental models available from the open literature.

Two novel hybrid neural network models (multiplicative and additive), each consisting of an ANN component and a physical component, have been developed and compared with the full-blown ANN model. The results from the hybrid models demonstrate comparable performance (in terms of cell voltage predictions) compared to the ANN model. Additionally, the hybrid models show performance gains over the physical model alone. The additive hybrid model shows better accuracy than that of the multiplicative hybrid model in our tests.     

Parameter identification of proton exchange membrane fuel cell via Levenberg-Marquardt backpropagation algorithm

It is essential to develop an accurate model of proton exchange membrane fuel cell (PEMFC) for a reliable operation and analysis, in which unknown parameters usually need to be determined. The inherent nonlinear, strong coupling, and diversification of PEMFC model seriously hinder traditional methods to identify the parameters. For the sake of overcoming these thorny obstacles, Levenberg-Marquardt backpropagation (LMBP) algorithm based on artificial neural networks (ANNs) is proposed for PEMFC parameter identification. Furthermore, the performance of LMBP is thoroughly evaluated and compared with four typical meta-heuristic algorithms under three cases. Simulation results indicate that LMBP performs a higher accuracy and faster speed for parameter identification. In particular, accuracy and convergence speed can achieve as much as 99.8% and 95.9% growth via LMBP, respectively.     

Aging prognosis model of proton exchange membrane fuel cell in different operating conditions

The aging prognosis model of Proton Exchange Membrane Fuel Cell (PEMFC) can predict the aging state of PEMFC to develop an effective prognostic maintenance plan. This paper proposes an aging prognosis model of PEMFC in different operating conditions based on the Backpropagation (BP) neural network and evolutionary algorithm. The influence of PEMFC current, hydrogen pressure, temperature, and relative humidity on the aging of PEMFC can be considered by the proposed method. Firstly, the aging prognosis model of PEMFC is built by the BP neural network. Then, the evolutionary algorithm including Mind Evolutionary Algorithm (MEA), Particle Swarm Optimization (PSO), and Genetic Algorithm (GA) is used to optimize the parameters of the established aging prognosis model of PEMFC. Finally, the accuracy of the proposed aging prognosis model is validated by 3 PEMFC aging experiments in different operating conditions. The results show that MEA, GA, and PSO can greatly improve the accuracy of the aging prognosis model of PEMFC. The MEA improves the accuracy by 10 times, while the computing time increases by 0.085s. The proposed MEA-BP that has a very short computing time can be applied to online applications.     

Effects of chlorides on the performance of proton exchange membrane fuel cells

This paper investigates the effects of cathode gases containing chloride ions on the proton exchange membrane fuel cell (PEMFC) performance. Chloride solutions are vaporized using an ultrasonic oscillator and mixed with oxygen/air. The salt concentration of the mixed gas in the cathode is set by varying the concentration of the chloride solution. Five-hour tests show that an increase in the concentration of sodium chloride did not significantly affect the cell performance of the PEMFC. It is found that variations in the concentration of chloride do not show significant influence on the cell performance at low current density operating condition. However, for high current density operating conditions and high calcium chloride concentrations, the chloride ion appears to have a considerable effect on cell performance. Experimental results of 108-h tests indicate that the fuel cell operating with air containing calcium chloride has a performance decay rate of 3.446 mV h⁻¹ under the operating condition of current density at 1 A/cm². From the measurements of the I-V polarization curves, it appears that the presence of calcium chloride in the cathode fuel gas affects the cell performance more than sodium chloride does.     

Stack shut-down strategy optimisation of proton exchange membrane fuel cell with the segment stack technology

In the previous researches, researchers mainly focus on the single cell which is far away from the practical application. In this paper, shut-down process is studied in a 5-cell stack with segment technology. In the unprotected group, the hydrogen/air boundary is observed, and the output voltage performance degrades greatly after 300 start-stop cycles. A 2-phase auxiliary load strategy is proposed to avoid the hydrogen/air boundary. The lifetime is extended. But a serious local starvation is observed during the shut-down process. And corrosion happened in the inlet region. To avoid the starvation, the second strategy is designed, which combines 2-phase auxiliary and air purge (2-phase load& air purge strategy). With the new strategy, the degradation of the stack after 1500 cycles is acceptable, and the carbon corrosion in the inlet is effectively reduced. It could conclude that the hydrogen/air boundary is the main cause of the degradation of fuel cell during an unprotected shut-down process. And a strategy only with auxiliary load may suffer from the local starvation. The purge process can avoid the vacuum effect in the fuel cell caused by the auxiliary load. Therefore, adding an air purge during the shut-down process is promising in vehicle fuel cell.     

Performance of a proton exchange membrane fuel cell stack with thermally conductive pyrolytic graphite sheets for thermal management

This work experimentally investigates the effects of the pyrolytic graphite sheets (PGS) on the performance and thermal management of a proton exchange membrane fuel cell (PEMFC) stack. These PGS with the features of light weight and high thermal conductivity serve as heat spreaders in the fuel cell stack for the first time to reduce the volume and weight of cooling systems, and homogenizes the temperature in the reaction areas. A PEMFC stack with an active area of 100 cm² and 10 cells in series is constructed and used in this research. Five PGS of thickness 0.1 mm are cut into the shape of flow channels and bound to the central five cathode gas channel plates. Four thermocouples are embedded on the cathode gas channel plates to estimate the temperature variation in the stack. It is shown that the maximum power of the stack increase more than 15% with PGS attached. PGS improve the stack performance and alleviate the flooding problem at low cathode flow rates significantly. Results of this study demonstrate the feasibility of application of PGS to the thermal management of a small-to-medium-sized fuel cell stack.     

Experimental analysis of dynamic characteristics on the PEM fuel cell stack by using Taguchi approach with neural networks

This study determines the optimum operating parameters for a proton exchange membrane fuel cell (PEMFC) stack to obtain small variation and maximum electric power output using a robust parameter design (RPD). The operating parameters examined experimentally are operating temperatures, operating pressures, anode/cathode humidification temperatures, and reactant flow rates. First, the dynamic Taguchi method is used to obtain the maximum and stable power density against the different current densities, which are regarded as the systemic inputs considered a signal factor. The relationship between control factors and responses in the PEMFC stack is determined using a neural network. The discrete parameter levels in the dynamic Taguchi method can be divided into desired levels to acquire real optimum operating parameters. Based on these investigations, the PEMFC stack is operated at the current densities of 0.4–0.8 A/cm². Since the voltage shift is quite small (roughly 0.73–0.83 V for each single cell), the efficiency would be higher. In the range of operation, the operating pressure, the cathode humidification temperature and the interactions between operating temperature and operating pressure significantly impact PEMFC stack performance. As the operating pressure increasing, the increments of the electric power decrease, and power stability is enhanced because the variation in responses is reduced.     

A novel three-dimensional, two-phase and non-isothermal numerical model for proton exchange membrane fuel cell

A three-dimensional, two-phase and non-isothermal model of a proton exchange membrane fuel cell (PEMFC) based on the previously developed model is established using the two-fluid method. This two-phase model considers the liquid water transport in both cathode and anode sides and accounts for the intrinsic heat transfer between the reactant fluids and the solid matrices. The latent heat of water condensation/evaporation is considered in the present model. The numerical results demonstrate that the lower cathode humidity is beneficial for cell performance. In the anode side, the water vapor can be condensed at high current density because the water vapor transport is less than the hydrogen consumption rate. Near the catalyst layer, the reactant fluid temperature is higher than the solid matrix temperature, and far from the catalyst layer, the temperature difference between the reactant fluid and the solid matrix decreases. Near the channel, the reactant fluid temperature is lower than the solid matrix temperature.     

Performance degradation prediction of proton exchange membrane fuel cell using a hybrid prognostic approach

The proton exchange membrane fuel cell has been widely used for industrial systems; however, its performance gradually degrades during use. Therefore, the study on the performance degradation prediction of fuel cells is helpful to extend its lifespan. In this paper, a novel hybrid approach using a combination of model-based adaptive Kalman filter and data-driven NARX neural network is proposed to predict the degradation of fuel cells. The overall degradation trend (i.e., irreversible degradation process) is captured by an empirical aging model and adaptive Kalman filter. Meanwhile, the detail degradation information (i.e., reversible degradation process) is depicted by the NARX neural network. Moreover, the correlation analysis of the reversible voltage time series is carried out to obtain the number of delays of the NARX neural network based on the autocorrelation function and the partial autocorrelation function. Then, the total degradation prediction is the sum of the overall degradation prediction and the detail degradation prediction. Finally, the prognostic capability of the proposed method is verified by two aging datasets, and the results show the effectiveness and superiority of the proposed method which can provide accurate degradation forecasting and remaining useful life.