Feasibility Assessment of Fast Numerical Simulations for Real-Time Monitoring and Control of PEM Fuel Cells

Computational models that ensure accurate and fast responses to the variations in operating conditions,such as the cell tem-perature and relative humidity(RH),are essential monitoring tools for the real-time control of proton exchange membrane(PEM)fuel cells.To this end,fast cell-area-averaged numerical simulations are developed and verifi ed against the present experiments under various RH levels.The present simulations and measurements are found to agree well based on the cell voltage(polarization curve)and power density under variable RH conditions(RH=40%,RH=70%,and RH=100%),which verifi es the model accuracy in predicting PEM fuel cell performance.In addition,computationally feasible reduced-order models are found to deliver a fast output dataset to evaluate the charge/heat/mass transfer phenomena as well as water production and two-phase fl ow transport.Such fast and accurate evaluations of the overall fuel cell operation can be used to inform the real-time control systems that allow for the improved optimization of PEM fuel cell performance.

Modelling Anti-Corrosion Coating Performance of Metallic Bipolar Plates for PEM Fuel Cells:A Machine Learning Approach

Proton exchange membrane(PEM)fuel cells have significant potential for clean power generation,yet challenges remain in enhancing their performance,durability,and cost-effectiveness,particularly concerning metallic bipolar plates,which are pivotal for lightweight compact fuel cell stacks.Protective coatings are commonly employed to combat metallic bipolar plate corrosion and enhance water management within stacks.Conventional methods for predicting coating performance in terms of corrosion resistance involve complex physical-electrochemical modelling and extensive experimentation,with significant time and cost.In this study machine learning techniques are employed to model metallic bipolar plate coating performance,diamond-like-carbon coatings of varying thicknesses deposited on SS316L are considered,and coating performance is evaluated using potentiodynamic polarization and electrochemical impedance spectroscopy.The obtained experimental data is split into two datasets for machine learning modelling:one predicting corrosion current density and another predicting impedance parameters.Machine learning models,including extreme gradient boosting(XGB)and artificial neural networks(ANN),are developed,and optimized to predict coating performance attributes.Data preprocessing and hyperparameter tuning are carried out to enhance model accuracy.Results show that ANN outperforms XGB in predicting corrosion current density,achieving an R2>0.98,and accurately predicting impedance parameters with an R2>0.99,indicating that the models developed are very promising for accurate prediction of the corrosion performance of coated metallic bipolar plates for PEM fuel cells.

A facile synthesis of high activity cube-like Pt/carbon composites for fuel cell application

High activity catalyst with simple low-cost synthesis is essential for fuel cell commercialization. In this study, a facile procedure for the synthesis of cube-like Pt nanoparticle （PtCube） composites with high surface area carbon supports is developed by mixing precursor of Pt with carbon supports in organic batches, hence, one pot synthesis. The PtCube grow with Vulcan XC-72 or Ketjen black, respectively, and then treated for 5.5 h at 185℃ （i.e., Ptcube5.5/V and PtcubeS.5/K）. The resulting particle sizes and shapes are similar; however, Ptcube5.5/K has a larger electrochemical active surface area （EASA） and a remarkably better formic acid （FA） oxidation performance. Optimization of the PtCube/K composites leads to Ptcubelo/K that has been treated for 10 h at 185℃. With a larger EASA, Ptcubelo/K is also more active in FA oxidation than the other Ptcube/K composites. Impedance spectroscopy analysis of the temperature treated and asprepared （i.e., untreated） Ptcube/K composites indicates that PtCube10/K is less resistive, and has the highest limiting capacitance among the PtCube/K electrodes. Consistently, the voltammetric EASA is the largest for Ptcube 10/K. Furthermore, PtCube10/K is compared with two commercial Pt/C catalysts, Tanaka Kikinzoku Kogyo （TKK）, and Johnson Matthey （JM）Pt/C catalysts. The TKK Pt/C has a higher EASA than Ptcube10/K, as expected from their relative particles sizes （3-4 nm vs. 6-7 nm for PtcubelO/K）, however, Ptcube10/K has a significantly better FA oxidation activity.