Machine learning and computational modeling informed cold-start design and optimization for proton exchange membrane fuel cells with cathode catalytic H₂-O₂ reaction heating

Abstract

The start-up performance of proton exchange membrane fuel cells in low-temperature environments directly affects their service life and market promotion prospects. However, it is still challenging to fully understand how different operating parameters synergistically intensify the cold startup efficiency of proton exchange membrane fuel cells. In this study, the cold-start performance of proton exchange membrane fuel cells is optimized via cathode catalytic H₂-O₂ reaction heating, integrated with machine learning for key indicator prediction and multi-objective optimization for operating parameter screening. The proposed strategy achieves a temperature rise exceeding 30 °C without external load at –20 °C, suppressing the peak ice volume fraction in the cathode catalyst layer to 3.28 vol % and ensuring post-start stability. Machine learning models can predict key cold-start indicators with high precision. SHapley Additive exPlanations analysis further reveals the complex nonlinear interactions between parameters and clarifies the key factors affecting cold-start performance. Non-dominated Sorting Genetic Algorithm-II optimization identifies Pareto-optimal solutions, demonstrating enhanced cold-start efficiency via synergistic regulation of reactant supply, temperature elevation, controlled anode back pressure, and coolant flow. These findings provide guidance for the engineering design and parameter regulation of proton exchange membrane fuel cells in cold-climate applications.

Graphical abstract

Keywords

Proton exchange membrane fuel cell / cold-start performance / machine learning / multi-objective optimization / SHapley Additive exPlanations

References

[1] Zhang J , Islam M S , Jambulingam M , Lim W M , Kumar S . Leveraging environmental corporate social responsibility to promote green purchases: the case of new energy vehicles in the era of sustainable development. Journal of Cleaner Production, 2024, 434: 139988

[2] Halder P , Babaie M , Salek F , Shah K , Stevanovic S , Bodisco T A , Zare A . Performance, emissions, and economic analyses of hydrogen fuel cell vehicles. Renewable & Sustainable Energy Reviews, 2024, 199: 114543

[3] Skarlis S , Molos T , Ayfantopoulou G , Nikiforiadis A , Bakouros L . Light commercial electric vehicles with hydrogen fuel-cell range extender: refueling strategy evaluation. Research in Transportation Business & Management, 2023, 50: 101040

[4] Tabe Y , Saito M , Fukui K , Chikahisa T . Cold start characteristics and freezing mechanism dependence on start-up temperature in a polymer electrolyte membrane fuel cell. Journal of Power Sources, 2012, 208: 366–373

[5] Higuchi Y , Yoshimune W , Kato S , Hibi S , Setoyama D , Isegawa K , Matsumoto Y , Hayashida H , Nozaki H , Harada M . et al. Experimental visualization of water/ice phase distribution at cold start for practical-sized polymer electrolyte fuel cells. Communications Engineer, 2024, 3(1): 33

[6] Baz F B , Elzohary R M , Osman S , Marzouk S A , Ahmed M . A review of water management methods in proton exchange membrane fuel cells. Energy Conversion and Management, 2024, 302: 118150

[7] Calili-Cankir F , Can E M , Ingham D B , Hughes K J , Ma L , Pourkashanian M , Lyth S M , Ismail M S . Patterned hydrophobic gas diffusion layers for enhanced water management in polymer electrolyte fuel cells. Chemical Engineering Journal, 2024, 484: 149711

[8] Ishikawa Y , Morita T , Nakata K , Yoshida K , Shiozawa M . Behavior of water below the freezing point in PEFCs. Journal of Power Sources, 2007, 163(2): 708–712

[9] Liang J , Fan L , Du Q , Yin Y , Jiao K . Ice formation during PEM fuel cell cold start: acceptable or not. Advanced Science, 2023, 10(24): 2302151

[10] Huo S , Jiao K , Park J W . On the water transport behavior and phase transition mechanisms in cold start operation of PEM fuel cell. Applied Energy, 2019, 233: 776–788

[11] Ivanova N , Baranov I , Kalinnikov A , Mensharapov R , Spasov D , Sinyakov M , Nikolaev I I , Ostrovsky S V , Grigoriev S A , Fateev V N . Cold start of proton exchange membrane fuel cell using build-in catalytic heater. Journal of Physics: Conference Series, 2024, 2723(1): 012004

[12] Pistono A O , Rice C A . Subzero water distribution in proton exchange membrane fuel cells: effects of preconditioning method. International Journal of Hydrogen Energy, 2019, 44(39): 22098–22109

[13] Amamou A , Boulon L , Kelouwani S . An adaptive cold start strategy of proton exchange membrane fuel cell based on maximum power mode. In: Proceedings of IEEE Vehicle Power and Propulsion Conference (VPPC). New York: IEEE, 2019,

[14] Amamou A A , Kelouwani S , Boulon L , Agbossou K . A comprehensive review of solutions and strategies for cold start of automotive proton exchange membrane fuel cells. IEEE Access: Practical Innovations, Open Solutions, 2016, 4: 4989–5002

[15] Zang L , Zhu X , Hao L . Study of three factors for commercial-size PEM fuel cell stack cold start: manifold, endplate, and assisted heating. Fuel, 2024, 373: 132335

[16] Yang Z , Jiao K , Wu K , Shi W , Jiang S , Zhang L , Du Q . Numerical investigations of assisted heating cold start strategies for proton exchange membrane fuel cell systems. Energy, 2021, 222: 119910

[17] Lin S , Su J , Shi L . Cold start simulation study at the key material level of fuel cell. Ionics, 2025, 31(7): 7105–7123

[18] Ge Z , Xu S , Fu X , Zhao Z . Improving the cold-start performance of proton exchange membrane fuel cells via precision engineering of key materials. Precision Chemistry, 2025, 3(4): 172–186

[19] Kim J , Yu J , Lee S , Tahmasebi A , Jeon C H , Lucas J . Advances in catalytic hydrogen combustion research: catalysts, mechanism, kinetics, and reactor designs. International Journal of Hydrogen Energy, 2021, 46(80): 40073–40104

[20] Luo Y , Jiao K . Cold start of proton exchange membrane fuel cell. Progress in Energy and Combustion Science, 2018, 64: 29–61

[21] Luo Y , Jia B , Jiao K , Du Q , Yin Y , Wang H , Xuan J . Catalytic hydrogen–oxygen reaction in anode and cathode for cold start of proton exchange membrane fuel cell. International Journal of Hydrogen Energy, 2015, 40(32): 10293–10307

[22] Guo Q , Luo Y , Jiao K . Modeling of assisted cold start processes with anode catalytic hydrogen-oxygen reaction in proton exchange membrane fuel cell. International Journal of Hydrogen Energy, 2013, 38(2): 1004–1015

[23] Sun S , Yu H , Hou J , Shao Z , Yi B , Ming P , Hou Z . Catalytic hydrogen/oxygen reaction assisted the proton exchange membrane fuel cell (PEMFC) startup at subzero temperature. Journal of Power Sources, 2008, 177(1): 137–141

[24] Guo H , Sun S , Yu H , Lu L , Xu H , Shao Z . Proton exchange membrane fuel cell subzero start-up with hydrogen catalytic reaction assistance. Journal of Power Sources, 2019, 429: 180–187

[25] Sun J , Yang X , Sun S , Shao Z . Investigation on the temperature uniformity and efficiency of cold start-up for proton exchange membrane fuel cell stack based on catalytic hydrogen/oxygen method. Journal of Power Sources, 2021, 496: 229755

[26] Shi L , Du C , Liu Z , Yi Y , Li R , Tang X , Su J , Qian L , Ma T . Identification of the cold start boundaries of proton exchange membrane fuel cells based on one dimensional multi-phase model. Renewable Energy, 2025, 240: 122222

[27] Wang J , Shan Z , Wang K . Sensitivity analysis of the proton exchange membrane fuel cell cold start process with the numerical method. Ionics, 2025, 31(8): 8193–8209

[28] Liu G , Lv S , Chen L , Tao W Q . Optimization of self-cold start for PEM fuel cell with serpentine flow field based on three-dimensional transient model: synergy of initial conditions and start-up strategies. Renewable Energy, 2026, 256: 124165

[29] Zhina W , Xu L , Jianqiu L , Ouyang M . Experimental study on cold start strategies of fuel cell stack: sensitivity analysis, optimization, and boundaries. Journal of the Electrochemical Society, 2024, 171: 6

[30] Liu Z , Mao X , Kang Z , Huang Y , Zhang Z , Xuan D . Multi-objective optimization of ice distribution uniformity and startup failure time for cold start of PEMFC catalyst layer based on CTnet and PSO-GA. Fuel, 2024, 363: 130922

[31] Jian Q , Zhao J . Experimental study on spatiotemporal distribution and variation characteristics of temperature in an open cathode proton exchange membrane fuel cell stack. International Journal of Hydrogen Energy, 2019, 44(49): 27079–27093

[32] Jin L , Wang X J , Zhu J W , Wang C F , Zhou T T , Zhang X W . Sensitivity analysis of proton exchange membrane fuel cell performance to operating parameters and its applicability assessment under different conditions. Energy Conversion and Management, 2021, 228: 113727

[33] Guo W , Wang R , Qiu Y , Wu L , Li K , Tang H . Whale algorithm optimized anode pressure controller for fuel cell systems in ejector recirculation mode. Energy and AI, 2025, 22: 100611

[34] Ming W , Sun P , Zhang Z , Qiu W , Du J , Li X , Zhang Y , Zhang G , Liu K , Wang Y . et al. A systematic review of machine learning methods applied to fuel cells in performance evaluation, durability prediction, and application monitoring. International Journal of Hydrogen Energy, 2023, 48(13): 5197–5228

[35] Xu Y , Li X , Cao Y , Liu K , Ming W . Review: application of machine learning methods to performance evaluation, lifetime prediction, and integrated management of hydrogen fuel cells. Environment, Development, and Sustainability, 2025,

[36] Soares R C , Silva J C , de Lucena J A , Filho A C L , de Souza Ramos J G G , Brito A V . Integration of Bayesian optimization into hyperparameter tuning of the particle swarm optimization algorithm to enhance neural networks in bearing failure classification. Measurement, 2025, 242: 115829

[37] Rastgou M , He Y , Lou R , Jiang Q . A comparison of metaheuristic optimizations with automated hyperparameter tuning methods in support vector machines algorithm for predicting soil water characteristic curve. Engineering Geology, 2025, 353: 108121

[38] Deb K , Pratap A , Agarwal S , Meyarivan T . A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 2002, 6(2): 182–197

[39] Zhao J , Wei W , Zhao L , Xie N , Yang T , Jiang K . Multi-objective optimization of PEMFC performance at different altitudes through integrated RSM, NSGA-II, EWM, and TOPSIS methodology. International Journal of Hydrogen Energy, 2025, 177: 151615

[40] Soon K L , Soon L T . A hybrid quantum neural network and classical gated recurrent unit for battery state of health forecasting incorporating SHAP analysis. Journal of Energy Storage, 2025, 136: 118596

[41] Zhang Y , Li J , Yu M , Chen X , Chen X , Shen J . Dominant factor identification and fast optimization of carnot battery by integrating SHAP and physics-guided neural network. Applied Energy, 2025, 401: 126641

[42] Zuo J , Lv H , Zhou D , Xue Q , Jin L , Zhou W , Yang D , Zhang C . Deep learning based prognostic framework towards proton exchange membrane fuel cell for automotive application. Applied Energy, 2021, 281: 115937

[43] Zhan Z , Yuan C , Hu Z , Wang H , Sui P C , Djilali N , Pan M . Experimental study on different preheating methods for the cold-start of PEMFC stacks. Energy, 2018, 162: 1029–1040

[44] Wang Q , Li B , Yang D , Dai H , Zheng J P , Ming P , Zhang C . Research progress of heat transfer inside proton exchange membrane fuel cells. Journal of Power Sources, 2021, 492: 229613

[45] Zhang W , Tao X , Li Q , Sun K , Chen R , Che Z , Wang T . Effect of startup modes on cold start performance of PEM fuel cells with different cathode flow fields. International Journal of Heat and Mass Transfer, 2024, 225: 125418

RIGHTS & PERMISSIONS

Higher Education Press