Fuel Cell Model Research Articles

Correlating electrochemical active surface area with humidity and its application in proton exchange membrane fuel cell modeling

The electrochemical active surface area has a vital influence on the performance of proton exchange membrane fuel cells. In this study, a universal semi-empirical equation of the electrochemical active surface area in the catalyst layer of fuel cells is proposed based on self-designed experiments, aiming to improve the predictive accuracy of numerical models under low humidity conditions. The generalization of this equation is also verified with the other nine catalyst layers of different compositions. At very low water content corresponding to low inlet humidification operations, the electrochemical active surface area is generally low and increases sharply with the increment of water content. When the water content exceeds a certain threshold value, the electrochemical active surface area becomes independent of water content. In addition, the electrochemical active surface area significantly decreases as the operating temperature increases. Consequently, including this newly proposed electrochemical active surface area equation in conventional full-cell models is of significant importance to accurately predict the activation loss and performance of fuel cells under both cold-start and normal operation conditions, especially at low humidification, which is validated against experimental data. In contrast with the conventional constant electrochemical active surface area assumption, this equation improves the model accuracy from about 70% to be higher than 90% under low humidity conditions.

Advanced parametric model for analysis of the influence of channel cross section dimensions and clamping pressure on current density distribution in PEMFC

The design of the flow field and the channel cross section of a bipolar plate as well as the intrusion effect of gas diffusion layers due to clamping pressure can significantly influence the performance of a fuel cell. The land and channel areas at a channel cross section can have different boundary conditions due to compression, intrusion of gas diffusion layers and transport phenomena. Todayapplication engineers are faced with the challenge of finding advantageous overall solutions from this plurality of multi-criteria parameters. In this thesis, a model approach is presented which combines a fully parametric analytical-empirical intrusion model with a 1D fuel cell model. This allows the calculation of current density distributions at the channel cross section, taking into account clamping pressure, channel parameters and operating conditions. Depending on the operating points considered, the results show that the maximum current densities occur in different areas of the channel cross section. In addition, with regard to the clamping pressure current density optima can be determined. Using this model approach, in an early development process application engineers will be able to compare hundreds of variants in a reasonable short time and thus determine advantageous fuel cell designs.

Damage Modeling of Solid Oxide Fuel Cells Accounting for Redox Effects

This work applies a multiscale mechanistic damage model developed for brittle ceramics and implemented in commercial finite element (FE) packages via user subroutines to study progressive damage in solid oxide fuel cells (SOFC) subjected to thermomechanical loading under normal operating and shutdown conditions including redox effects. The damage model captures the micromechanics of stiffness reduction due to material porosity change and microcracking and integrates the as-obtained stiffness reduction law into a continuum damage mechanics (CDM) formulation for the evolution of microcracks up to fracture. The volumetric “swelling” that occurs during redox is treated in constitutive modeling similarly to thermal expansion, but swelling strains are irreversible. This damage model was first validated through predictions of strength and stress-strain response for the SOFC electrode materials. Next, it has been applied to predict the potential for degradation in a generic planar SOFC stack with large active area cells. Multicell stack models were simulated in both co-flow and counter-flow configurations. In addition, a constant temperature redox cycle was also simulated to capture overall cell electrode damage due to volumetric swelling of the nickel (Ni)-based anode in the anode-supported cells.

Optimal Values of Unknown Parameters of Polymer Electrolyte Membrane Fuel Cells Using Improved Chaotic Electromagnetic Field Optimization

In this article, the improved chaotic electromagnetic field optimization (ICEFO) algorithm is adopted to generate the optimal values of unknown parameters of fuel cells (FCs). The mathematical model of polymer electrolyte membrane FC (PEMFC) considers nonlinear and complex optimization problems with different control variables. The sum of squared error between the measured and computed stack voltages is considered as the main objective function. The performance of the ICEFO algorithm is tested on three PEMFC stacks. Moreover, sensitivity and statistical measures are presented to confirm the reliability and accuracy of ICEFO. In addition, the effect of changing the cell temperature and reactants pressures is studied for more validation of ICEFO. Furthermore, the results obtained by ICEFO are competitive compared with other optimization methods. These results confirm the effectiveness of ICEFO in solving the optimization problem of PEMFC parameter estimation. Finally, based on the optimization results a Simulink model for PEMFC is developed to examine the dynamic performance of the PEMFCs.

Vehicle-to-grid application with hydrogen-based tram

Hydrogen is acquiring a central role in energy uses, aiming to support and, then, replace conventional sources. Innovative smart-grids are based on new concepts of energy production and distribution, focus of research studies and preliminary prototypes. In this field, fuel cell-based vehicle-to-grid applications can be a useful system as a back-up power device during energy oscillations and as a transportation mode, especially in public service. In the present paper, a fuel cell hybrid tram is numerically tested, modelling dynamically each main component, with the aim to operate as a power and heat supplier, during the night, and as an urban light rail vehicle, during the day. The fuel cell model is developed, highlighting the principal elements of stack, auxiliary and heat-water management sub-systems. Performance improvements, such as regenerative braking and cogenerative processes, are taken into account, conceptualizing an innovative cooling method. In addition, a 3D CAD model is built to define a suitable layout for the hybrid powertrain, following safety guidelines. In the simulation campaigns, encouraging results are achieved for the case study analysed: for daily operations, the fuel cell-based powertrain provides approximately 2 MWh, consuming 108 kg of hydrogen (48 kg in vehicle-to-grid mode and 60 kg in mobility mode) and reaching an overall efficiency of approximately 43%. In addition, during vehicle-to-grid mode, the tram supplies thermal power capable of warming up at 45 °C almost 0.2 kg/s of tap water on average, obtaining a cogenerative efficiency that exceeds 60%. The promising performance achieved could represent the first steps of the proof of feasibility for this innovative technology.

Mathematical Modeling of an Electrotechnical Complex of a Power Unit Based on Hydrogen Fuel Cells for Unmanned Aerial Vehicles

The issues of mathematical and numerical simulation of an electrical complex of a power plant based on hydrogen fuel cells with a voltage step-down converter were considered. The work was aimed at developing a mathematical model that would provide for determining the most loaded operation mode of the complex components. The existing mathematical models do not consider the effect of such processes as the charge and discharge of the battery backup power supply on the power plant components. They often do not consider the nonlinearity of the fuel cell output voltage. This paper offers a mathematical model of an electrical complex based on the circuit analysis. The model combines a well-known physical model of a fuel cell based on a potential difference and a model of a step-down converter with a battery backup power supply developed by the authors. A method of configuring a fuel cell model based on the experimental current–voltage characteristic by the least-squares method has been proposed. The developed model provides for determining currents and voltages in all components of the power plant both in the nominal operating mode and in the mode of limiting the power consumed from the fuel cell when the battery backup power supply is being charged. The correctness of the calculated ratios and the mathematical model has been confirmed experimentally. Using the proposed model, a 1300 W power plant with a specific power of 529.3 W∙h/kg was developed and tested.

Influence of Sherwood Number at Channel-Gdl Interface on the Performance of PEM Fuel Cell

In this paper, an enhanced isothermal, two-phase one-dimensional semi-analytical PEM fuel cell model is developed. Anode catalyst layer is simulated analytically and cathode catalyst is simulated numerically with agglomerate method. A “three-step” validation procedure is performed to achieve good agreement with experimental data. Single-phase flow and two-phase flow patterns in bioplate channel are considered with corresponding Sherwood number correlations. The results indicate that oxygen concentration drop at Channel-GDL interface decreases with the increase of Sherwood number. Thus, considering the Sherwood number as a constant is not appropriate in fuel cell model. The Sherwood number of film flow is smaller than that of slug flow, which represent higher mass transfer resistances under water flooding condition. The present model can reflect the main two-phase transport process in PEM fuel cell with low computational cost and can be used in primary design and control of fuel cell.

Coupling Experiment and Modelling to Characterize and Predict Performances of PEMFC

In the past few decades, many improvements have been achieved regarding the PEMFCs components, the cell/stacks design as well as the architecture of the overall system. Hence, the level of maturity of PEMFC systems recently became sufficient to enable their commercialization for automotive applications (FCEV Miraï by Toyota, FCEV Nexo by Hyundai, etc.). However, despite great initial performance, PEMFC systems still suffer technological limitations, such as their initial cost and overall durability in real life operation, which prevents widespread industrial deployment. Some of these limitations can be overcome by combining experiments and modelling to characterize and better understand the behaviour of the carbon-supported platinum (Pt/C) electrocatalyst, its utilization/effectiveness in PEMFC catalyst layers and to be able to predict the performance and durability of PEMFC. This should help making the most relevant choices to accelerate the development processes of PEMFC.In this work, physico-chemical and electrochemical measurements are performed from the scale of the raw Pt/C materials up to the complete catalyst layer, to gather as much information as possible on the catalytic layer micro-structure and its operating properties. Coupling experiments in classic RDE and in PEMFC differential-single cell (1.8 cm² under high reactant stoichiometry) enables to fully characterize the electrochemical properties and behaviour of our materials. These different measurements are performed under various ideal and well-controlled operating conditions: cell temperature (T), relative humidity (RH), partial pressure of O2 (PO2 ) were chosen to be relevant and to dissociate and characterize as much as possible the different electro-chemical mechanisms involved in the catalyst layers of PEMFC. In parallel; several models have been developed to simulate one or combined mechanisms leading to the performance [1] and degradation of cell components[2], [3]. In fact, models aim at isolating and quantifying the contribution of each mechanism by choosing relevant operating conditions. Based on our experimental work and data sets, the behaviour of the Pt/C electrocatalysts has been studied in order to introduce new electrocatalytic features and especially the Pt surface oxide formation and reduction which is linked to Pt surface state. This preliminary step was then further developed into, a complete performance model for the O2 reduction reaction at the cathode to better describe the physical and electrochemical phenomenon involved in 0D and 1D catalyst layers during fuel cell operation. In this new description, the position of the main adsorption/desorption peaks are well captured by the simulation compared to experiment keeping complete physical meaning.This work has partially received funding from the Fuel Cells and Hydrogen 2 Joint Undertaking under grant agreement No. 875025 (FURTHER-FC project). This Joint Undertaking receives support from the European Union’s Horizon 2020 Research and Innovation programme, Hydrogen Europe and Hydrogen Europe Research.[1] B. Randrianarizafy, P. Schott, M. Chandesris, M. Gerard, and Y. Bultel, “Design optimization of rib/channel patterns in a PEMFC through performance heterogeneities modelling,” International Journal of Hydrogen Energy, vol. 43, no. 18, pp. 8907–8926, May 2018, doi: 10.1016/j.ijhydene.2018.03.036.[2] T. Jahnke, G. A. Futter, A. Baricci, C. Rabissi, and A. Casalegno, “Physical Modeling of Catalyst Degradation in Low Temperature Fuel Cells: Platinum Oxidation, Dissolution, Particle Growth and Platinum Band Formation,” J. Electrochem. Soc., vol. 167, no. 1, p. 013523, Nov. 2019, doi: 10.1149/2.0232001JES.[3] G. Maranzana, A. Lamibrac, J. Dillet, S. Abbou, S. Didierjean, and O. Lottin, “Startup (and Shutdown) Model for Polymer Electrolyte Membrane Fuel Cells,” J. Electrochem. Soc., vol. 162, no. 7, pp. F694–F706, 2015, doi: 10.1149/2.0451507jes. Figure 1

Numerical Investigation of Tapered Flow Field Configurations for Enhanced Polymer Electrolyte Membrane Fuel Cell Performance

A tapered flow field configuration (FFC) is proposed to improve the oxygen transport, water removal and performance of polymer electrolyte membrane fuel cells. By performing a three-dimensional multiphase fuel cell model, the influence of the tapered FFC on the internal physicochemical process and overall cell performance are numerically investigated in this study. The tapered FFC without and with consideration of the electric contact resistance (ECR) between bipolar plates and gas diffusion layers are comparatively evaluated. Compared with the conventional FFC, for the tapered FFC without considering the ECR, an increased ratio of the side length for the inlet to that for the outlet (LI/O) enhances oxygen transport, water removal and cell performance. However, for tapered FFCs considering the ECR, an increased ratio of LI/O initially increases and then weakens oxygen transport, water removal and cell performance. Nevertheless, regardless of the tapered FFC with or without considering the ECR, a decreased ratio of LI/O degrades the overall cell performance. In comparison with all tapered FFC designs, the optimal tapered FFC design with an LI/O of 1.2 exhibits a more uniform reactant and current density distributions, which reduces the coefficient of variation of the current density and oxygen molar concentration by approximately 21.4% and 8.5%, thereby improving overall cell performance.

Modeling of Fuel Cell Stack for High-Speed Computation and Implementation to Integrated System Model

:1-dimentional (1D) modeling methods of fuel cell stack are investigated. To ensure numerical simulation of life-long system operation in permissible calculation time and accuracy, physical/empirical models of mass transfer, heat transfer, and electrochemistry in fuel cell stack with proper resolution are developed. These models are validated and verified by the actual fuel cell system data collected in a variety of operating conditions, such as low to high loads, operating temperatures, and total pressures. Modeling and implementation: For a high-speed computation, 2D (along-the-channel + across-the-channel distribution) model was reduced to 1D (across-the-channel distribution) model by introducing weighting function for state variables at inlet and outlet of the fuel cell stack (flowrate, pressure, temperature, and molar fractions of gaseous species), as shown in Fig.1.The physics implemented in the fuel cell stack model are listed in Table 1. Of the models listed in the table, the short-circuit current model, effective enthalpy voltage model, and direct-combustion reaction model were newly developed and implemented. The other models were employed from the representative literature [1][2][3][4]. The model parameters were calibrated by the microscopic measurement results of geometrical structures as well as electrochemical test results measured at a variation of current, gas composition, humidity, and temperature using small cells [5][6].For reducing the computational load, the models were integrated based on a discrete-time explicit numerical solver without any iterative calculations for numerical convergence. To achieve high accuracy even in such a simple framework of the numerical solving methods, the features of mass transport dynamics were expressed by discrete first-order lag model with the unique time-constants, which were determined empirically. All the models were implemented on MATLAB/SIMULINK platform and the interface of the fuel cell stack model was designed so that it could be easily integrated with the balance-of-plants models (air-supply, hydrogen-supply, and cooling subsystems) and the existing libraries of the fuel cell system controllers. Results and discussion Extensive testing was conducted and a lot of validation data were collected in the variety of operating conditions such as low to high loads, operating temperatures, and total pressures. The validation data were utilized for the validation of the fuel cell stack model accuracy and modeling strategies.An example of the validation results is shown in Fig.2. Time-series data of the interfacial conditions of the fuel cell stack (current, temperature, flowrate, pressure, and gas composition) are inputted to the fuel cell stack model, and voltage and resistance are calculated by the model. Then, calculated and measured values are compared. Newly developed models (the short-circuit current model, effective enthalpy voltage model, and direct-combustion reaction model) were found to be beneficial to close a gap of the acceleration / deceleration transient features between dynamic fuel stack responses in the validation data and the calculation results.The computational time of the numerical simulation is listed in Tables 2 and 3. A considerable reduction in computation time (more than 50-times faster than real time) was confirmed even under the standard-performance computational resources. References R.B. Bird, W.E. Stewart, E.N. Lightfoot, Transport Phenomena, Revised Second Edition, John Wiley & Sons, New York (2006).S.V. Patankar, Numerical Heat Transfer and Fluid Flow, McGraw–Hill, New York (1980).T.E. Springer, T.A. Zawodzinski, and S. Gottesfeld, “Polymer electrolyte fuel cell model”, J. Electrochem. Soc., 138(8), 2334–2342 (1991).D.M. Bernardi and M.W. Verbrugge, “A mathematical model of the solid-polymer-electrolyte fuel cell,” J. Electrochem. Soc., 139(9), 2477–2491 (1992).T. Tsukamoto et al., “Three-dimensional numerical simulation of full-scale proton exchange membrane fuel cells at high current densities,” J. Pow. Sour., 588, 229412 (2021).N. Nonoyama et al., “Analysis of oxygen-transport diffusion resistance in proton-exchange-membrane fuel cells,” J. Electrochem. Soc., 158(4), B416–B423 (2011). Figure 1

Parameter Identification of a Quasi-3D PEM Fuel Cell Model by Numerical Optimization

Polymer electrolyte membrane fuel cells (PEMFCs) supplied with green hydrogen from renewable sources are a promising technology for carbon dioxide-free energy conversion. Many mathematical models to describe and understand the internal processes have been developed to design more powerful and efficient PEMFCs. Parameterizing such models is challenging, but indispensable to predict the species transport and electrochemical conversion accurately. Many material parameters are unknown, or the measurement methods required to determine their values are expensive, time-consuming, and destructive. This work shows the parameterization of a quasi-3D PEMFC model using measurements from a stack test stand and numerical optimization algorithms. Differential evolution and the Nelder–Mead simplex algorithm were used to optimize eight material parameters of the membrane, cathode catalyst layer (CCL), and gas diffusion layer (GDL). Measurements with different operating temperatures and gas inlet pressures were available for optimization and validation. Due to the low operating temperature of the stack, special attention was paid to the temperature dependent terms in the governing equations. Simulations with optimized parameters predicted the steady-state and transient behavior of the stack well. Therefore, valuable data for the characterization of the membrane, the CCL and GDL was created that can be used for more detailed CFD simulations in the future.

The prediction of the polarization curves of a solid oxide fuel cell anode with an artificial neural network supported numerical simulation

The presented study focuses on a numerical simulation of the transport phenomena inside a solid oxide fuel cell anode. The classical mathematical model leads to a notable discrepancy between measured and predicted overpotentials. One of the possible reasons is the assumption of the constant values of electrochemical reaction charge transfer coefficients. A modified formulation of the problem includes data-driven correction of reaction charge transfer coefficients in the electrochemical reaction model. Here we show a dedicated computational scheme in which an artificial neural network updates the charge transfer coefficients depending on the operational conditions and the available datasets. The neural network was trained on twelve experimental data points of an anode's polarization curve obtained from the literature. The training set contained data for the anode operating in two different temperatures - 800 °C and 900 °C. The test set contained additional six data points for an anode operating at 1000 °C. Charge transfer coefficients were proposed by the Artificial Neural Network as a functional relation of the temperature and withdrawn current. The results of the predictions are juxtaposed with the experimental data from the literature. It was shown that an Artificial Neural Network could improve an electrochemical reaction model in Solid Oxide Fuel Cell modeling.

Solid Oxide Fuel Cell Modeling Using Numerical Method and Neural Network. (Dept. E)

Solid Oxide Fuel Cell Modeling Using Numerical Method and Neural Network. (Dept. E)

Optimal parameter estimation strategy of PEM fuel cell using gradient-based optimizer

The exact parameter estimation of the fuel cell model is considered a critical stage in delivering a consistent emulation for the fuel cell system characteristics. The aim of this is to suggeste a robust methodology based on the Gradient-based Optimizer (GBO) to identify the best parameters of PEM fuel cell (PEMFC). Three distinct types of PEM fuel cells: 250 W FC stack, BCS 500 W, and SR-12 500 W, were used to demonstrate the superiority of the GBO. To confirm the superiority of GBO, the results were compared with those obtained using different optimizers such as salp swarm algorithm (SSA), heap-based optimizer (HBO), differential evolution (DE), whale optimization algorithm (WOA), moth-flame optimization algorithm (MFO), sine cosine algorithm (SCA), and Harris's hawk optimizer (HHO). During the optimization process, the unknown parameters of PEM fuel cells are used as decision variables, whereas the objective function needs to be minimum is represented by the sum square error between the measured data and estimated data. In addition, the obtained results by GBO are compared with other methods achieved in the literature. The superiority of GBO in determining the optimal parameters of different PEM fuel cells is proved.

A fast fuel cell parametric identification approach based on machine learning inverse models

In this paper a computationally efficient optimization approach to the parametric identification of a fuel cell equivalent circuit model is presented. It is based on the inverse model and on machine learning regressions. During the training phase, the inverse model is built numerically by means of advanced regression approaches, i.e., the support vector machine regression, the least squares support vector machine regression and the Gaussian process regression. The training set is synthetically generated to the aim of exploring the parameter space and to characterize different stack operating conditions, including normal and faulty ones. The accuracy of the considered approaches is investigated by employing a test set including many experimental data, consisting of impedance spectra measured through the electrochemical impedance spectroscopy and referring to very different stack operating conditions. The results show that all the considered machine learning methods allow to identify the parameters of the fuel cell model with a low computational burden, so that they fit with the hardware resources of low cost embedded processors. This feature allows to envisage that the proposed approaches are good candidates for a model-based on-line diagnosis of fuel cell stacks.

Application of the improved chaotic grey wolf optimization algorithm as a novel and efficient method for parameter estimation of solid oxide fuel cells model

In this paper, important functional parameters of solid oxide fuel cells are identified by introducing a novel high-speed optimization method, namely adaptive chaotic grey wolf optimization algorithm. The suggested optimization method is obtained by combining the adaptive grey wolf optimization and chaotic grey wolf optimization algorithms. The chaotic algorithm is applied to the basic grey wolf optimization to achieve higher convergence speed, keep the population's diversity, and provide an initial population with uniform distribution. Besides, a nonlinear convergence factor is defined for balancing the global and local exploration abilities. Employing the improved convergence factor resulted in a new version of the grey wolf optimization algorithm, namely adaptive grey wolf optimization algorithm. Adaptive chaotic grey wolf optimization algorithm adopts the advantages of both chaotic grey wolf optimization and adaptive grey wolf optimization methods simultaneously. The adaptive grey wolf optimization algorithm is applied to a 5 kW dynamic tubular stack. The results of the simulation report the lowest values of mean squared error, higher accuracy, higher robustness, and high convergence speed for the adaptive grey wolf optimization algorithm compared to some well-known optimization methods. Besides, the proposed method shows a good agreement with experimental results with lower computational difficulty.

Model parameters estimation of a proton exchange membrane fuel cell using improved version of Archimedes optimization algorithm

Proton Exchange Membrane (PEM) fuel cell is one of the most popular fuel cells because of its higher efficiency among the other fuel cells. Because of the expensive materials in designing of this type of fuel cell, it should be first design and simulated in the best and optimum way to reduce the construction costs as much as possible. In the present study, a new model identification is proposed for optimal parameters identification of the PEM fuel cells. The major idea in this study is to provide a new optimal methodology to parameters estimation of the unknown variables in the PEM fuel cell model so that the absolute error (IAE) between the estimated data based on the proposed model and the real data has been minimized. The proposed method uses a new improved design of Archimedes Optimization Algorithm (IAOA) to this purpose. The designed model is then implemented on two practical case studies and the results are compared with some well-known methods. Final results shows that the proposed method with 0.10 and 0.14 error values for Nexa and NedStack PS6 models, respectively, provides the best solution among the other comparative methods.

Investigation of a cost-effective strategy for polymer electrolyte membrane fuel cells: High power density operation

Polymer electrolyte membrane (PEM) fuel cell technology needs to overcome the cost barrier in order to compete with the internal combustion engines (ICEs) for transportation application. A viable approach is to raise fuel cell's power output without increasing its size and Pt loading in the catalyst layers (CLs). In this strategy, the cost per kW power output can be proportionally reduced due to the increased power density. This paper examines this strategy by exploring several important aspects that influence fuel cell performance under high power or current density using a three-dimensional (3-D) fuel cell model. It is shown that local CLs may be subject to low oxygen concentration under a high current density of 2 A/cm2, causing low reaction rate near the outlet, especially under the land. Additionally, the oxygen reduction reaction (ORR) rate may be subject to a large through-plane variation under 2 A/cm2, raising ohmic voltage loss in the CL. Two additional cases are investigated to improve fuel cell performance under 2 A/cm2: one has a 5 times thinner CL with the same ORR kinetics per membrane electrode assembly (MEA) area and the other has a 5 times thinner CL with 5 times higher ORR kinetics. The results show the output voltage is raised approximately from 0.5 V to 0.554 V in the former CL case and further to 0.606 V for the latter CL. To enable high-efficiency operation (e.g. >50%), thinner CLs with high ORR kinetics and GDLs with better transport properties are one research and development (R&D) direction.

2-D + 1-D PEM fuel cell model for fuel cell system simulations

The water management is critical for the operation of PEM fuel cells and has a strong impact on its performance and durability. The aim of this work is the simulation-based investigation of the operation of a PEM fuel cell system with the special focus on its water management.In order to analyze these dependencies correctly, a 2-D + 1-D PEM fuel cell stack model has been developed, which on the one hand has a high level of modelling details and on the other hand meets high requirements concerning its runtime, to enable acceptable simulation times for fuel cell system simulations. The fuel cell model is integrated into an AVL Cruise-M fuel cell system simulation.An analysis is presented comparing a system operation with a fuel cell in co- and counter-flow configuration with a special focus on the local and overall water management.

Extreme learning machine based meta-heuristic algorithms for parameter extraction of solid oxide fuel cells

A precise, fast, and robust parameter extraction technique of solid oxide fuel cell models is extremely crucial for optimal control and behavior analysis. In this paper, a novel extreme learning machine based method is proposed to extract unknown parameters of solid oxide fuel cell models including electrochemical model and simple electrochemical model. At first, extreme learning machine is applied to overcome two thorny obstacles (e.g., data shortage and noised data) via predicting additional data and updating noised data. Then, both original data collected from a 5 kW solid oxide fuel cell stack and processed data are transferred to effectively guide eight prominent meta-heuristic algorithms for effective parameter extraction. The performance of extreme learning machine is thoroughly investigated in two typical operation conditions through a comprehensive comparison based on various training data. Simulation results validate that the proposed approach can effectively contribute to searching efficient model parameters along with high accuracy, prominent stability, high speed, and great robustness. Particularly, the accuracy of parameter extraction for electrochemical model and simple electrochemical model can be improved by 49.3% and 65.6% at most, respectively.