Modeling method of an integrated fuel cell system: In this study, the actual FC-system in commercial fuel cell electric vehicle (FCEV), 2nd-generation MIRAI [1], is selected as the target of the validation and verification of the models. It consists of the FC-stack and sub-systems of air, hydrogen (H2), and cooling systems as shown in Fig. 1. The concepts of modelling methods of the FC-system are shown in Fig. 2, where the system configuration is depicted as the function-block diagram. In each function block, the entire system is broken-down to the component level and physical models of the individual components of FC-stack [2], air compressor [3], throttle valve [4], intercooler [5], and piping elements [5] are implemented. The state variables such as total pressure, flowrate, temperature, and gas composition are defined only at the centre of each component and the distribution of the state variables inside each component is not considered for high speed computation of the dynamic system behaviour. The differential equations of mass balance, mole balance, and energy balance across the fluid circuit shown in Fig. 2 are solved by finite difference method. The differentiated linear algebraic equation of mass balance is expressed by Eq. 1. The air, H2, and cooling system models are integrated with FC-stack mass transport and electrochemical models to build an integrated FC-system model in the Simulink block diagram. Results and discussion: The integrated FC-system model was validated and verified with the actual data collected with 2nd-generation MIRAI by comparing the simulation outputs and the system response data in same input and boundary conditions and the acceptable accuracy was confirmed. The relationships among a FC-stack material properties of PEM, CL, and GDL, FC-stack performance, and the system performance of the system efficiency, acceleration response, and heat generation rate are investigated. An example where cathode GDL substrate is removed and only MPL is remained is shown in Figs. 3–6. Fig. 3 is the simulation input conditions of target FC-system power and ambient wind velocity in front of radiator, where the steady state and dynamic transient in low to high load are combined and the constant wind velocity is assumed. Fig. 4 is the simulation results of I–V and I–R performance comparison. The removal of the GDL substrate improved oxygen transport and reduced FC-resistance by 10 mΩ and raised FC-voltage by 20 mV at 2.0 A/cm2. The break-down of voltage improvement of 20 mV consists of 5.5 mV reduction of resistance overpotential and 14.5 mV by concentration overpotential as far as MPL can deliver O2 under the ribs of the flow channels. Fig. 5 is the simulation results of the dynamic FC-system responses of overall heat generation rate, FC-coolant outlet temperature, and overall H2 consumption amount during an entire pattern. It is confirmed that the peak FC-coolant outlet temperature is lowered from 85 to 80 °C following the peak heat generation rate is reduced by 16 %, which also contributes to prevent the PEM and CL from drying. H2 consumption was reduced 4.3 %. Conclusions: The modeling method for an integrated FC-system which consists of the FC-stack, air system, H2 system, and cooling system is investigated. This model was validated and verified by the actual FC-system data collected with the FC-system implemented in the commercial FCEV and an acceptable accuracy was confirmed. The impact of oxygen transport resistance of cathode GDL substrate on the FC-system performance was investigated. It demonstrates impacts of cell configuration including choice of components and materials on the system performance can be investigated by numerical simulation. It is beneficial for an entire FC-system development process. Acknowledgement : This work was supported by the FC-Platform Program: Development of design-for-purpose numerical simulators for attaining long life and high performance project (FY 2020–2022) conducted by the New Energy and Industrial Technology Development Organization (NEDO), Japan.