Modeling and Control of the Fuel Cell/Gas Turbine hybrid system has been a complex task based on first principles models.   It has been demonstrated that the first principle models are time consuming and no suitable to design and implement.  This has been because the lack on the models to include a lot of phenomena present on the system such as coupling of variables, parameter interaction, nonlinearities, hysteresis, etc.   It is also very complex to model valves and turbomachinery with high accuracy and avoid uncertainties on the output of the models.  System Identification (SI) technique was proposed to identify discrete transfer functions from input-output experimental data of a Solid-Oxide Fuel Cell and Gas Turbine hybrid system, realized through the Hybrid Performance (HyPer) facility at the U.S. Department of Energy, National Energy Technology Laboratory (NETL).    The performance index to measure the fit of the model with the data gathered from the system has been most of the time below to 85%.  That results indicate that uncertainties were present on the models obtained.    This work was focus on analyze the impact of the variability on model parameters on the gains of the PID controllers designed to control the cathode air flow with the hot air valve of the HyPer system.  One factor at the time variability has been the method employed for the parameter sensitivity analysis.  The sensitivity analysis shown in this work will help to strengthening the knowledge base of the models uncertainties and help to improve controller performance, predictability, and implementability.