This paper investigates a new approach that incorporates the predicted renewable energy power as well as the predicted load consumption in the power flow management of a stand-alone hybrid renewable energy system. The studied renewable system comprises photovoltaic panels, wind turbine, as primary renewable energy sources and uses as second source a fuel cell which provides electric power from previously stored hydrogen produce by an electrolyzer. Using stochastic models and propagating wind speed, solar irradiance uncertainties through the system, a new fuzzy logic controller is design and validated with a reported experimental data. Unlike the reported power management system, the control system incorporate a one-step prediction of the DC link net power with its uncertainty in order to make an appropriate decision which improves the hydrogen and the fuel cell lifetime. Through a comparative study, its has been demonstrated that integrating power prediction and its uncertainty gives better results than an optimization-based approach proposed in literature. Copyright (C) 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.