Within the last few years, the trend towards more distributed, renewable energy sources has led to major changes and challenges in the electricity sector. To ensure a stable electricity distribution in this changing environment, we propose a robust energy management approach to deal with uncertainty occurring in microgrids. For this, we combine robust optimization with a rolling horizon framework to obtain an algorithm that is both, tractable and can deal with the considered uncertainty. The main contribution of this work lies within the development and testing of a dynamic scheduling tool, which identifies good starting time slots for the rolling horizon. Combining this scheduling tool with the rolling horizon framework results in a dynamic rolling horizon model, which better integrates uncertainty forecasts and realizations of uncertain parameters into the decision-making process. A case study reveals that the dynamic rolling horizon model outperforms the classical version by up to 57% in costs and increases the local use of PV by up to 11%.