Design and validation of a predictive energy management strategy for self-consumption in tertiary buildings

This work presents a predictive energy management strategy for self-consumption in tertiary buildings. The self-consumption is composed of a photovoltaic generation and a battery. The energy management strategy is composed of a forecast module, high-level strategy and real-time adaptative control. Due to the daily forecast, significant data was available 24 hours in advance, allowing the energy management strategy to take advantage. The high-level strategy defines the battery's operation mode for each hour of the day. The real-time adaptative control corrects the possible errors with instant measurements and generates real-time battery commands and its operation mode. With this approach, a reduction of 16.17 % of the electric bill was obtained by comparing it to a scenario without a battery and its correspondent strategy. The development was integrated and validated in a test bench, obtaining a 60.43 % grid independence increase.