Multi-carrier optimal power flow of energy hubs by means of ANFIS and SQP

Due to the climate change and the decrease in fossil fuel reserves, the industrial and tertiary sectors have been focused on the implementation of advanced energy management systems in order to improve their energy efficiency and reduce their overall emissions. One way to achieve that goal, which is also the focus of this work, is by optimizing the energy use in their operation processes. This paper presents a hybrid optimization method, combined by neuro-fuzzy inference systems and the quadratic programming optimization method, to calculate the short-term demand forecasting of a multi-carrier energy system and to optimize its energy flow. The objective of the optimization is to fulfill the system's energy demands and minimize a set of established optimization criteria. Moreover, the algorithm considers the system's dynamics and inertias in order to guarantee that the obtained results present a feasible and stable operation strategy for the energetic plant. The method has been applied and validated under real conditions in a car manufacturing plant of Spain, in the framework of a FP7 European research project using online production and consumption data.