Energy hub optimization applied on car manufacturing plants

The challenge of climate change is real and growing. Every car manufacturing plant must look deep into its operations and use every opportunity to play its part in reducing overall emissions. Furthermore, due to the increased competition of the sector, they must turn to solutions that can reduce their costs, but maintaining the products quality on high standards. One way to achieve that goal, which is also the focus of this work, is by reducing the energy use in the plants operation processes. This paper presents a hybrid optimization methodology, combined by neuro-fuzzy inference systems and the quadratic programming algorithm, to calculate the short-term demand forecasting of a plant, and to optimize its energy flow. Moreover, the dynamic models of the plant's equipment are considered into the optimization process, to calculate the dynamic system response and the equipment's inertias. Finally, the algorithm optimizes the operation of the plant with objective to fulfill the energy demand, minimizing several optimization criteria. The presented methodology has been tested and evaluated in an automotive factory plant of Spain using real production and consumption data.