Characterization of an Energy Consumption Model for a Net Zero Energy Building Laboratory
Abstract
Energy saving and the integration of renewable energies are critical issues in Net Zero Energy Buildings (NZEB). In this context, the availability of methods for the prediction of the Electrical Load Demand (ELD) is extremely important mainly due to its relevance for an appropriate energy management, sizing of NZEB energy systems, and, especially, for the implementation of suitable energy control strategies, such as energy storage. This paper describes the development and assessment of an electricity demand short-term predictive Artificial Neural Network (ANN) model for a characteristic laboratory within an NZEB located at Almería (Southeastern Spain). As the model is aimed to be used as part of advanced building energy control schemes, some specific requirements, as a tradeoff between accuracy and simplicity, have been considered. The work presented in this paper contains both a description of the algorithms and reference data for an appropriate development of this kind of models. Moreover, a detailed discussion of the obtained ANN model which has been validated using real data obtained from the NZEB used as case-study has been included.
Keywords
Net zero energy buildings, Electrical load demand, Artificial neural network, Radial basis functions