Neural Networks and Support Vector Regression Models to Evaluate Energy Efficiency-Noise Reduction Coefficient Correlation

Abstract

The main purpose of this manuscript is to investigate the correlation potential between energy efficiency (EE) and the noise reduction coefficient (NRC) using advanced machine learning techniques, including support vector regression (SVR) and neural networks (NN) models. This research uses real datasets of Heating/Cooling energy consumption and Noise reduction coefficient NRC of four residential buildings, namely A, B, C, and D, each with a common floor area of 80 m<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> but differing thermophysical properties of their envelopes. The buildings A, B, C, and D are insulated, respectively, with a single clay brick wall, a single concrete brick wall, a double clay brick wall insulated with polystyrene, and a double clay brick wall with a medium of air gap. The analysis of our proposed models presents a moderate relationship between EE and NRC, but no strong correlation is observed. Also, the replacement of the polystyrene layer with an air gap between double brick walls in building D gives rise to interesting energy savings compared to building C and the other buildings under study. It doesn't remain easy to provide a strong correlation between EE and NRC, but it may be achievable in buildings featuring homogeneous and thicker wall structures.