Sustainability Research of Building Systems Based on Neural Network Predictive Models and Life Cycle Assessment (LCA)–Emergy–Carbon Footprint Method

Abstract

Facing the abnormal climate changes and the goal of carbon neutrality, the ecological sustainability research of building systems has become a focus of attention for experts in this field. However, the definition of sustainable buildings is broad. This article discusses the quantitative analysis of sustainable buildings from the perspectives of an ecological emergy and carbon footprint. It also establishes the long-term sustainability of buildings through predictive neural networks. The research findings indicate that the emergy and carbon emissions during the operational and materials phases dominate the entire system. The calculation and analysis of the emergy sustainability indicator (ESI) demonstrate a decreasing trend in the sustainability of the building system over three time periods (10 years, 20 years, and 30 years), with results of 0.58, 0.238, and 0.089, respectively. As the operational time increases, carbon emissions from the building system also increase, further exacerbating the pressure on the building and reducing its overall sustainability. To address this dilemma, sustainable retrofit measures have been proposed, such as rainwater harvesting and embedded applications of distributed energy sources, which reduce the burden of emergy and carbon emissions. The effectiveness of these measures has been validated in this article, demonstrating their potential to enhance building sustainability and providing references for architects and building managers.