머신러닝 기술을 활용한 건축물 사례기반 온실가스 감축 의사결정 지원 모델 구축

Abstract

Purpose: Many countries have implemented policies to reduce greenhouse gas (GHG) emissions in public buildings, emphasizing the leading role of the public sector. In Korea, in order to achieve a 30% reduction in GHG emissions by 2030, public agencies must set annual targets or quotas. However, the lack of experts and support are the biggest obstacles to achieving this reduction target. Methods: This study constructed a GHG evaluation database (DB) and Data set based on energy end uses, GHG reduction technology with the aim of decision making about GHG reduction with minimal building information and limited expert knowledge. The GHG evaluation DB was built using data from the Commercial Building Energy Consumption Survey (CBECS), an energy consumption survey of 6,720 public and commercial buildings by the US Department of Energy. In addition, a DB for evaluating the reduction amount of greenhouse gas reduction technology was established with reference to 1,206 greenhouse gas reduction technology application projects by the Korea Energy Survey. The database was used for constructing data set, we developed a machine learning-based GHG reduction decision support model. Result: Additionally, the case study of domestic public buildings, the economic and environmental benefit of applying greenhouse gas reduction technology were evaluated. The evaluated building can reduce about 111 tonCO₂-eq and convert it into economic profit of 36 million won, confirming the applicability of the model.