TY - JOUR
T1 - Applications of Clustering and Isolation Forest Techniques in Real-Time Building Energy-Consumption Data
T2 - Application to LEED Certified Buildings
AU - Kim, Jonghoon
AU - Naganathan, Hariharan
AU - Moon, Soo Young
AU - Chong, Oswald
AU - Ariaratnam, Samuel
N1 - Publisher Copyright:
© 2017 American Society of Civil Engineers.
PY - 2017/10/1
Y1 - 2017/10/1
N2 - Buildings are the largest consumer of energy in the United States from various sectors that includes transportation, industry, commercial, and residential buildings. Leadership in Energy and Environmental Design (LEED) certification program, home energy rating system (HERS), and American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) standards are developed to improve the energy efficiency of the commercial and residential buildings. However, these programs, codes, and standards are used before or during the design and construction phases. For this reason, it is challenging to track whether buildings still could be energy efficient post construction. The primary purpose of this study was to detect the anomalies from the energy consumption dataset of LEED institutional buildings. The anomalies are identified using two different data mining techniques, which are clustering, and isolation Forest (iForest). This paper demonstrates an integrated data mining approach that helps in evaluating LEED energy and atmosphere (EA) credits after construction.
AB - Buildings are the largest consumer of energy in the United States from various sectors that includes transportation, industry, commercial, and residential buildings. Leadership in Energy and Environmental Design (LEED) certification program, home energy rating system (HERS), and American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) standards are developed to improve the energy efficiency of the commercial and residential buildings. However, these programs, codes, and standards are used before or during the design and construction phases. For this reason, it is challenging to track whether buildings still could be energy efficient post construction. The primary purpose of this study was to detect the anomalies from the energy consumption dataset of LEED institutional buildings. The anomalies are identified using two different data mining techniques, which are clustering, and isolation Forest (iForest). This paper demonstrates an integrated data mining approach that helps in evaluating LEED energy and atmosphere (EA) credits after construction.
UR - http://www.scopus.com/inward/record.url?scp=85025089060&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85025089060&partnerID=8YFLogxK
U2 - 10.1061/(ASCE)EY.1943-7897.0000479
DO - 10.1061/(ASCE)EY.1943-7897.0000479
M3 - Article
AN - SCOPUS:85025089060
SN - 0733-9402
VL - 143
JO - Journal of Energy Engineering
JF - Journal of Energy Engineering
IS - 5
M1 - 04017052
ER -