Energy Consumption Patterns and Characteristics of College Dormitory Buildings Based on Unsupervised Data Mining Method

The college building is a large energy consumer with high density of consumption. However, less attention paid to buildings, particularly dormitory buildings. Based on the one-year historical data collected from 20 buildings located in Wuhan, China, this study aims propose three-stage strategy identify and analyze consumption patterns characteristics dormitories detail, including determining patterns, analyzing key based four indexes, examining three influencing factors (occupants’ gender floor orientation location rooms). results show that heavy users (around 10% all occupants) consume around 20% total have narrowest comfort temperature range. light users, 42% occupants, only approximately 27% energy. Their different tolerance coldness main reason contributing males top corner tend significantly more hot weather. This would help campus facilities understand use behavior occupants formulate adequate policies so as improve management campuses.