A study on temperature-setting behavior for room air conditioners based on big data. (July 2020)

• Record Type: Journal Article
• Authors: Yan, Lu; Liu, Meng; Xue, Kai; Zhang, Ziwei
• Is Part Of: Journal of building engineering. Volume 30(2020)
• Publication Date: 2020-07
• Subjects: Occupant behavior -- Room air conditioner -- Set-point temperature -- Data mining -- Cluster analysis; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2020.101197
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: The set-point temperature of room air conditioners (RACs) is extremely critical for cooling energy consumption of residential buildings. However, current research on temperature-setting behavior is limited owing to the limitations of data acquisition. This study aims to identify the typical temperature-setting patterns for RACs and explore the association of temperature-setting behavior with other RAC operation characteristics. The data obtained from the big data cloud platform of an RAC manufacturer were analyzed in this study. These data consist of measured data from 966 bedroom RACs (BRACs) and 321 living room RACs (LRACs). First, the RAC operation characteristics, involving five aspects, namely, set-point temperature, set wind speed, indoor temperature, operation duration, and energy consumption, were extracted from the raw data by transforming, aggregating, and merging the bottom-level measured data. Subsequently, cluster analysis was performed to identify various and typical temperature-setting behavior patterns. Five typical temperature-setting patterns for BRACs and six typical patterns for LRACs were obtained. Afterwards, data mining methods of difference analysis and association analysis were employed to explore the differences and association, respectively, of different temperature-setting patterns with other operation characteristics of RACs (e.g., set wind speed, indoor air temperature, operation duration, and energy consumption). The results of this study can provide researchers with references of temperature-setting strategies in residential building energy simulation and quantify the energy impacts of diverse temperature-setting patterns in residential buildings. Highlights: Data obtained from the big data cloud platform of a room air conditioner manufacturer was analyzed. Five typical temperature-setting patterns for BRACs and six typical patterns for LRACs were obtained. Frequently used and most used set-point temperature, and ranges of temperature adjustment were identified for each pattern. Association of each temperature-setting pattern with other RAC operation characteristics was analyzed.