Analyzing Determinants of Energy Consumption for Heating and Cooling in Apartment Units - Comparison of Linear and Nonlinear Statistical Models

To develop detailed and effective energy saving plans for existing buildings, determinants of actual energy consumption should be identified first. Linear statistical models that have widely used in prior studies presents a major limitation in treating nonlinear problems. Therefore, any determinant having nonlinear relationship with the energy consumption has been hardly found. To address this problem, this study proposes a novel way to discover hidden determinants, using both linear and nonlinear models: multiple linear regression (MLR) and decision tree (DT). This study used energy consumption and characteristics data of 53 apartment units in Seoul, which were collected by real-metering and field survey. Through MLR and DT models, building, system, and occupant characteristics that significantly affect each of energy consumption for heating and cooling were identified. As a result, some of determinants were common in both models while some were not (e.g. year of building permit, COP of air conditioner, and number of employed residents). The result implies that it is desirable to analyze determinants of energy consumption using both linear and nonlinear models instead of relying on a single model.