Analysis of Parameters which Affects Prediction of Energy Consumption in Buildings using Partial Least Square (PLS) Approach

Abstract

The development of energy consumption prediction model is an integral part of the management and improvement of building energy efficiency in order to save the energy and to reduce the environmental impact. There are factors that affect energy in buildings which are heating and cooling. Building materials, ventilation, building direction, and building area are the important factors to determine building energy efficiency. The study aims to find the role of input variable (independent) on the output variable (dependent) in the form of Heating Load (HL) and Cooling Load (CL). The study employs Partial Least Square (PLS) analysis method which is a variant-based Structural Equation Modeling analysis known as SEM-PLS. The result of the study indicates that the estimation of inner model of the direct influence of Orientation on Cooling Load and Heating Load is not significant. It means the size of the Orientation value does not significantly affect the increase/decrease in Cooling Load and Heating Load. While the direct effect of Overall Height, Wall Area, and Surface Area on Cooling Load and Heating Load is significant. It means that the value of Overall Height, Wall Area, and Surface Area has a significant effect on increasing/decreasing Cooling Load and Heating Load. The results of this study are expected to be useful to help building designers, especially related to energy efficiency in the buildings. In addition, the development of this model can be used as an alternative in determining the factors that affect comfort in a building.