An analysis on energy consumption of two different commercial buildings in Malaysia

The increment rate of gross domestic product (GDP) and electricity consumption had been consistent until the mid-1990s where Malaysia experience a slump in GDP but our electricity consumption continued to increase until now. To reduce the energy consumption, we need to perform an energy analysis based on factors that affect the energy consumption and trends of data from two different commercial buildings. There are five main factors which are equipment's, outside temperature, building structure, operating hour and people. Among these factors, temperature will be considered to analyse energy consumption in two different commercial buildings in Malaysia. The motivation to conduct this analysis is to establish the benchmarking of the energy efficiency of the commercial buildings since it has not established yet in Malaysia. In this paper, Multilayer Perceptron (MLP) based on Artificial Neural Networks (ANN) has been implemented for energy efficiency analysis. The results of this analysis showed that the energy prediction by using artificial neural network is better than traditional method used by industry which is linear regression where linear regression show the highest error square for both buildings which is 30491.23 for Skywarth and 91738.31 for Skymage building. By comparing the results of two different buildings, we can conclude that outside temperature plays an important role in determining energy consumption of commercialize building. For future study, an advanced method such as k-Nearest Neighbor or Support Vector Machine can be used to predict the energy consumption, so we could obtain better prediction results.