A Smart Building Energy Management using Internet of Things (IoT) and Machine Learning

Abstract

Energy consumed by smart buildings in Malaysia accelerates continuously due to the growth of country’s population. This occurrence creates awareness towards energy management team to organize and manage their energy consumption systematically. The main purpose of this research is to analyse and predict the energy consumptions in order to achieve a better energy management of a commercial smart building towards efficiency. In this paper, historical data of hourly consumption of maximum demand collected at the selected tenants of the smart commercial building implemented with Internet of Things (IoT) has been analysed using statistical method computed with the formula of mean, variance, skewness and kurtosis. k-nearestneighbour (k-NN) method has been applied on the data of consumptions for prediction process by using three different values of nearest neighbour. The predicted data has been separated into different training and testing ratios which are 70% and 30%. Root mean square error (RMSE) is proposed in this paper to evaluate the performance of predicted data. The results showed that the nearest neighbour with k = 5 is the most accurate since it provides the lowest average RMSE value with 5.73, 8.54 and 0.35 for each tenant of the building respectively. This model will be used as a reference to predict the energy usage for the upcoming period. In the future, data that has been predicted can be integrated into the available system for user monitoring and controlling purposes.