Prediction Methodology of Energy Consumption Based on Random Forest Classifier in Korean Residential Apartments
Fazli Wahid · Advanced science and technology letters · 2015
Recently, we are focusing on managing the energy system automatically. The smart grids perform much functionality like a self-healing capability, a self-resistance to external attacks, and actively engage the consumers. Also we provide high quality power to consumers in the operation environment of smart grids. In this paper, we propose a prediction methodology based on random forest classifier of energy consumption in Korean residential apartments. The prediction consists of three stages namely data retrieval, data processing and prediction. The prediction results help the energy suppliers to make decisions for the provision of energy to different apartments according to their demand.