Energy consumption prediction of a smart home using non-intrusive appliance load monitoring

The increasing need for energy has been a major problem in recent years. In view of this problem, energy saving and reduction of energy consumption are strongly encouraged. The residential sector accounts an important part of final energy consumption and is therefore a major challenge for improving energy efficiency. In this work, individual energy consumption is determined from measurements taken downstream at the energy meter using a single current and a single voltage sensor, without a learning phase or knowledge of the equipment inside the home. This non-intrusive appliance load monitoring (NIALM) method has several advantages: it allows us to process the load curves and to extract useful information for the identification of the uses and to prevent the most energy consuming appliances. In addition, we will apply the Auto Regressive Moving Average with eXternal inputs (ARMAX) model to predict the energy consumption. These two approaches will allow us to better analyze the management, control, metering and billing system of consumption in order to ensure better energy efficiency in buildings.