Neural Networks Based Home Energy Management System in Residential PV Scenario

In this paper we propose and design a home energy management system using artificial intelligence. The device, monitoring home loads, detecting and forecasting photovoltaic (PV) power production and home consumptions, informs and influences users on their energy choices. A neural network self-learning prediction algorithm is used to forecast, over a determined time horizon, the power production of the PV plant and the consumptions of the house. The online learning algorithm is based on a Radial Basis Function (RBF) network and combines the growing criterion and the pruning strategy of the minimal resource allocating network technique. Furthermore a novel method to simulate electrical consumptions and evaluate the potential benefits of a Demand Side Management is developed. The proposed solution has been experimentally tested in 3 houses with 3.3 KWp PV plant.