Data‐driven forecasting of local <scp>PV</scp> generation for stochastic <scp>PV</scp> ‐battery system management

Power systems face more uncertainty by increasing photovoltaic system installations on the roof of buildings. To optimally manage energy and available flexibility in a building, stochastic optimization is used to take an optimal decision under minimize operational cost. In optimization, scenario set as input represent random variable, PV generation this case. paper, data-driven method proposed obtain distribution variable later generate sets representing day-ahead installed building. This only based historical it does not require any other external data such weather forecasts. A machine learning-based technique applied forecast production following generating for decision-making. Later, PV-battery management problem formulated two-stage while generated optimization. The algorithm tested scheduling commercial informed real-life measurement data. results show that able capture providing cost-optimal reliable solution application problem. Without using data, error reaches NRMSE = 11.89, there 5% reduction cost Moreover, easy implement building system.