BIPV Modeling with Artificial Neural Networks: Towards a BIPV Digital Twin

Modeling the photovoltaic (PV) energy output with high accuracy is essential for predicting and analyzing performance of a PV system. In particular cases building-integrated building-attached systems (BIPV BAPV, respectively) time-varying partial shading conditions are relevant added difficulty modeling power conversion. The availability laser imaging detection ranging (LIDAR) data to create very-high-resolution elevation digital models can be effectively used computing at resolution. this work, an artificial neural network (ANN) has been model generation different BIPV arrays on 5 min basis using meteorological solar irradiance on-site conditions, as well patterns estimated from surface inputs. ANN validated three years 5-min-basis monitored showing very (6–16% relative error depending façade). proposed methodology combines computation powerful machine learning algorithms vertical under conditions. results presented here prove also capability techniques towards creation twin specific case that complements conventional monitoring strategies in diagnosis anomalies.