AI-Based Solar Power Generation Forecasting in Building Integrated Photovoltaic (BIPV) Enabled Smart Buildings

Building Integrated Photovoltaic (BIPV) systems have been increasingly deployed in modern smart buildings for renewable energy harvesting. However, the solar power output from BIPV systems varies considerably depending on the fluc- tuating environmental factors like solar irradiance, temperature, humidity, and wind speed. This has often caused difficulties in planning and utilizing solar power. In this work, an Artificial Intelligence-based solar power forecasting model for BIPV-enabled smart buildings has been proposed using a hybrid deep learning approach. The proposed model has been developed using Artificial Neural Network (ANN), Support Vector Regression (SVR), and Long Short-Term Mem- ory (LSTM) networks for predicting solar power output using historical photovoltaic output data and various meteorological factors. The experimental results have been presented to show the effectiveness of the proposed solar power forecasting model using an Attention-based LSTM network, which has demonstrated improved prediction accuracy compared to traditional machine learning approaches. The proposed system can be effectively utilized for optimizing energy usage in smart buildings and minimizing dependence on traditional grid supply.