Prediction of Solar Energy Production Using Low-Cost Data Logger and ANN Algorithm

The urgent need for sustainable energy solutions drives the exploration of solar energy as a viable and abundant resource. This chapter delves into a novel approach that combines low-cost data loggers with Artificial Neural Networks (ANN) to predict solar energy production with high accuracy. The study emphasizes the importance of collecting comprehensive environmental and solar irradiance data, which are crucial for developing precise predictive models. The ANN algorithm, known for its adaptability and accuracy, is employed to analyze the collected dataset, achieving an impressive R2 score of 0.95. This high level of precision positions the methodology as a robust tool for solar energy planning and management. The chapter also discusses the structure of the predictive system, including data acquisition, transmission, storage, and preprocessing, as well as the architecture of the ANN model. Furthermore, it highlights the potential applications in smart cities, where IoT-enabled sensors and devices monitor solar panel performance in real-time, optimizing energy production and supporting the transition to cleaner, sustainable energy sources. The findings underscore the effectiveness of the approach in forecasting energy production from photovoltaic systems, paving the way for advancements in solar energy forecasting and management.