Artificial Intelligent-Based Techniques in Solar Radiation Applications

Abstract

The evolving smart grid emerges as a response to the challenges posed by the unreliability and inefficiency of the traditional electric grid. This transformation is crucial due to issues like the aging infrastructure and the intermittency of renewable energy sources, particularly solar radiation. The smart grid is anticipated to facilitate two-way power flows and introduce innovative technologies. This study explores the impact of smart grid technologies, particularly those supported by artificial intelligence (AI), on-demand load, future energy consumption, and energy management services. The focus is on AI-based systems applied in solar energy applications, aiming to enhance efficiency and reduce costs. Various AI techniques, including neural network methods, are examined for their role in addressing challenges such as forecasting, fault diagnosis, and control in solar radiation applications. The research introduces and compares three AI models—gated recurrent unit (GRU), artificial neural network (ANN), and long short-term memory model (LSTM)—for predicting solar irradiance. The outcomes emphasize the versatility of AI, not only in solar systems but also in extending its applications to other renewable energy systems like wind and diverse fields such as security, reliability, and stability.