Sigmoid Kernel Canonical Correlation Analysis (SCKCCA) for PV Output Power and Meteorological Factors

The global shift towards renewable energy sources has spotlighted solar energy as a pivotal solution for sustainable power generation. As the adoption of photovoltaic (PV) panels grows, understanding the intricate relationship between PV output power and varying weather conditions becomes paramount. Traditional linear models have provided foundational insights, but often fall short in capturing the nuanced interactions among multiple weather variables. Recent state-of-the-art approaches, leveraging machine learning and advanced statistical methods, have enhanced predictive accuracy but often lack in-depth interpretability. This paper introduces the Sigmoid Kernel Canonical Correlation Analysis (SCKCCA) as a novel approach to bridge this gap. By integrating the sigmoid kernel into Canonical Correlation Analysis, SCKCCA captures non-linear correlations between weather conditions and PV output, offering both predictive precision and clear interpretability. Our exploration underscores the significance of SCKCCA in the broader context of solar energy research, emphasizing its potential to influence future renewable energy strategies and policies.