Dimensionless comparison of solar radiation time series data to address seasonality.

The intermittency of solar power presents significant challenges for grid and energy systems, primarily owing to the unstable and chaotic characteristics of solar radiation time series caused by the combination of seasonality and stochasticity. Currently, there is a lack of simple and easily interpretable methods for describing or transforming solar radiation time series that can directly benefit various solar energy applications. In this study, a deseasonalization method for solar radiation time series based on a transformation matrix is introduced. This method is based on the assumption that daily solar radiation patterns can be scaled by day length and radiation intensity and involves transforming daily solar radiation patterns to eliminate differences due to seasonal effects and geographical location. The proposed method is developed and validated using long-term hourly data collected from five observation stations in Japan: Abashiri, Sapporo, Tokyo, Hiroshima, and Fukuoka, spanning the year from 2000 to 2022. The results indicate that the deseasonalized solar radiation time series exhibits three characteristics: (1) improved correlation in time series, (2) comparable radiation patterns, and (3) seasonal scalability. This study provides a foundational approach to simplifying the solar radiation time series, offering potential for broader applications in solar energy forecasting.