Statistical Analysis of Solar Irradiance Variability

• Published in: IEEE Power & Energy Society General Meeting pp. 1 - 5
• Main Authors: Nikolopoulos, Angelos R., Batzelis, Efstratios I., Lewin, Paul, Nikolaou, Nikolaos
• Format: Conference Proceeding
• Language: English
• Published: IEEE 21.07.2024
• Subjects: Energy resolution; Fluctuations; Forecasting; Gaussian mixture model; Gaussian Mixture Model (GMM); Meteorology; photovoltaic (PV) forecasting; Photovoltaic systems; Power systems; Solar irradiance; State estimation; Statistical analysis
• ISSN: 1944-9933
• DOI: 10.1109/PESGM51994.2024.10689164

Solar photovoltaic (PV) generation forecasting is an important tool to power system operators, but struggles under conditions of intermittent solar irradiance. Although studying and forecasting irradiance itself has been the subject of much research, little progress has been made on the variability (or fluctuation) of irradiance and its statistical properties, despite it being an important parameter in generation forecasting, state estimation and other power system applications. This paper takes a close look into the statistical nature of irradiance variability and shows that it can be sufficiently modeled by a Gaussian Mixture Model (GMM) of six components. Furthermore, an investigation on the required time resolution demonstrates that sub-minute resolution is necessary to accurately capture irradiance variability. The analysis is performed on a one-second resolution irradiance dataset provided by NREL.