What question did this study set out to answer?

The aim is to analyze hybrid models that integrate physical and machine learning approaches for photovoltaic power forecasting.

July 30, 2026Solar EnergyOpen Access

Hybrid photovoltaic power forecasting by combining physical and machine learning models

View Full Paper

Ask AI

Bookmark

Authors

Hybrid models enhance forecasting accuracy for photovoltaic power in varying climates, indicating improved reliability.

The aim is to analyze hybrid models that integrate physical and machine learning approaches for photovoltaic power forecasting.
Two hybrid approaches were implemented: physics-informed ML and physics-guided models.
Data from PV systems in Brazil and the Netherlands were utilized for validation.
Models ranged from classical to complex formulations to determine performance impact.
Machine learning methods, particularly nonlinear ensemble techniques, significantly improved forecasting accuracy.
Performance of hybrid models showed weak sensitivity to physical model refinements, impacting metrics by less than 1%.
The combination of physical and ML helped enhance accuracy and reliability in photovoltaic forecasting.

Santos et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark