What question did this study set out to answer?

The aim is to improve the accuracy of photovoltaic power forecasting using measured weather and generation data.

March 25, 2026Journal of Korea Multimedia Society0 citationsOpen Access

A Study on a Hybrid Photovoltaic Power Forecasting Model Based on Measured Weather and Generation Data

CPChaeyeon Park YHYeongwoo Ha YLYeunghak Lee

Puntos clave

The aim is to improve the accuracy of photovoltaic power forecasting using measured weather and generation data.
Developed a week-ahead forecasting framework based on hourly PV generation and meteorological data.
Constructed a dataset with 8,033 records spanning May 2024 to April 2025.
Applied cyclic time encodings, lag features, rolling statistics, and interaction features to capture data nuances.
Compared performance of machine learning models, particularly XGBoost, against deep learning models.
XGBoost achieved the best performance with an RMSE of 70.84 W and an R² of 0.925.
A Ridge-based stacking ensemble improved accuracy, achieving an RMSE of 62.21 W and an R² of 0.943.
Lightweight stacking was shown to enhance forecasting stability under limited data conditions.

Resumen

Accurate photovoltaic(PV) power forecasting is essential for stable grid operation, yet PV output varies sharply with weather. This study proposes a measured-data-based week-ahead (168-hour) forecasting framework using on-site PV generation and hourly meteorological observations. We construct an hourly dataset of 8,033 records (May 2024–April 2025) and apply cyclic time encodings, lag features, rolling statistics, and interaction features to capture temporal dependency and periodicity. Using timeordered splits, we compare tree-based machine learning models with deep learning baselines. Among single models, XGBoost achieves the best performance (RMSE 70.84 W, R² 0.925). A Ridge-based stacking ensemble further improves accuracy (RMSE 62.21 W, R² 0.943). These results suggest that lightweight stacking improves stability for week-ahead PV forecasting under limited measured data.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo