Aug 2026· Applied Sciences· Vol 16, pp. 8261· 1 citation· 32 references
TL;DR
A ramp-aware quantile regression Temporal Fusion Transformer (RQR-TFT) that jointly estimates PV power quantiles and the probability of a future ramp event and demonstrates improved ramp-sample coverage and overall interval quality.
Abstract
Photovoltaic (PV) power interval forecasting models are commonly trained on data dominated by non-ramp samples, which may weaken uncertainty characterization during rapid power changes. This study proposes a ramp-aware quantile regression Temporal Fusion Transformer (RQR-TFT) that jointly estimates PV power quantiles and the probability of a future ramp event. Ramp labels are constructed from the normalized power change between adjacent sampling instants. A shared Temporal Fusion Transformer (TFT) encoder extracts temporal representations from historical PV power and meteorological variables, and two output branches perform quantile forecasting and ramp-event identification. Ramp-sample-weighted quantile loss and positive-class-weighted classification loss are jointly optimized to increase the influence of minority ramp samples. The proposed method is evaluated for 4 h ahead forecasting using measurements collected from a 50 MW PV power station during 2019–2020. For the nominal 90% prediction interval, RQR-TFT achieves a ramp-sample prediction interval coverage probability (PICPR) of 0.864, an overall prediction interval normalized average width (PINAW) of 0.209, and an overall normalized interval score (NIS) of 0.365. The area under the precision–recall curve for ramp-event identification is 0.906. The results demonstrate improved ramp-sample coverage and overall interval quality, although ramp-sample coverage remains below the nominal level.
Accurate photovoltaic (PV) power forecasting is becoming increasingly important for the secure and economical operation of renewable-rich power systems. However, conventional data-driven forecasting methods generally rely on sufficient site-specific historical observations, limiting their applicability to newly commiss...
Ping Zhou, Fu-Jie Cai, W. Jiang et al.· Energies· 0 citations
ABSTRACT Photovoltaic (PV) power forecasting is a foundational capability for operating power systems with high shares of variable renewable generation. However, the methodological landscape is fragmented across physical-based models, statistical time-series approaches, machine learning (ML), and increasingly diverse h...
Jokūbas Jonuška, R. Damaševičius, Diana Belova-Plonienė et al.· International Journal of Gre...· 1 citation
Data-driven machine learning (ML) models forecast photovoltaic (PV) power accurately, but the conventional Random Forest (RF) samples candidate predictors uniformly and therefore ignores established physical knowledge about how meteorological variables drive PV generation. This paper proposes the Domain-Aware Random Fo...
S. Nyirenda, Musa Ndiaye, Esau Zulu et al.· Energies· 0 citations
Accurate day-ahead forecasting of photovoltaic (PV) power and electric load is essential for reliable power system operation under growing renewable penetration and demand complexity. Both series, however, are strongly nonstationary, with multiscale dynamics driven by external factors. This study develops a multiscale...
Weather transitions can cause rapid fluctuations in wind power and complicate the construction of prediction intervals. This study presents a robust error-correction framework that balances interval coverage and width. Multiple backward cloud transformation based on sampling with replacement (MBCT-SR) is combined with...
Uncertainty-bounded photovoltaic (PV) power monitoring is essential for reliable field monitoring, as field meteorological measurements may be affected by sensor noise, calibration bias, and outlier contamination, while PV power output also varies with changing environmental conditions. These measurement-related effect...
Heng-Yi Su, Tzu-Chieh Lin· IEEE Transactions on Instrum...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.