Jul 2026· International Journal of Green Energy· Vol 23, pp. 2731 - 2746· 0 citations· 37 references
Abstract
ABSTRACT Ultra-short-term photovoltaic (PV) power forecasting is vital for power systems with high renewable penetration. Under volatile weather, PV output is dominated by short-term fluctuations, while long-range history becomes weakly informative or even noisy. However, most forecasters rely on fixed temporal receptive fields, failing to adapt their dependency span to changing meteorological regimes. This paper proposes a State-Guided Continuous Temporal-Range Modulation Network (SGTRM) to dynamically regulate the usable temporal context according to evolving weather states. SGTRM first employs a hierarchical causal temporal encoder to extract multi-scale representations from historical PV and NWP sequences. It then introduces a differentiable distance-decay bias into the attention weights, so that distant observations are adaptively down-weighted and the effective temporal range can be adjusted continuously rather than by switching among predefined scales. Experiments show that SGTRM consistently outperforms strong baselines across seasons and weather regimes. Compared with the Transformer baseline, SGTRM reduces the average nRMSE by 35.9% and performs robustly under rainy conditions. Visualization further supports its consistency with atmospheric evolution. These findings suggest that modeling temporal dependency as a continuously regulated and meteorology-conditioned process provides a more flexible and physically interpretable framework for robust ultra-short-term PV forecasting under complex weather environments.
Accurate short-term photovoltaic (PV) power forecasting is critical for secure grid operation and economic dispatch, yet its performance is often degraded by non-stationary irradiance fluctuations induced by cloud transients and weather regime shifts. To address this challenge, this paper proposes a physics-guided temporal fusion architecture built on the Temporal Fusion Transformer (TFT) for multi-horizon probabilistic PV forecasting. The model fuses historical plant measurements, Numerical Weather Prediction (NWP) variables, and solar-geometry features, and introduces a cross-attention fusion module to align future meteorological drivers with the most relevant historical context for improved ramp responsiveness. Physical plausibility is promoted by soft physics-consistency regularization, including non-negativity, rated-capacity bounds, clear-sky envelope constraints, and ramp-rate penalties, while uncertainty is quantified via multi-quantile regression to produce calibrated prediction intervals. Experiments on real-world PV datasets across different seasons and weather conditions show that the proposed approach consistently improves deterministic accuracy (MAE and RMSE) and probabilistic performance (PICP and CRPS) over representative baselines, with particularly robust behavior under regime-shift and ramp events. Overall, combining attention-based temporal fusion with physics-guided learning provides a practical and scalable solution for reliable probabilistic PV power forecasting in highly variable atmospheric conditions.
Pei-Xiang Wu· European Conference on Elect...· 0 citations
The rapid expansion of photovoltaic (PV) systems poses significant challenges to grid stability. Hybrid Energy Systems (HES) are intended to alleviate this volatility, yet their coordinated dispatch often remains suboptimal due to communication delays and ramp-rate constraints. Accurate ultra-short-term PV power forecasting is therefore essential, as it enables preemptive control and timely dispatch adjustments that unlock the full potential of HES. In this study, we propose a novel AI hybrid forecasting framework that integrates a rule-based model with a Decomposition Linear (DLinear) Long Short-Term Memory (LSTM) deep learning core, representing, to the best of our knowledge, a novel integration of a decomposition-based linear model (DLinear) with LSTM networks for ultra-short-term PV power forecasting. The DLinear component decomposes the time series into trend and remainder sequences, which are then independently modeled by separate LSTM networks to capture distinct dynamics. Using data from a 300 kWp PV power station, the framework achieves an average daily prediction accuracy exceeding 93% for both 5-min and 15-min horizons. The model reliably tracks power variations under sunny and rainy conditions, while under volatile cloudy weather its accuracy decreases but still captures essential fluctuation patterns. These results demonstrate the potential of the proposed framework for improving the dispatch and operational reliability of hybrid energy systems. However, further validation across additional seasons and sites is needed to establish broader generalizability.
Fu-Yan Huang, Gang Xiao, Keqin Wang et al.· Energies· 0 citations
Accurate short-term photovoltaic (PV) power forecasting is essential for secure grid operation, economic dispatch, and efficient utilization of renewable energy. However, PV output exhibits strong nonlinearity and nonstationarity due to rapidly varying meteorological conditions, cloud movements, and system uncertainties. Conventional statistical models and single-source deep learning approaches often fail to fully exploit the rich multi-source information available in modern PV plants, such as historical power, on-site meteorological measurements, and clear-sky or numerical weather prediction (NWP) features. In this paper, we propose MSF-TransPV, a Multi-Source Fusion Transformer framework for short-term PV power forecasting. The model adopts a multi-branch temporal encoder that separately processes historical PV output, meteorological variables, and optional clear-sky/NWP-derived features, mapping them into a shared latent space. A cross-variable multi-head attention module is then introduced to explicitly capture the dependencies between PV dynamics and atmospheric conditions, enabling fine-grained interaction across different feature sources. On top of the fused representation, a temporal Transformer encoder models long-range temporal dependencies and feeds a lightweight decoder that supports both point forecasting and probabilistic forecasting via quantile regression. Experimental results on real-world PV datasets demonstrate that MSF-TransPV consistently outperforms persistence, statistical baselines, recurrent neural networks, and vanilla Transformer models in terms of RMSE, MAE, and normalized error metrics, while also providing reliable prediction intervals. These results indicate that explicit multi-source fusion and cross-variable attention are highly effective for PV forecasting under complex and rapidly changing weather conditions.
Xiao-Mei Wang, Pei-Xuan Xu, Xiao-Hui Wang· European Conference on Elect...· 0 citations
: Accurate probabilistic wind farm power forecasting is essential for reserve scheduling, dispatch decision-making, and risk-aware operation under high levels of wind power penetration. However, short-term wind power sequences exhibit strong nonstationarity, heterogeneous environmental variables exhibit time-varying predictive relevance under different meteorological regimes and operating states, and historical operating states and future numerical weather prediction (NWP) variables contribute differently over the forecasting horizon. In addition, direct quantile forecasts may suffer from quantile crossing or physically inconsistent wind-speed-power responses. To address these issues, this paper proposes a physics-regularized gated model for short-term probabilistic wind farm power forecasting, using hourly Local Peak Power (LPP) as the operational evaluation target. The temporal encoder follows the PatchTST patching strategy to capture both short-term ramping behavior and longer-range temporal dependence. A group-gated environmental variable selection and fusion module is designed to adaptively emphasize physically relevant meteorological variable groups and conditionally integrate historical representations with future NWP information. Moreover, a monotone multi-quantile prediction structure with a physics-based regularization term is introduced to improve probabilistic coherence and wind-speed-power consistency. Experiments are conducted against ten representative baselines covering empirical, tree-based ensemble, recurrent, convolutional, and Transformer-based probabilistic forecasting methods. The proposed model achieves the lowest mean absolute error (MAE), root mean squared error (RMSE), and continuous ranked probability score (CRPS) and obtains a 90% prediction interval coverage probability (PICP90) of 0.9001 and a 90% prediction interval normalized average width (PINAW90) of 0.3123, indicating near-nominal interval coverage with a moderate interval width. Relative to the strongest baseline for each metric, the proposed model reduces MAE by 4.92% compared with LightGBM-QR and reduces RMSE and CRPS by 2.78% and 1.94% compared with XGBoost-QR, respectively. Statistical significance analysis then confirms the overall performance gains, while ablation, interpretability, and sensitivity analyses verify the effectiveness of group-gated fusion, the future NWP branch, and physics-based regularization. These results demonstrate that the proposed model provides accurate, reliable, and physically consistent probabilistic forecasts for short-term wind farm operation.
Zhong Chen, Zenan Wang, Siyu Chen et al.· Energy Engineering· 0 citations
Accurate high-resolution photovoltaic power forecasting remains challenging because rapid weather changes and heterogeneous plant configurations produce nonlinear variations and extreme errors. Existing data-driven methods capture complex meteorological relationships but may generate physically inconsistent forecasts, whereas simplified physical models cannot fully represent site specific behavior. This study developed a physics-informed temporal fusion transformer framework combining a transformer–bidirectional long short-term memory forecasting branch with soft irradiance–power and temperature–power constraints. The framework was evaluated using 70,176 observations from Solar Station Site 5, recorded at 15-minute intervals, and independently trained and calibrated across eight solar stations. On the Site 5 test set, it achieved a mean absolute error of 1.4253 megawatts, a root mean squared error of 3.8612 megawatts, and a coefficient of determination of 0.9734. It outperformed standalone temporal fusion transformer, gated recurrent unit, extreme gradient boosting, and random forest; relative to standalone temporal fusion transformer, mean absolute and root mean squared errors decreased by 29.7% and 15.1%, respectively. Sensitivity analysis selected 0.05 as the best nonzero physics weight for Site 5 and revealed a trade-off between predictive accuracy and physical consistency. Residual analysis identified transient and measurement-related outliers. Across the eight stations, coefficients of determination exceeded 0.84 at six sites and reached 0.9804 at Site 6, supporting generalizability across heterogeneous installations following station-specific calibration. Deployment latency remained 0.0864 milliseconds per sample, supporting operational use for renewable-energy integration, grid scheduling, and low-carbon power management.
M. Nasir, Muhammad Farhan Hanif, Muhammad Tahir Hassan et al.· Clean Energy· 0 citations
As the global energy mix shifts toward cleaner sources, the large-scale grid integration of photovoltaic (PV) power poses severe challenges to microgrid frequency stability and security. From a fundamental physical perspective, the solar radiation driving photovoltaic conversion consists of electromagnetic waves on the micrometer scale; as these waves traverse the atmosphere, they undergo intense Rayleigh and Mie scattering caused by cloud dynamics and aerosol attenuation. This atmospheric degradation results in highly nonlinear, transient fluctuations in the effective power reaching the ground. Consequently, computationally intensive full-wave simulation models are impractical for real-time dispatch, while existing purely data-driven deep learning algorithms—lacking physical interpretability—are prone to overfitting and prediction failure under non-stationary meteorological conditions. To bridge this gap between physics and algorithms, this study proposes a novel Physics-Informed Temporal Convolutional Network (PI-TCN) architecture. The framework utilizes Global Horizontal Irradiance (GHI) and Diffuse Horizontal Irradiance (DHI) as inputs to implicitly reconstruct the electromagnetic wave's energy attenuation trajectory, employing 1D causal dilated convolutions to eliminate temporal lag. Furthermore, the model innovatively incorporates non-negative electromagnetic energy boundaries and first-order wave derivatives as penalty functionals during backpropagation, thereby constraining model weights to converge within a physically feasible domain. Benchmarking against a three-year high-resolution dataset from the Desert Knowledge Australia Solar Centre (DKASC) demonstrates that the PI-TCN achieves an exceptionally high coefficient of determination (R2) of 0.9524 and an inference latency of merely 0.08 milliseconds; notably, it attains a Matthews Correlation Coefficient (MCC) of 0.8412 in capturing extreme ramp events. By utilizing the Jacobian matrix of partial derivatives to fully deconstruct the network's "black-box" nature, this research establishes a robust and highly interpretable new paradigm for the convergence of computational electromagnetics and artificial intelligence.
