This study provides an in-depth comparative analysis of four state-of-the-art neural architectures, confirming that high-fidelity point forecasts and rigorously quantified uncertainty can be achieved simultaneously, providing a clear path toward more dependable PV dispatch, reserve allocation, and market participation.
Abstract
Reliable short-term photovoltaic (PV) power forecasts are pivotal to grid balancing, intraday market clearing, and asset optimization. This study provides an in-depth comparative analysis of four state-of-the-art neural architectures–a convolutional-recurrent hybrid (CNN-LSTM), an LSTM-Autoencoder, a standalone LSTM, and a time-series Transformer–trained on three years of irradiance and power data with one-minute sampling interval from two meteorologically contrasting PV sites (semi-arid Colorado and desert Nevada). Each model is evaluated in a probabilistic forecasting setting, where kernel density estimation (KDE)-based residual post-processing and quantile extraction are used to obtain calibrated prediction intervals. The benchmark adopts an irradiance-driven forecasting framework, using historical irradiance observations as inputs and future PV power as the prediction target, thereby enabling a controlled assessment of each architecture’s ability to model the underlying irradiance-to-power relationship. Robustness is probed through three stressors: truncated training histories (1–3 years), temporal coarsening (1-, 5-, and 15-minute records), and up to 30% randomly or block-removed observations. Across all tests, CNN-LSTM consistently delivers the narrowest and most reliable intervals, leveraging its convolutional front-end to detect rapid cloud-edge ramps while the LSTM tail preserves long-range context. Compared with a strong LSTM benchmark, the hybrid reduces mean interval width by 22%, lowers the Continuous Ranked Probability Score (CRPS; a scoring rule measuring the distance between the forecast distribution and the observed outcome - lower is better) by 11%, and boosts empirical coverage by 5% (achieving 98.2% reliability). These advantages remain intact under data scarcity, coarse sampling, and substantial missingness, highlighting the critical role of architectures that unite local feature extraction with sequential memory. The findings confirm that high-fidelity point forecasts and rigorously quantified uncertainty can be achieved simultaneously, providing a clear path toward more dependable PV dispatch, reserve allocation, and market participation.
The large-scale integration of photovoltaic (PV) systems into modern power grids impacts operational challenges in reducing the intermittency of solar irradiance. Short-term forecasting, especially with accurate result is essential for grid stability, economic dispatch, and demand-side management. However, the development of robust deep-learning models is frequently constrained by the limited readiness of high-quality and labeled actual generation data. This paper presents a simulation-driven methodology to address this data scarcity problem. A synthetic PV generation dataset is created using a deterministic mathematical model. This method captures the diurnal solar cycle, augmented with Gaussian stochastic noise to simulate the effects of cloud cover and atmospheric turbulence. A Long Short-Term Memory (LSTM) network in two layer is trained on the synthetic dataset and subsequently validated against real hourly PV generation data for Bali, Indonesia. This data was obtained from the PVGIS-ERA5 database. The method performs 1-hour-ahead sequence-to-point forecasting using a 6-hour retrospective window. A persistence baseline model is employed for comparative benchmarking. This experiment found that the LSTM outperforms the persistence baseline on both datasets significantly. For the empirical Bali PV dataset, the LSTM attains a Mean Absolute Error (MAE) and a Root Mean Square Error (RMSE) of 29.75 W and 43.31 W, respectively, with a corresponding R2 value of 0.9474. Compared to the persistence benchmark's MAE of 69.00 W, the network successfully cuts prediction error by 56.9%. These results validate both the simulation-driven training approach and the LSTM's capability for short-term solar forecasting under tropical conditions.
I. D. Saputra, Nicola Schulz, I Nyoman Kusuma Wardana et al.· 2026 International Conferenc...· 0 citations
A systematic comparison of Long Short-Term Memory and transformer-based architectures for deterministic short-term PV power forecasting using publicly accessible data from multiple climatic regions highlights the advantages of attention-based sequence modeling for PV applications and offers practical guidance on feature design, input horizon selection, and hyperparameter ranges for future data-driven PV forecasting studies.
Marcel Lüdecke, Elias Oppermann, Michel Meinert et al.· e+i Elektrotechnik und Infor...· 0 citations
With the global shift toward green energy, solar photovoltaic (PV) power has expanded rapidly. However, the unpredictable nature of PV generation, caused by changing weather conditions like irradiance and temperature, challenges grid stability and power scheduling. Therefore, developing smart forecasting models for high-precision PV power prediction is essential for modern grid management. This paper introduces an optimized forecasting framework using Long Short-Term Memory (LSTM) networks. By integrating historical power generation data with localized meteorological factors, a multivariate predictive model was developed and validated using empirical data from a 100 MW PV plant. Based on Pearson correlation analysis, four key features—temperature, direct normal irradiance (DNI), relative humidity, and cloud cover—were chosen as the main drivers of PV output. A multivariate LSTM model was then trained and carefully tested using time series cross-validation. Results show the multi-feature architecture consistently outperforms and surpasses single-feature benchmarks. Specifically, the model achieved a peak R2 of 0.9864 and minimum MAE of 1.4057 kW. Across four validation sets, R2 remained stable (0.9755–0.9836), with most errors tightly bounded within ±5 kW. These findings confirm the model’s excellent accuracy and its value for improving power system dispatch and resource planning.
Accurate short-term photovoltaic (PV) power fore-casting is important for energy-aware production planning in manufacturing environments integrating renewable sources. This paper benchmarks six deep learning architectures—LSTM, BiLSTM, GRU, CNN–LSTM, CNN–BiLSTM, and CNN–GRU—for PV power forecasting at 30-minute resolution under a strictly controlled protocol (identical data split, preprocessing, input window, and evaluation metrics). Experiments on a real annual PV dataset show that CNN–LSTM achieves the best RMSE in the univariate setting (PV power only) with RMSE = 1.153 and R = 0.972, while LSTM attains the lowest MAE (MAE = 0.564). In a multivariate extension (PV + weather), CNN–LSTM remains the best hybrid model (RMSE = 1.148, R = 0.969), with only marginal RMSE improvement over the univariate CNN– LSTM. Beyond point forecasting, the proposed pipeline converts the best model output into planning-ready uncertainty inputs by generating Monte Carlo trajectories from validation residuals and reducing them into compact multi-day scenarios with probability weights over a weekly horizon. These weighted scenarios are designed to be directly injected as PV availability inputs in uncertainty-aware production scheduling and energy management.
Assiya Zahid, Lamia Hammadi, Patrice Leclaire et al.· International Conference on...· 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.
Fuyan Huang, Gang Xiao, Keqin Wang et al.· Energies· 0 citations
Accurate photovoltaic (PV) power forecasting is essential for reliable microgrid operation, efficient energy dispatch, and improved utilization of renewable energy resources. Existing forecasting methods often have limited capacity to represent the nonlinear relationships between meteorological conditions and PV power output. They also tend to underrepresent the temporal dynamics of PV generation and the physical principles governing photovoltaic energy conversion. To address these limitations, this study proposes a hybrid forecasting framework, CNN-LSTM-PINNs, that integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Physics-Informed Neural Networks (PINNs). In the proposed framework, CNNs extract spatial dependencies among multivariate meteorological variables, LSTM networks capture temporal dependencies in PV generation, and PINNs incorporate soft physical constraints derived from photovoltaic energy conversion mechanisms. The proposed model is evaluated using publicly available datasets from three large-scale PV power stations in China, with observations recorded at 15-min intervals. The empirical results show that CNN-LSTM-PINNs outperform the conventional CNN-LSTM benchmark across the primary station-level datasets. Relative to the benchmark model, the proposed framework reduces Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) and improves the coefficient of determination (R2). These results indicate that embedding physical constraints into data-driven forecasting models can improve PV power prediction accuracy. The model also shows stronger robustness and generalization performance under heterogeneous operating conditions, although its effectiveness is contingent on relatively stable data distributions. Feature-importance analysis further indicates that global horizontal irradiance (GHI) and irradiance-derived variables are the most informative predictors of PV power output. Overall, this study provides a physics-informed hybrid modeling approach for high-resolution PV power forecasting in microgrid applications.
Jiabo Gou, Xiaoqiao Liao, Sheng Li et al.· Energies· 0 citations