Explore the vast range of titles available.

Accurate solar radiation forecasting is essential to operate power systems safely under high shares of photovoltaic generation. This paper compares the performance of several machine learning algorithms for solar radiation forecasting using endogenous and exogenous inputs and proposes an ensemble feature selection method to choose not only the most related input parameters but also their past observations values. The machine learning algorithms used are: Support Vector Regression (SVR), Extreme Gradient Boosting (XGBT), Categorical Boosting (CatBoost) and Voting-Average (VOA), which integrates SVR, XGBT and CatBoost. The proposed ensemble feature selection is based on Pearson coefficient, random forest, mutual information and relief. Prediction accuracy is evaluated based on several metrics using a real database from Salvador, Brazil. Different prediction time-horizons are considered: 1 h, 2 h and 3 h ahead. Numerical results demonstrate that the proposed ensemble feature selection approach improves forecasting accuracy and that VOA performs better than the other algorithms in all prediction time horizons.

In this paper, an improved clustered adaptive neuro-fuzzy inference system (ANFIS) to forecast an hour-ahead solar radiation data for 915 h is introduced. First, we have classified the history data of solar radiation time series to decrease the input sample size using clustering methods. Three methods are used, namely, fuzzy c-means (FCM), subtractive clustering, and grid partitioning. These methods allow classifying the input data into groups; each group has similar properties that help to understand the correlation between the data and by consequence simplify the forecasting process. Second, we designed an ANFIS structure that takes both advantages of fuzzy theory to describe the uncertain phenomena of the data and artificial neural network algorithm, which has a self-learning ability. Finally, by combining clustered data and ANFIS model, an hour-ahead forecasting is achieved, and it was validated using measured data. The advantage of the proposed method is that provides the ability to use implicitly the information associated with the forecasting problem, without a priori knowledge of the relationships between the different variables solar radiation. The comparison results show that the ANFIS with FCM clustering model gives the best results with RMSE equals to 112 W/m2 and high values of FS.

Microgrid is becoming an essential part of the power grid regarding reliability, economy, and environment. Renewable energies are main sources of energy in microgrids. Long-term solar generation forecasting is an important issue in microgrid planning and design from an engineering point of view. Solar generation forecasting mainly depends on solar radiation forecasting. Long-term solar radiation forecasting can also be used for estimating the degradation-rate-influenced energy potentials of photovoltaic (PV) panel. In this paper, a comparative study of different deep learning approaches is carried out for forecasting one year ahead hourly and daily solar radiation. In the proposed method, state of the art deep learning and machine learning architectures like gated recurrent units (GRUs), long short term memory (LSTM), recurrent neural network (RNN), feed forward neural network (FFNN), and support vector regression (SVR) models are compared. The proposed method uses historical solar radiation data and clear sky global horizontal irradiance (GHI). Even though all the models performed well, GRU performed relatively better compared to the other models. The proposed models are also compared with traditional state of the art methods for long-term solar radiation forecasting, i.e., random forest regression (RFR). The proposed models outperformed the traditional method, hence proving their efficiency.

The integration of Photovoltaic (PV) systems into grid has a detrimental effect on grid stability, dependability, reliability, efficiency, economy, planning and scheduling. Thus, a reliable PV output prediction is necessary for grid stability. This paper presents a detailed review on PV power forecasting technique. A detailed evaluation of forecasting techniques reveals that solar irradiance is highly correlated with PV output which makes it mandatory to classify the weather as well as analyze the cloudy motion. The researchers in literature dominantly preferred very short term and short-term horizon for PV forecasting. Different PV forecasting models categorized into physical, persistence, statistical, machine learning and hybrid model used in literature are studied and evaluated using various performance metrics such as root mean square error and mean absolute error for accuracy. A thorough comparison of various forecasting techniques suggests that deep neural networks with ensemble technique or hybrid techniques supersedes the traditional approaches of PV power forecasting in terms of efficiency and accuracy. Graphical abstract

Accurate short-term solar radiation forecasting is essential for the reliable integration of photovoltaic systems into modern power grids, particularly in regions characterized by strong climatic heterogeneity. This study proposes a Climate-Aware Hybrid Kolmogorov–Arnold Network (CA-HKAN) framework for forecasting hourly Global Horizontal Irradiance (GHI) under diverse atmospheric regimes. The framework integrates an intrinsically interpretable spline-based Kolmogorov–Arnold Network with a feed-forward neural network through a deterministic switching mechanism governed by Extreme Value Theory (EVT). EVT is employed to derive climate-specific clearness-index thresholds, which are scaled to delineate stable and volatile irradiance regimes. These thresholds deterministically activate the interpretable spline-based component under physically stable conditions, while a neural network fallback is engaged during volatile or extreme atmospheric states. The proposed approach is evaluated using hourly meteorological and irradiance data from five climatically distinct regions in Saudi Arabia, representing desert, coastal, mountainous, and transitional environments. Experimental results demonstrate that the proposed CA-HKAN framework achieves predictive accuracy competitive with modern deep learning baselines, such as CNN–BiLSTM models, across all regions while maintaining physical consistency, including non-negativity and realistic irradiance bounds. Compared with standalone models, the hybrid approach offers a favorable balance between accuracy, robustness, and transparency. Ablation analyses further confirm the complementary roles of the hybrid components and the effectiveness of EVT-based regime control. Overall, the CA-HKAN framework provides a practical and interpretable solution for climate-aware solar radiation forecasting, supporting trustworthy deployment in sustainable energy systems operating under heterogeneous and non-stationary atmospheric conditions.

This paper aims to develop the long short-term memory (LSTM) network modelling strategy based on deep learning principles, tailored for the very short-term, near-real-time global solar radiation (GSR) forecasting. To build the prescribed LSTM model, the partial autocorrelation function is applied to the high resolution, 1 min scaled solar radiation dataset that generates statistically significant lagged predictor variables describing the antecedent behaviour of GSR. The LSTM algorithm is adopted to capture the short- and the long-term dependencies within the GSR data series patterns to accurately predict the future GSR at 1, 5, 10, 15, and 30 min forecasting horizons. This objective model is benchmarked at a solar energy resource rich study site (Bac-Ninh, Vietnam) against the competing counterpart methods employing other deep learning, a statistical model, a single hidden layer and a machine learning-based model. The LSTM model generates satisfactory predictions at multiple-time step horizons, achieving a correlation coefficient exceeding 0.90, outperforming all of the counterparts. In accordance with robust statistical metrics and visual analysis of all tested data, the study ascertains the practicality of the proposed LSTM approach to generate reliable GSR forecasts. The Diebold–Mariano statistic test also shows LSTM outperforms the counterparts in most cases. The study confirms the practical utility of LSTM in renewable energy studies, and broadly in energy-monitoring devices tailored for other energy variables (e.g., hydro and wind energy).

Energy communities have recently gained significant attention as local entities that empower neighborhoods to contribute actively to the clean energy transition by adopting solar energy. However, the variability of weather conditions makes PV energy production highly unpredictable, emphasizing the need for accurate prediction and forecasting to ensure efficient operation and balance supply and demand. This study investigates the use of machine learning models to predict PV energy generation from multiple household rooftop photovoltaic (PV) systems within an energy community, with solar irradiance serving as the sole input parameter. Furthermore, various deep learning architectures were also explored to forecast solar radiation and determine the optimal model configuration. The results show that the Random Forest model performed better than the other models tested, achieving the lowest error metrics for PV energy prediction. For solar radiation forecasting, the GRU model demonstrates good performance compared the other models.

This study proposes hybrid solar radiation temporal modeling approaches to support the design of clean energy systems using deep learning techniques and statistical distribution fitting. Solar radiation data are analyzed using a probability distribution to determine whether they follow a known statistical pattern, focusing on total solar radiation on a tilted surface (MJ/m 2 ) ( ). Maximum likelihood estimation (MLE), whale optimization algorithm (WOA), and particle swarm optimization (PSO) are used to optimize the process of estimating probability distribution parameters. Subsequently, the cumulative distribution function (CDF) is constructed, and a particular distribution profile is applied to replace the inherent randomness in data during the preparation phase of estimation model inputs. In the next step, innovative hybrid temporal modeling approaches based on CDF are developed using long short-term memory networks (LSTMs), gated recurrent units (GRUs), and extreme gradient boosting (XGBoost) algorithms. Model results are evaluated through Jensen-Shannon divergence (JSD) analysis. Thus, the DOD-Boost framework is established. According to the findings from comprehensive analyses, DOD-Boost models that integrated a modeling approach for , optimization techniques, data preprocessing strategies, and temporal modeling achieved highly accurate predictions. Among all tested models, the Weibull (WOA) – LSTM – XGBoost model achieved the best distributional accuracy, with the lowest JSD value of 0.0084. The JSD metric was prioritized as it provides a more comprehensive assessment of performance by measuring the similarity of the predicted and actual data distributions, which is more informative than simple point predictions for energy planning. Consequently, this study provides a transferable hybrid model for PV-based energy planning that can also be used in developing countries.

Accurate ultra-short-term solar radiation forecasting is critical for renewable energy integration and power grid stability, yet operational systems exhibit systematic biases and accumulating errors during transitional weather conditions. This research develops an integrated error correction framework that synergistically combines multimodal vision transformers with physics-constrained neural networks to refine satellite-based solar forecasts. The proposed architecture extracts complementary spatiotemporal features from visible, infrared, and water vapor satellite channels through hierarchical cross-modal attention mechanisms, while simultaneously enforcing fundamental physical principles including energy conservation and radiative transfer constraints as soft regularization terms during training. Experimental validation using four years of Himawari-8 geostationary satellite observations and ground measurements from 47 meteorological stations demonstrates that the framework achieves 18.7% RMSE reduction compared to best-performing baselines, with systematic bias reduced from 12.7 W/m² to 1.2 W/m². Ablation studies reveal synergistic interactions between multimodal fusion and physics-aware learning, with the combined approach delivering benefits exceeding individual components. The model maintains computational efficiency suitable for operational deployment, processing 21 forecasts per second on consumer-grade hardware while respecting radiative energy consistency across all predictions.

Recently, the increasing prevalence of solar energy in power and energy systems around the world has dramatically increased the importance of accurately predicting solar irradiance. However, the lack of access to data in many regions and the privacy concerns that can arise when collecting and transmitting data from distributed points to a central server pose challenges to current predictive techniques. This study proposes a global solar radiation forecasting approach based on federated learning (FL) and convolutional neural network (CNN). In addition to maintaining input data privacy, the proposed procedure can also be used as a global supermodel. In this paper, data related to eight regions of Iran with different climatic features are considered as CNN input for network training in each client. To test the effectiveness of the global supermodel, data related to three new regions of Iran named Abadeh, Jarqavieh, and Arak are used. It can be seen that the global forecasting supermodel was able to forecast solar radiation for Abadeh, Jarqavieh, and Arak regions with 95%, 92%, and 90% accuracy coefficients, respectively. Finally, in a comparative scenario, various conventional machine learning and deep learning models are employed to forecast solar radiation in each of the study regions. The results of the above approaches are compared and evaluated with the results of the proposed FL-based method. The results show that, since no training data were available from regions of Abadeh, Jarqavieh, and Arak, the conventional methods were not able to forecast solar radiation in these regions. This evaluation confirms the high ability of the presented FL approach to make acceptable predictions while preserving privacy and eliminating model reliance on training data.