Solar Radiation Prediction Research Articles

IoT and machine learning models for multivariate very short‐term time series solar power forecasting

AbstractIn solar energy generation, the inherent variability caused by cloud cover and weather events often leads to sudden fluctuations in power outputs. Addressing this challenge, the authors’ study focuses on very short‐term solar irradiance (SI) prediction. Leveraging multivariate time series datasets, the authors improve very short‐term SI predictions. To achieve accurate very short‐term SI predictions, the authors employ machine learning techniques throughout the forecasting process. Additionally, the authors’ work pioneers the integration of the Internet of Things (IoT) into solar power systems, a novel approach in the field. The authors leverage LoRa (long range) technology for low‐cost, low‐power, and long‐range wireless control networks. The authors’ study focuses on SI forecasting using long short‐term memory and bi‐directional long short‐term memory (Bi‐LSTM) models, achieving high accuracy. The SI forecasts are evaluated in terms of root‐mean‐square error (RMSE) and mean absolute error in relation to meteorological data and sky image data. The improvement in performance can be attributed to the Bi‐LSTM's bidirectional nature, allowing it to incorporate future information during training, thereby enhancing its predictive capabilities. Overall, the results suggest that the Bi‐LSTM model is more suitable for accurately forecasting SI, particularly in scenarios requiring short‐term predictions based on rapidly changing environmental factors.

A radiant shift: Attention-embedded CNNs for accurate solar irradiance forecasting and prediction from sky images

The continuous increase in solar power integration with energy systems can be attributed to the push for cleaner energy use globally, highlighting the importance of accurate solar forecasts. Conventional prediction methods, although valuable, often fall short of delivering the precision required for dynamic energy management systems due to their inability to effectively capture the intricate relationships inherent in solar irradiance variations. This research addresses this limitation by introducing a novel approach that harnesses the potential of Attention-embedded Convolutional Neural Networks (ATT_CNN) to forecast Global Horizontal Irradiance (GHI), Direct Normal Irradiance (DNI), and Diffuse Horizontal Irradiance (DHI) for intra-hour solar forecasting utilizing sequences of sky images. The main contribution of this study is the integration of attention mechanisms into the CNN architecture, strategically designed to enhance their efficacy in predicting GHI, DNI, and DHI by fostering an adaptive focus on sky image features. First, we use SRRL dataset with GHI, DNI, and DHI features as predicting labels, then the Mean Bias Error (MBE), Root Mean Square Error (RMSE), and Forecasting RMSE skill score (FSS) are used to evaluate the model's performance. This study utilized six lead times and four sequence lengths, based on this the best combination is: Sequence length: 4 with Minutes: 20. This combination provides a balanced and optimal performance with low RMSE (62.75 W/m2), low MBE (2.71 W/m2), and a high FSS (38.81), indicating good accuracy, minimal bias, and high skill score. Furthermore, when compared with state-of-the-art models, the proposed model yielded superior results. This advancement holds profound implications for the optimization of solar power utilization within mainstream energy systems, further underscoring the significance of cutting-edge deep learning (DL) techniques in advancing sustainable energy technologies. The findings of this study indicate that integrating attention mechanisms is essential to enhance the accuracy and reliability of forecasts, and using longer sequences of images can further improve forecasting performance.

AI-Driven Solar Energy Generation and Smart Grid Integration A Holistic Approach to Enhancing Renewable Energy Efficiency

This paper comprehensively analyzes AI-driven solar energy generation and smart grid integration, focusing on enhancing renewable energy efficiency. The study examines applying advanced artificial intelligence techniques in optimizing solar power production, forecasting, and grid management. Machine learning algorithms, including Support Vector Regression (SVR) and Artificial Neural Networks (ANN), are evaluated for effectiveness in solar irradiance prediction and PV system performance estimation. The integration of AI in smart grids is explored, highlighting its role in demand-side management, energy storage optimization, and grid stability control. A holistic approach to improving renewable energy efficiency is proposed, encompassing integrated AI frameworks for solar-plus-storage systems, multi-objective optimization techniques for energy management, and AI-enabled microgrids and virtual power plants. The paper also addresses the challenges and future trends in AI application to renewable energy systems, including scalability issues, regulatory considerations, and ethical implications. By leveraging big data analytics and advanced AI algorithms, this research demonstrates the potential for significant improvements in overall system efficiency, reliability, and sustainability of solar energy systems integrated with smart grids.

Forecasting of global solar radiation: A statistical approach using simulated annealing algorithm

In recent years, the share of renewable energy sources in current energy production has been increasing due to the depletion of fossil fuel resources, increasing energy needs and environmental concerns. Solar energy, one of the most important renewable energy sources depending on being clean, sustainable and environmentally friendly energy source. Understanding the solar radiation value is crucial for maximizing the potential of solar energy and ensuring the efficient operation of solar energy systems. In this paper, it is aimed to develop a new global solar radiation (GSR) prediction model by using simulated annealing algorithm (SAA). Modeling solar radiation data with significant variability is a challenging task that necessitates the use of nonlinear approaches. Although the SAA is widely used in engineering and natural sciences, it has not previously been used to develop a GSR prediction model. While developing the SAA forecast model for the Adana region, long-term sunshine duration and solar radiation data obtained from the Turkish General Directorate of State Meteorology were used, as well as geographical features such as latitude and longitude of the selected region. The performance and applicability of the proposed model was examined by comparing it with different GSR estimation methods developed in the literature. The primary contribution of this study lies in the introduction of an inventive model, which has been constructed for the first time in the assessment of GSR, to the current collection of literature. The produced results were statistically compared with the observed data using six separate performance and error measures. The results of all years showed that the relative percentage error (RPE) less than 11.66 %, the mean percentage error (MPE) does not exceed 12.55 %, the correlation coefficient (R2) greater than 0.98, the mean absolute percentage error (MAPE) test results appear to vary close to the value 10, the sum of squared relative error (SSRE) test results seem to approaching 0 and the t-statistic test less than 4.06 for the annual GSR. It has been observed that the developed SAA GSR forecasting model has a successful performance when compared to different forecasting models popularly used in the literature. According to the statistical error results, the estimated GSR data from the novel SAA-GSR forecast model aligns well with the measured meteorological values. Given the efficacy of the established SAA model in GSR estimation, this study is anticipated to offer valuable insights and contributions to the application of the SAA approach in other energy applications.

Application of Quantum Neural Network for Solar Irradiance Forecasting: A Case Study Using the Folsom Dataset, California

Merging machine learning with the power of quantum computing holds great potential for data-driven decision making and the development of powerful models for complex datasets. This area offers the potential for improving the accuracy of the real-time prediction of renewable energy production, such as solar irradiance forecasting. However, the literature on this topic is sparse. Addressing this knowledge gap, this study aims to develop and evaluate a quantum neural network model for solar irradiance prediction up to 3 h in advance. The proposed model was compared with Support Vector Regression, Group Method of Data Handling, and Extreme Gradient Boost classical models. The proposed framework could provide competitive results compared to its competitors, considering forecasting intervals of 5 to 120 min ahead, where it was the fourth best-performing paradigm. For 3 h ahead predictions, the proposed model achieved the second-best results compared with the other approaches, reaching a root mean squared error of 77.55 W/m2 and coefficient of determination of 80.92% for global horizontal irradiance forecasting. The results for longer forecasting horizons suggest that the quantum model may process spatiotemporal information from the input dataset in a manner not attainable by the current classical approaches, thus improving forecasting capacity in longer predictive windows.

Prediction of Global Solar Irradiance on Parallel Rows of Tilted Surfaces Including the Effect of Direct and Anisotropic Diffuse Shading

Prediction of Global Solar Irradiance on Parallel Rows of Tilted Surfaces Including the Effect of Direct and Anisotropic Diffuse Shading

Solar irradiance prediction with variable time lengths and multi-parameters in full climate conditions based on photovoltaic greenhouse

Photovoltaic power generation can provide energy for greenhouses and achieve high quality and high yield of crops. In reality, solar irradiance is fluctuating and intermittent. Thus, the key to ensure efficient photovoltaic power use under greenhouse environmental conditions is to provide an accurate prediction of solar irradiance. Yet, currently studies on irradiance prediction in terms of the variable time lengths, multi-parameter and full climate conditions are rather limited. To improve the comprehensive performance of the prediction model, this paper proposes models of irradiance time series prediction such as Pyraformer, Informer, Transformer and TimesNet. Those models were tested based on the synergistic combination of weather conditions (WC), sunshine time accumulation (STA/h), instantaneous total irradiance (ITI/(W/m2)) and irradiance daily accumulation (IDA/(MJ/m2)). The model performance was rigorously evaluated with 9 prediction lengths, 5 training days, 4 seasons and 5 days for 9 weather conditions. The results showed that the Transformer model had the best overall prediction performance for STA, ITI, IDA and WC at different time steps of 10 min to 24 h. All models were suitable for predicting time series within 1 h. However, TimesNet model was not suitable for predicting time series with steps outside 1 h. On the other hand, by using the sun combination dataset, the Transformer model had the best performance at a time step of 10 min. The mean absolute error (MAE), mean-square error (MSE), root mean squared error (RMSE) and coefficient of determination (R2) of ITI were 0.118 W/m2, 0.059 W/m2, 0.243 W/m2 and 93.9 %, respectively. When exploring the minimum dataset, with the increase of data samples, the prediction effect of TimesNet showed an increasing trend. While, Transformer had the best prediction effect for the dataset with one year of use. When exploring seasonality, Pyraformer model had the best prediction effect on winter and summer, and TimesNet had the best prediction effect on autumn and spring. Local prediction of 9 climate conditions showed that the effects of snow and dust storm were not ideal. The research results showed that the characterization factors that were closely linked to irradiance. The prediction scheme proposed in this paper combined the advantages of different time steps, different factor combination datasets, different data volumes and seasonality, which greatly improves the generalization ability of the model. This study can provide a reference for irradiance prediction and more refined management of photovoltaic (PV) greenhouse.

Advancements and Challenges in Solar Radiation Prediction: A Review of Machine Learning Approaches

Solar irradiance prediction holds paramount importance in optimizing the efficacy and dependability of solar power systems. This evaluative manuscript delves into an array of machine learning methodologies applied in solar irradiance prediction, encompassing conventional methodologies. It further delves into the ramifications of meteorological data, geographical and temporal parameters, data preprocessing, feature curation methodologies, amalgamated learning methodologies, and composite models. Moreover, the manuscript investigates the latent advantages of amalgamating orbital visuals and terrestrial sensors to augment prognostic capacities. Ultimately, the review underscores the exigency for broader datasets, refined feature curation methodologies, and composite models to attain heightened precision and resilience.

Hybrid model for intra-day probabilistic PV power forecast

Renewable generation forecasting has become an essential element of power system management because of the ever-increasing share of renewable power fed into the grid. Solar energy is one of the most common and well-known sources of energy in existing networks. Due to its intermittent and non-linear nature, accurate prediction of solar irradiance is essential for ensuring the reliable integration of photovoltaic (PV) plants with the grid system and for effectively managing supply and demand. The dispatching problem is typically addressed by generating a forecast for a few sample plants and scaling the result of these forecasts by the PV power generation capacity connected to the electric system. This research presents a comprehensive comparison of various methods for predicting PV power generation in fixed and single axis tracking PV systems at two different locations. These methods include time-series statistical regression, parametric physical approaches, machine learning (ML), and ensemble techniques, aimed at intra-day forecasts ranging from 1 to 6 h ahead. The study introduces a novel method for developing a parametric linear model for fixed PV systems and uses model output statistics (MOS) to improve forecast accuracy. Additionally, this study details various optimization algorithms designed to refine model parameters for creating a hybrid model. It also examines key factors influencing PV power forecasts and applies post-processing Kalman filter techniques. The study provides a detailed analysis of deterministic and probabilistic forecasting models, offering not just the most probable power production estimate, but also conveying the associated uncertainty. The results demonstrate a highly skilled ML based hybrid model to forecast PV power for different technologies at two different locations.

Unlocking renewable energy potential: Harnessing machine learning and intelligent algorithms

This review article examines the revolutionary possibilities of machine learning (ML) and intelligent algorithms for enabling renewable energy, with an emphasis on the energy domains of solar, wind, biofuel, and biomass. Critical problems such as data variability, system inefficiencies, and predictive maintenance are addressed by the integration of ML in renewable energy systems. Machine learning improves solar irradiance prediction accuracy and maximizes photovoltaic system performance in the solar energy sector. ML algorithms help to generate electricity more reliably by enhancing wind speed forecasts and wind turbine efficiency. ML improves the efficiency of biofuel production by optimizing feedstock selection, process parameters, and yield forecasts. Similarly, ML models in biomass energy provide effective thermal conversion procedures and real-time process management, guaranteeing increased energy production and operational stability. Even with the enormous advantages, problems such as data quality, interpretability of the models, computing requirements, and integration with current systems still remain. Resolving these issues calls for interdisciplinary cooperation, developments in computer technology, and encouraging legislative frameworks. This study emphasizes the vital role of ML in promoting sustainable and efficient renewable energy systems by giving a thorough review of present ML applications in renewable energy, highlighting continuing problems, and outlining future prospects

Multi-site solar irradiance prediction based on hybrid spatiotemporal graph neural network

Constructing accurate spatiotemporal correlations is a challenging task in joint prediction of multiple photovoltaic sites. Some advanced algorithms for incorporating other surrounding site information have been proposed, such as graph neural network-based methods, which are usually based on static or dynamic graphs to build spatial dependencies between sites. However, the possibility of the simultaneous existence of multiple spatial dependencies is not considered. This paper establishes a spatiotemporal prediction model based on hybrid spatiotemporal graph neural network. In this model, we apply adaptive hybrid graph learning to learn composite spatial correlations among multiple sites. A temporal convolution module with multi-subsequence temporal data input is used to extract local semantic information to better predict future nonlinear temporal dependencies. A spatiotemporal adaptive fusion module is added to address the issue of integrating diverse spatiotemporal trends among multiple sites. To assess the model's predictive performance, nine solar radiation observation stations were selected in two different climatic environments. The average root mean square error (RMSE) of the constructed model was 38.51 and 49.90 W/m2, with average mean absolute error (MAE) of 14.72 and 23.06 W/m2, respectively. Single-site and multi-site prediction models were selected as baseline models. Compared with the baseline models, the RMSE and MAE reduce by 3.1%–20.8% and 8.9%–32.8%, respectively, across all sites. The proposed model demonstrates the effectiveness of improving accuracy in forecasting solar irradiance through multi-site predictions.

Prediction of Global Solar Radiation across the Region of Northern Sudan

Available data for solar resources indicates that northern Sudan has solar potentials among the highest in world. However, in this vast area (about 471×103 km2) only two stations, at Hudeiba and Dongola, have records for global solar radiation. Four other stations (at Atbara, Karima, Abu-Hamed and Wadi-Halfa) in this region have records for sunshine duration. In this study, recorded data at Hudeiba and Dongola stations have been used to develop 15 models for estimation of monthly average daily global solar radiation on a horizontal surface in northern Sudan. The developed models have been extensively evaluated by using 14 statistical indicators. All developed models revealed good performance (percentage errors within ±4.5%). Then, the most accurate regional model (Model M3) has been used to predict the global solar radiation at Atbara, Karima, Abu-Hamed and Wadi-Halfa stations. Results showed that the whole region receives abundant solar energy throughout the year. The yearly average of daily global solar radiation over the entire region is 23.16 MJ/m2. Furthermore, the developed models can be used to predict global radiation at adjacent places that lack such data.

Meta-Learning Guided Weight Optimization for Enhanced Solar Radiation Forecasting and Sustainable Energy Management with VotingRegressor

Solar radiation prediction plays a crucial role in renewable energy management, impacting various decision-making processes aimed at optimizing the utilization of solar resources and promoting sustainability. Ensemble regression methods, notably VotingRegressor, have emerged as promising tools for accurate solar radiation forecasting. By integrating predictions from multiple base estimators, ensemble methods have the potential to capture intricate patterns inherent in solar radiation data. However, achieving optimal predictive performance with ensemble methods heavily relies on the careful weighting assigned to each base estimator, presenting a significant challenge. In this study, a novel approach is presented to enhance solar radiation prediction by utilizing meta-learning techniques to optimize the weighting mechanism in the VotingRegressor ensemble. Meta-learning, a subfield of machine learning focusing on learning algorithms across different tasks, provides a systematic framework for learning to learn. This enables models to adapt and generalize more effectively to new datasets and tasks. Our proposed methodology demonstrated significant improvements, with the VotingRegressor with meta-learning techniques achieving an RMSE of 8.7343, an MAE of 5.42145, and an R² of 0.991913. These results mitigate the need for manual weight tuning and improve the adaptability of the VotingRegressor to varying solar radiation conditions, ultimately contributing to the sustainability of renewable energy systems. The methodology involves a comprehensive exploration of meta-learning techniques, encompassing gradient-based optimization, reinforcement learning, and Bayesian optimization.

Bionic fusion perspective: Audiovisual-motivated integration network for solar irradiance prediction

Accurate and reliable prediction of solar irradiance (SI) is an important requirement to develop solar energy while a challenging task due to stochastic and nonlinear data characteristics. Additionally, most deep networks show powerful prediction capabilities but lack the supports from biological science, reflecting that bionically-inspired networks in SI analysis are still not enough explored. To this end, this paper proposes an audiovisual-motivated Transformer-CNN integration network, called ATI-net, for predicting SI. The audiovisual cognition gives a superior design framework for ATI-net with signal capture, signal analysis, and prediction blocks. In the first block, through mimicking the function of both eye and ear in external signal conversion, multi-scale features are extracted by incorporating multi-branch convolutions with varying kernels, where the Mish function addresses the problem that traditional ReLU function stops learning when the input is negative. In the second block, through mimicking the function of left and right hemispheres in neuronal signal analysis, two structures triggered by Transformers and convolutions are designed to remember temporal evolutionary rules, where residual connections are beneficial to mine deep information and avoid forgetting. In the third block, through mimicking the function of a higher brain region in generating understanding, the above information is integrated to make the SI prediction. Besides, the nonlinear dependencies and linear relationships are independently extracted and integrated into the ATI-net, which not only reduces information interference but is consistent with the “divide and conquer” idea. Experimental results show that the ATI-net outperforms 18 benchmarks, and average improvements of root mean squared error (RMSE) are 26.28% and 26.01% for two datasets, respectively. In summary, the ATI-net is one of the reliable alternatives to SI prediction.

Prediction of daily global solar radiation in different climatic conditions using metaheuristic search algorithms: a case study from Türkiye

Today’s many giant sectors including energy, industry, tourism, and agriculture should closely track the variation trends of solar radiation to take more benefit from the sun. However, the scarcity of solar radiation measuring stations represents a significant obstacle. This has prompted research into the estimation of global solar radiation (GSR) for various regions using existing climatic and atmospheric parameters. While prediction methods cannot supplant the precision of direct measurements, they are invaluable for studying and utilizing solar energy on a global scale. From this point of view, this paper has focused on predicting daily GSR data in three provinces (Afyonkarahisar, Rize, and Ağrı) which exhibit disparate solar radiation distributions in Türkiye. In this context, Gradient-Based Optimizer (GBO), Harris Hawks Optimization (HHO), Barnacles Mating Optimizer (BMO), Sine Cosine Algorithm (SCA), and Henry Gas Solubility Optimization (HGSO) have been employed to model the daily GSR data. The algorithms were calibrated with daily historical data of five input variables including sunshine duration, actual pressure, moisture, wind speed, and ambient temperature between 2010 and 2017 years. Then, they were tested with daily data for the 2018 year. In the study, a series of statistical metrics (R2, MABE, RMSE, and MBE) were employed to elucidate the algorithm that predicts solar radiation data with higher accuracy. The prediction results demonstrated that all algorithms achieved the highest R2 value in Rize province. It has been found that SCA (MABE of 0.7023 MJ/m2, RMSE of 0.9121 MJ/m2, and MBE of 0.2430 MJ/m2) for Afyonkarahisar province and GBO (RMSE of 0.8432 MJ/m2, MABE of 0.6703 MJ/m2, and R2 of 0.8810) for Ağrı province are the most effective algorithms for estimating GSR data. The findings indicate that each of the metaheuristic algorithms tested in this paper has the potential to predict daily GSR data within a satisfactory error range. However, the GBO and SCA algorithms provided the most accurate predictions of daily GSR data.

Temporal Evolution, Oscillation and Coherence Characteristics Analysis of Global Solar Radiation Distribution in Major Cities in China’s Solar-Energy-Available Region Based on Continuous Wavelet Transform

The majority of the energy required for human survival is derived either directly or indirectly from solar radiation, thus it is important to investigate the periodic fluctuations in global solar radiation over time. In this study, six cities—Harbin, Shenyang, Beijing, Shanghai, Wuhan, and Guangzhou—located in the utilizable areas of solar energy in China, were selected, and the periodicity of the daily global solar radiation of these six cities with time were investigated by means of wavelet power spectrum analysis. Furthermore, Harbin, which has the lowest monthly average of global solar radiation in the cold of winter, was selected to explore the temporal evolution relationship between global solar radiation and other meteorological factors, and air quality factors by wavelet coherence analysis. Among the meteorological factors, the correlation between global solar radiation and sunshine duration exhibits the highest level of consistency. Global solar radiation demonstrates a positive association with atmospheric temperature/wind speed/precipitation over the annual cycle. Conversely, it displays a negative correlation with atmospheric pressure during this time frame. Additionally, on shorter timescales, global solar radiation is negatively correlated with air humidity and precipitation. Among air quality factors, the relationship between global solar radiation and CO/NO2/O3/PM2.5/PM10/SO2 primarily manifests in the broader annual cycle time scale. Within this context, global solar radiation shows a positive correlation with O3, while displaying negative associations with the other five air quality indicators. These findings contribute to urban environmental planning and the effective management and utilization of solar radiation, thereby providing valuable insights to guide the future development of cross-regional comprehensive energy utilization projects under diverse climatic and geographical conditions. Additionally, they serve as a reference for subsequent research aimed at enhancing the accuracy of global solar radiation prediction models.

Dual-branch deep learning architecture for enhanced hourly global horizontal irradiance forecasting

Solar energy, a pivotal renewable resource, faces challenges posed from the inherent randomness and volatility of solar irradiance. Effective control and utilization of solar energy systems rely on the accuracy of solar irradiance prediction models. Temporal data of solar irradiance often involves a mixture of long-term patterns represented by periodicity and short-term patterns represented by nonlinearity, for which traditional approaches are difficult to balance. In this study, a novel deep learning approach employing a dual-branch architecture is proposed to forecast hourly global horizontal irradiance. The proposed model comprises two parallel branches, namely the global temporal extractor and the local temporal extractor. Two branches effectively capture complex combinations of global and local temporal patterns from multivariable input data. To enhance the effectiveness of feature extraction, the model incorporates feature-temporal attention modules and self-attention modules within the extractors, which can re-calibrate feature weights and emphasize crucial information. To strengthen the model's robustness, an autoregressive linear model is integrated in parallel, compensating for nonlinear output. The proposed model has been validated on multiple public datasets. Experimental results show the performance of the proposed model has been greatly improved, in which achieves a remarkable 41.76% improvement of forecasting accuracy over baseline models. The superiority the proposed model is also underscored through comparative analysis against state-of-the-art methods.

Optimal Operation of Cogeneration Power Plant Integrated with Solar Photovoltaics Using DLS-WMA and ANN

Focusing on mitigating global challenges arising from hydrocarbon-based sources, the integration of cogeneration power plants with solar photovoltaics offers a viable solution. The intermittent nature of renewable resources presents a challenge to the consistent performance of cogeneration systems. To address these issues, this work introduces a novel framework for integrating cogeneration power plants (CGPPs) with solar photovoltaic systems. The key innovation of this research lies in its dual-algorithm approach that seamlessly blends cogeneration power plants with solar photovoltaic. This study proposed an integrated approach, employing the Derivative Log Sigmoid-Woodpecker Mating Algorithm (DLS-WMA) and Optimized Artificial Neural Networks (O-ANN), to combine cogeneration power plants with solar photovoltaics in industrial distribution systems. The methodology is aimed at achieving a cost-effective, efficient system design, enhancing the efficiency of cogeneration power plants, and introducing energy storage batteries for uninterrupted power generation under diverse atmospheric conditions and loads. Additionally, the proposed system includes rechargeable batteries for energy storage to support critical services when the solar plant is offline and the CGPP cannot meet the power demand. The industrial system’s photovoltaic component is tuned using the DLS-WMA for cost minimization and O-ANN for solar irradiance prediction, ensuring continuous power flow by optimizing both the photovoltaic system and the cogeneration power plant (CGPP) system. Real-time datasets are used to compare the results obtained by this new approach with those of the previous state-of-the-art algorithms. The error with O-ANN prediction is 1.2%, compared to 4.1% with the existing WMA-ANN technique, while the cost-benefit with DLS-WMA shows a 9% improvement over the WMA-ANN technique. The experimental outcomes demonstrate the efficiency of this new approach. Collaboration with industry stakeholders and policymakers is crucial for the large-scale deployment of this system, facilitating the adoption of sustainable energy practices in industrial distribution systems.

An integrated skip convolutional network with residual learning and feature extraction for point and interval prediction of solar radiation

Spinning reserve based on solar radiation prediction can ensure the secure operation of large-scale grid-connected photovoltaic systems, but irrational spinning reserve can lead to substantial economic losses and even grid collapse. Therefore, it is imperative to identify characteristic patterns of solar radiation and establish an effective solar radiation prediction model. However, existing hybrid models often struggle to efficiently recognize and leverage input features, compromising the robustness of the model. To address this limitation, this study proposes an integrated skip-convolutional network with residual learning and feature extraction (InSCNet), which enhances the capability to represent information data and thereby improves the stability and accuracy of solar radiation prediction by employing a series of feature extraction and prediction blocks with residual learning. InSCNet includes a sophisticated feature extraction module to effectively address the underutilization of feature information, and it incorporates residual learning, skip-convolution, and recursive networks in the prediction module, reducing the risk of gradient vanishing or gradient explosion while enhancing prediction accuracy. Additionally, an interval estimation method with adaptively chosen distributions is introduced, which extends the prediction interval estimation method. The proposed InSCNet is rigorously evaluated using datasets from three major cities in Pakistan. Experimental results demonstrate that InSCNet outperforms existing solutions in both point and interval predictions.

A hybrid graph attention network based method for interval prediction of shipboard solar irradiation

Solar energy ships came into being to reduce carbon emissions of global shipping and fossil fuels consumption. Accurate prediction of photovoltaic power generation can improve the economical operation of solar energy ships and reduce the power fluctuation of ship power systems. This study proposes a novel hybrid method to forecast ultra-short-term solar irradiation in a solar energy ship along the east coast of China. An improved graph attention framework is trained to describe the dynamic spatial-temporal topology among five weather stations and a solar energy ship. Initial interval prediction results are then determined by an improved Divided Period Optimization (DPO) method. According to distribution characteristics of the prediction error of shipboard solar irradiation, a Dynamic Prediction Interval Modification (DPIM) method is proposed to further optimize the prediction accuracy and interval width of ultra-short-term solar irradiation. Comparison experiments show that the dynamic connection matrix effectively improves the prediction accuracy, and the DPIM method reduces the PI width of 28.84%–37.79 % in shipboard solar irradiation. The proposed prediction method can accurately predict the shipboard solar irradiation at the period of significant disturbance, and provide effective technical support for the economic scheduling of solar energy ships.