Solar Radiation Model Research Articles

Comparison of optical spectra between asteroids Ryugu and Bennu: I. Cross calibration between Hayabusa2/ONC-T and OSIRIS-REx/MapCam

Asteroids (162173) Ryugu and (101955) Bennu observed by Hayabusa2 and Origins, Spectral Interpretation, Resource Identification, and Security-Regolith Explorer (OSIRIS-REx) share many global properties, but high-spatial-resolution spectral observations by the telescopic Optical Navigation Camera (ONC-T) and MapCam detected subtle but significant differences (e.g., opposite space weathering trends), which may reflect differences in their origin and evolution. Comparing these differences on the same absolute scale is necessary for understanding their causes and obtaining implications for C-complex asteroids. However, ONC-T and MapCam have a large imager-to-imager systematic error of up to 15% caused by the difference in radiometric calibration targets. To resolve this problem, we cross calibrated albedo and color data between the two instruments using the Moon as the common calibration standard. The images of the Moon taken by ONC-T and MapCam were compared with those simulated using photometry models developed from lunar orbiter data. Our results show that the cross-calibrated reflectance of Ryugu and Bennu can be obtained by upscaling the pre-cross-calibrated reflectance of Bennu by 13.3 ± 1.6% at b band, 13.2 ± 1.5% at v band, 13.6 ± 1.7% at w band, and 14.8 ± 1.8% at x band, while those for Ryugu are kept the same. These factors compensate for the imager-to-imager bias caused by differences in targets used for radiometric calibration and solar irradiance models used for data reduction. Need for such large upscaling underscore the importance of using the cross-calibrated data for accurately comparing the Ryugu and Bennu data. The uncertainty in these factors show that the reflectance of Ryugu and Bennu can be compared with <2% accuracy after applying our results. By applying our cross calibration, the geometric albedo of Bennu became consistent with those observed by ground-based telescopes and the OSIRIS-REx Visible and InfraRed Spectrometer (OVIRS). Our result can be simply applied by multiplying a constant to the publicly available data and enables accurate comparison of the optical spectra of Ryugu and Bennu in future studies.

A new evolutionary forest model via incremental tree selection for short-term global solar irradiance forecasting under six various climatic zones

The increasing integration of solar sources into the energy mix presents significant challenges, particularly in short-term energy management. Accurate solar irradiance forecasts can greatly assist solar power plant operators and energy network managers in making informed decisions about energy production and consumption. This paper aims to develop a new accurate forecasting model for short-term global solar irradiance based on an innovative evolutionary forest approach. Our model, baptized EFITS, performs incremental tree selection through appropriate evolutionary operators maintaining a good tradeoff between accuracy and diversity, generating progressively near-optimal decision trees to construct the final evolutionary forest forecaster. This new evolution process also automatically selects near-optimal input parameters, enhancing the overall model accuracy and generalization ability. Six climatically diverse locations in Morocco and three types of inputs (endogenous, exogenous, and hybrid) are used to assess the performance of the proposed. The results demonstrate that our proposed model exhibits excellent performance across all studied sites and horizons. Among all input types, hybrid inputs delivered the best forecasting accuracy across all studied sites and horizons. Notably, the continental climate site (Bni Mellal) achieved the highest accuracy, with nRMSE ranging from 4.94% to 7.54% and nMBE from 0.71% to −0.46% for 1 to 6 h forecasts. Conversely, Ifrane city, characterized by a humid temperate climate, showed the lowest accuracy, with nRMSE ranging from 10.34% to 18.94% and nMBE from 1.21% to −1.54%. Finally, a detailed comparison with benchmarking models (random forest, bagging, gradient boosting, single decision tree, bidirectional long short-term memory network, and scaled persistence models), revealed that our model consistently outperforms them across all tested scenarios, locations, and forecasting horizons.

Preflight Spectral Calibration of the Ozone Monitoring Suite-Nadir on FengYun 3F Satellite

The Ozone Monitoring Suite-Nadir (OMS-N) instrument is the first hyperspectral remote sensor in the ultraviolet band of China’s Fengyun series satellites. It can be used to detect several kinds of atmospheric constituents. This paper describes the prelaunch spectral calibration of the OMS-N onboard FengYun 3F. Several critical spectral parameters including the spectral resolution, spectral dispersion, and the instrument spectral response function were determined through laser-based measurements. A secondary peak of the instrument spectral response function from the short wavelength side of the ultraviolet band was found, and the possible influence on data applications was analyzed using a reference solar model and radiative transfer model. The results indicate that the spectral resolution and spectral accuracy of OMS-N meet the mission requirements. However, the asymmetries in the instrument spectral response function in the ultraviolet band were found near nadir rows, which are expressed as the “asymmetric central peak” and “secondary peak”. The analysis results show that if the influences of the instrument spectral response function “asymmetric central peak” and “secondary peak” in the ultraviolet band are ignored, they will bring an error as large as 5% at the center of the absorption line.

On-Grid Photovoltaic (PV) - Battery - PLN for Smart Home System

Electricity is one of basic human needs. However, PLN's ability to meet customer demands is hampered by its limitations. On the other hand, the sunny geographical advantage of Makassar city can be utilized as a new renewable and environmentally friendly energy source in a smart home. Smart house is a family residence that is able to synergize electricity usage based on the habits of its residents with the help of smart technology so that comfort, safety and efficiency of using electrical energy are obtained. The utilization of solar cell hybrid power – battery – PLN can be implemented in addition to meeting the needs of electricity load in the smart home, it can also contribute excess energy to fulfill off-grid building load. Monte Carlo Simulation (MCS) is carried out at the beginning of data processing by randomly generating 24-hour models of solar irradiance and smart home load requirements along with weather conditions. PLN not only takes over fulfilling the needs of the smart home load when there is less and or no sunlight and minimum battery capacity conditions, but also it will charge the battery capacity up to 100% every midnight. On average, the daily load requirement for a smart home is almost half the energy produced by PVs, which are 12,439 kW and 24,509 kW respectively. Furthermore, the smart home hybrid power is capable of producing 8,946 MW of excess energy in a year to serve the off-grid building load needs.

An interpretable framework for modeling global solar radiation using tree-based ensemble machine learning and Shapley additive explanations methods

Machine learning techniques provide an effective and cost-efficient solution for estimating solar radiation for solar energy utilization. However, the reported machine learning-based solar radiation models fail to offer comprehensive explanations for their outputs. Therefore, this study aims to tackle this issue by developing machine learning models that are both accurate and interpretable. To achieve the objective, this study evaluated the performance of tree-based ensemble algorithms, using optimized combinations of model input parameters for different climate zones in China. The results showed that the extreme gradient boosting (XGBoost) models demonstrated the highest overall accuracy, model stability, and generalization ability. At the national scale, the developed XGBoost models yielded an average R2, MAE, and RMSE of 0.939, 1.226 MJ/m2, and 1.663 MJ/m2, respectively, showing significant improvements of 2.13–27.78% in RMSE compared to recently reported models. Most importantly, the state-of-the-art SHapley Additive exPlanations (SHAP) technique was integrated with the developed XGBoost models to enhance model interpretability in terms of global and local feature importance, as well as the interaction effects between model features. The results of the SHAP value analysis demonstrated the robustness of sunshine duration in modeling global solar radiation, revealing thresholds where its values undergo a shift from negative to positive effects on model output. SHAP interaction values illustrated the interaction effects among features in the developed solar radiation model, uncovering the model's complex non-linear relationships. Additionally, this study provided explanations for individual instances based on the SHAP method. Overall, this study provided an accurate, reliable, and transparent machine learning model and an enlightening framework for modeling global solar radiation at sites without observations.

A Life-Long Learning XAI Metaheuristic-Based Type-2 Fuzzy System for Solar Radiation Modeling

A Life-Long Learning XAI Metaheuristic-Based Type-2 Fuzzy System for Solar Radiation Modeling

Optimization Control Strategy for Transition Season Blinds Balancing Daylighting, Thermal Discomfort, and Energy Efficiency

Natural lighting in office buildings is often affected by excessive solar heat gain and discomfort glare, and the transitional seasons are no exception. Therefore, addressing the issue of natural lighting in buildings has always been a challenge in developing shading control strategies. This study designed a model-based calculation method for shading blind control indicators. The method combines existing physical models for sky conditions, solar radiation, heat transfer, thermal comfort, glare, and illuminance calculations and incorporates modifications and additions. Additionally, an equivalent energy-saving model for shading was established. A shading blind control strategy balancing indoor light and the thermal environment with energy savings during transitional seasons was proposed. In transitional seasons, this method can eliminate 100% of glare, reduce discomfort illuminance duration by 81.3% and heat discomfort duration by 87.5% and save 81.3% of lighting energy consumption when air conditioning is used. By comparing the simulation results of an office building in Xi’an with the results obtained from this calculation method, the absolute error percentage was found to be 6.83%, verifying the reliability of the calculation method. Finally, the proposed control strategy was compared with common methods such as no blinds, end-angle control, and fixed-angle control to evaluate its performance in terms of daylighting, thermal comfort, and energy savings. The results showed that the control strategy proposed in this study has significant advantages with respect to various performance indicators.

Accumulative Heat Stress in Ruminants at the Regional Scale under Changing Environmental Conditions

Environmental heat stress is implicated in various animal health issues in ruminants, including reproduction rates, mortality rates, and animal physical quality. During extremely hot weather, there is often no overnight equilibration of animal temperature with its cooling effect, and the accumulated heat load becomes an important factor in animal health for ruminants such as sheep. Using the heat load index (HLI), a heat load model is used as an indicator of heat stress on an hourly basis and annually, using downscaling models for temperature, humidity, solar radiation, and wind speed, in both spatial and temporal cases, across several example sites in regional Victoria. Analysis is provided on the performance of the downscaling models and various adaptation and mitigation options are discussed and tested. These options include using different tree planting patterns to modify solar radiation exposure and wind effects, with mixed results because adding shading structures may also diminish the effect of wind-based cooling. The modelling experiments indicated that (1) heat stress is likely to increase under future climate conditions and could represent a serious threat to the health of small ruminants; (2) adaptation measures by means of tree planting to provide shade may not be sufficient to alleviate projected heat stress; and (3) other adaptation measures will need to be considered. Indicative results for heat stress under potential future environments are provided for 2030, 2050, and 2070. Also discussed is the performance of wind speed modelling, and the effect of heat stress on animal growth and ram fertility.

Validation of a 3D Local-Scale Adaptive Solar Radiation Model by Using Pyranometer Measurements and a High-Resolution Digital Elevation Model.

The result of the multidisciplinary collaboration of researchers from different areas of knowledge to validate a solar radiation model is presented. The MAPsol is a 3D local-scale adaptive solar radiation model that allows us to estimate direct, diffuse, and reflected irradiance for clear sky conditions. The model includes the adaptation of the mesh to complex orography and albedo, and considers the shadows cast by the terrain and buildings. The surface mesh generation is based on surface refinement, smoothing and parameterization techniques and allows the generation of high-quality adapted meshes with a reasonable number of elements. Another key aspect of the paper is the generation of a high-resolution digital elevation model (DEM). This high-resolution DEM is constructed from LiDAR data, and its resolution is two times more accurate than the publicly available DEMs. The validation process uses direct and global solar irradiance data obtained from pyranometers at the University of Salamanca located in an urban area affected by systematic shading from nearby buildings. This work provides an efficient protocol for studying solar resources, with particular emphasis on areas of complex orography and dense buildings where shadows can potentially make solar energy production facilities less efficient.

Comparative analysis of single and hybrid machine learning models for daily solar radiation

This study investigates the estimation of daily solar radiation (SR) through various machine learning (ML) models, including the k-nearest neighbor algorithm (KNN), support vector regression (SVR), and random forest (RF), both individually and in combination with the wavelet transform (WT). The assessment of these models is based on meteorological data spanning three decades (1981–2010) from the province of Kütahya in Türkiye. To address the inherent uncertainty in these data-driven models, the quantile regression method is employed for uncertainty analysis. Statistical metrics, such as mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2), mean prediction interval (MPI), and prediction interval coverage probability (PICP), are utilized to evaluate the effectiveness and uncertainties of the models. The SVR and KNN models exhibit comparable performances concerning both predictive accuracy and uncertainty levels. However, hybrid models, such as KNN-WT, RF-WT, and SVR-WT, display enhanced accuracy compared to individual ML models, as indicated by statistical performance criteria. Notably, the SVR-WT model, incorporating inputs such as sunshine duration, air temperature, wind speed, and relative humidity, outperforms other models in terms of RMSE (2.174 MJ/m2), MAE (1.721 MJ/m2), R2 (0.923), MPI (28.55), and PICP (0.80) for the testing dataset. In conclusion, the integration of WT significantly improves the performance of ML models, providing valuable insights for the design and operation of solar energy systems, where precise daily SR estimation is critical for optimal and cost-efficient operation.

Effects of arcade design on wind and thermal environment inside an idealized 2D urban canyon with realistic solar heating

The arcade design has been widely used in the architecture design since it can provide space for commercial retails and improve local thermal comfort. However, the influences of different arcade design parameters on wind and thermal environment under realistic solar heating conditions are not fully understood, which hinders further application of the arcade design. This paper aims to investigate the effects of different arcade design parameters on wind and thermal environment under different solar heating conditions. Numerical wind flow, solar radiation and heat transfer simulations of a two-dimensional deep street canyon with aspect ratio of three are performed. The steady Reynolds-Averaged Navier-Stokes model with standard k−ε turbulence model is employed in this investigation, while considering solar radiation model. Moreover, the realistic solar heating simulations are achieved by setting different local solar times, i.e., LST0900 (leeward wall heating), LST1200 (street ground heating) and LST1500, (windward wall heating). Under realistic solar heating conditions, the building wall heating and the street ground heating are shown to evidently influence the strength and location of the predominant recirculation inside the street canyon, especially for the windward wall heating. Moreover, analysis results have shown that the arcade width has larger influences on wind and thermal environment than that of the arcade height. Particularly, the wind velocity increases 2.1 to 3.8 times, the air temperature decreases 0.42–2.24 °C and the Physiological Equivalent Temperature value drops between 2.2 °C and 7.0 °C when the arcade width increases to half of building width.

Global Solar Radiation and Its Interactions with Atmospheric Substances and Their Effects on Air Temperature Change in Ankara Province

On the analysis of solar radiation and meteorological variables measured in Ankara province in Türkiye from 2017 to 2018, an empirical model of global solar radiation was developed. The global solar radiation at the ground and at the top of the atmosphere (TOA) was calculated and in good agreement with the observations. This model was applied to compute the losses of global solar radiation in the atmosphere and the contributions by atmospheric absorbing and scattering substances. The loss of global solar radiation in the atmosphere was dominated by the absorbing substances. The sensitivity test showed that global solar radiation was more sensitive to changes in scattering (described by a scattering factor S/G, S and G are diffuse and global solar radiation, respectively) than to changes in absorption. This empirical model was applied to calculate the albedos at the TOA and the surface. In 2017, 2018, and 2019, the computed albedos were 28.8%, 27.8%, and 28.2% at the TOA and 21.6%, 22.1%, and 21.9% at the surface, which were in reasonable agreement with satellite retrievals. The empirical model is a useful tool for studying global solar radiation and the multiple interactions between solar energy and atmospheric substances. The comparisons of global solar radiation and its loss in the atmosphere, as well as meteorological parameters, were made at some representative sites on the Earth. Some internal relationships (between G and the absorbing and scattering substances, air temperature and atmospheric substances, air temperature increase and latitude, etc.) were found. Thus, it is suggested to thoroughly study solar radiation, atmospheric substances, and climate change as a whole system and reduce the direct emissions of all atmospheric substances and, subsequently, secondary products (e.g., CO2 and non-CO2) in the atmosphere for the achievement of slowing down climate warming.

Correlations between Total and Spectral Solar Irradiance Variations

We compare short-term (seven solar rotations), wavelength-dependent temporal variations in spectral solar irradiance (SSI) with those from the total solar irradiance (TSI). Using space-based measurements, we empirically find good correlations across most of the visible and near-infrared (NIR) spectral range, suggesting that the TSI time variability can provide a useful estimate of SSI variability. These empirically determined correlations are consistent with physics-based bolometric variations, providing a straightforward wavelength-dependent parameterization of the SSI variability given a known change in the TSI. Using a solar-irradiance model to distinguish the facular and sunspot contributions, which are responsible for nearly all the irradiance variability on timescales longer than a day, we confirm these results and determine the correlation contributions due to each magnetic activity type individually. The correlations determined from the model agree in functional form to those of the empirical data, although we do note differences near opacity minimum (1.6 μm). Our results provide a simple TSI-based estimate of the time dependence of the spectral solar variability across the ultraviolet to NIR spectral regions, with the TSI accounting for 94% of the variability in the SSI over the 400–1200 nm range.

Using of the Weibull distribution in developing global solar radiation forecasting models

AbstractIn this study, a solar radiation prediction model was developed using the Weibull distribution function (WDF). The main purpose of this study is to develop a new model for solar radiation (SR) estimation using the WDF and to bring this model to the literature. Although the WDF is widely used in wind energy forecasting due to its compatibility with wind speed data, it has not been used before in solar radiation forecasting model development. In this study, it is aimed to make SR estimation with the WDF for the first time. SR data for all regions where this new model was developed and its performance was examined were obtained from the General Directorate of Meteorology. In order to decide the success of these developed models, five different statistical metrics (RPE, MPE, MAPE, SSRE, and t‐stat) were discussed in the study. When the results are evaluated in general, it is seen that the new developed model has an acceptable performance. According to all test results in the region where the model was developed, it is seen that the new developed model showed the best performance with 0.0353, 0.4068, 7.1851, 0.1144, and 0.0162 values, respectively. The performance of this new developed model was observed in three different regions. When the statistical test results in these regions were examined, it was seen that the new model developed had a similar performance to the popular solar forecasting models widely used in the literature.

Monthly Global Solar Radiation Model Based on Artificial Neural Network, Temperature Data and Geographical and Topographical Parameters: A Case Study in Spain

Solar energy plays an essential role in the current energy context to achieve sustainable development while supplying energy needs, creating jobs, and protecting the environment. Many solar radiation models have provided valid estimates at many different locations, using appropriate input variables for specific climatic conditions, but predictions are less accurate on a regional scale. Since radiometric weather stations are relatively dispersed, even in the most developed countries, it is interesting to develop indirect models based on measurements that are common in secondary network stations. This paper develops a monthly global solar radiation model based on a simple neural network structure, using temperature, geographical, and topographical data from 105 meteorological stations, representative of the whole of peninsular Spain. A hierarchical clustering procedure was employed to select the data used to train and validate the model. To avoid functional dependencies between parameters and variables, which hinder the generality of the model, all input and output variables are dimensionless. The estimates fit the 1260 monthly data with RRMSE values of about 6%, which improves results obtained previously, using regression models, and proves that simplicity is compatible with the generality and accuracy of a model, even in large regions with very varied characteristics.

Analysis of Energy and Thermal Performance of High-Altitude Airship under Variable Attitude

The thermal problem of high-altitude airships has an essential impact on position control and energy system performance. Adjusting the airship’s attitude angle causes differences in thermal performance during position alterations. This paper studies an airship’s energy and thermal performance under variable attitudes. We establish an airship solar radiation and thermal model to analyze power output under different thermal conditions. The thermal performance of airships at varying pitch angles is investigated using computational fluid dynamic (CFD) software (Fluent V6.3.26). To determine the optimal distribution of pitch angles under various conditions, we have developed an optimization model that considers both the presence and absence of thermal influence. We have also assessed the impact of airship geometric parameters on the optimal pitch angle, considering the diversity of airship shapes. Our results demonstrate that pitch angle alterations significantly influence airships’ temperature and flow field distribution. But the degree of necessity of considering the thermal effect in calculating the optimal pitch angle distribution varies depending on the date and latitude, with the most vital need observed during low-latitude summer and the weakest during high-latitude winter. The findings of this research have significant reference value in selecting operation strategies and the control of operating performance for high-altitude airships.

Reference Evapotranspiration Evaluation Using Solar Radiation Estimated by ANN and Empirical Models

Aims: The reference evapotranspiration (ETo) estimation with Penman-Monteith or Priestley-Taylor methods requires measurements of temperature, radiation, humidity, and wind velocity. In this study, we evaluated the estimations of ETo by Penman-Monteith (ETo-PM) and Priestley-Taylor (ETo-PrT) methods using indirect methods of calculating solar radiation (Rs). Place and Duration of Study: Daily meteorological measurements from two stations in northern Greece were used for the development of solar radiation models and ETo calculation. Methodology: The indirect methods of calculating solar radiation (Rs) are based on Artificial Neural networks (ANN) technology and models using the multi-linear regression method (MLR). Three different ANN and MLR models were derived. The Hargreaves method is also used. The evaluation of the indirect Rs derived models and the ETo estimation by the two methods was performed with the use of correlation coefficients (r), root mean square error (RMSE), and efficiency (EF) indexes. Results: The statistics of ETo estimation at the two stations showed that the r and EF values, between the estimated ETo using the indirect Rs models and estimated ETo using Rs measured, were greater than 0.963 and 0.918, respectively, while the RMSE values were lower than 0.646 mm d-1. The statistics of Rs models, showed that the r and EF values were greater than 0.860 and 0.605, respectively, while the RMSE values were lower than 4.47 MJ m-2d-1. Conclusion: The results of ANN models in comparison to MLR models, when using the same input variables, are consistent between them. These findings indicate that the Penman-Monteith and Priestley-Taylor methods can accurately predict ETo using Rs values estimated indirectly through the examined methods and models.

Statistical inter-comparison study of empirical models to estimate the monthly-average daily global irradiation on tilted south oriented surfaces

Accurate determinations of the global solar irradiation received by a tilted surface during the average day of each month (HG (s, 0)) are a prerequisite in different solar energy applications, particularly in design methods. Validation of empirical models to estimate such values is of paramount importance, given the limited number of solar radiation observation sites and meteorological stations capable of providing the real information. Using the information of global irradiation on horizontal surface (HG) recommended in the Schedule of Conditions System Technologies of Low Temperature (SCT) for 52 Spanish provinces, four solar irradiation models of estimation were investigated: Liu and Jordan [1], Aláiz [2], Klein [3] and the one proposed by Censolar. A comparative study based on the statisticians of the distribution of the calculated information indicates that the values of HG (s, 0) estimated with the model of Aláiz agreed well with the reference dates. On the contrary, the Censolar model has got the maximum diversion.

Composition of feature selection techniques for improving the global horizontal irradiance estimation via machine learning models

The rise in photovoltaic capacity installed globally and the sporadic nature of solar resources underline the significance of solar radiation forecasting for grid integration. The choice of relevant climatic variables for estimating global horizontal irradiance (GHI) is crucial because it influences estimation accuracy. To keep this in mind, different feature selection techniques such as filter, wrapper, and embedded are computed and compared to identify a pertinent input variable for the prediction of GHI. In order to reduce the dimensionality of the input features, it is pertinent to remove unnecessary variables using feature selection techniques. This comprehensive study explores the use of tree-based ensemble machine learning (ML) models in the modeling of global solar radiation. Eighteen statistical and ML regressor algorithms, along with ten feature selection methods are evaluated based on statistical comparisons of their computational effectiveness and estimation errors. The lowest mean absolute error (MAE) and root mean squared error (RMSE) of 0.9938 and 2.5327 achieved by the extra-tree regressor algorithm reveals that it outperforms the other models on the test dataset. The integration of feature selection techniques improves predictive capabilities, reducing MAE and RMSE by 13.15% and 5.24%, respectively, for the Mutual Information method. This study also identifies influential features critical to accurate GHI estimation, such as solar zenith angle, diffuse horizontal irradiance, and direct normal irradiance, which provide valuable insights into the dynamics governing solar radiation estimation.

A hybrid machine-learning model for solar irradiance forecasting

Abstract Nowcasting and forecasting solar irradiance are vital for the optimal prediction of grid-connected solar photovoltaic (PV) power plants. These plants face operational challenges and scheduling dispatch difficulties due to the fluctuating nature of their power output. As the generation capacity within the electric grid increases, accurately predicting this output becomes increasingly essential, especially given the random and non-linear characteristics of solar irradiance under variable weather conditions. This study presents a novel prediction method for solar irradiance, which is directly in correlation with PV power output, targeting both short-term and medium-term forecast horizons. Our proposed hybrid framework employs a fast trainable statistical learning technique based on the truncated-regularized kernel ridge regression model. The proposed method excels in forecasting solar irradiance, especially during highly intermittent weather periods. A key strength of our model is the incorporation of multiple historical weather parameters as inputs to generate accurate predictions of future solar irradiance values in its scalable framework. We evaluated the performance of our model using data sets from both cloudy and sunny days in Seattle and Medford, USA and compared it against three forecasting models: persistence, modified 24-hour persistence and least squares. Based on three widely accepted statistical performance metrics (root mean squared error, mean absolute error and coefficient of determination), our hybrid model demonstrated superior predictive accuracy in varying weather conditions and forecast horizons.