Solar Model Research Articles

Causal Attention Deep-learning Model for Solar Flare Forecasting

Solar flares originate from the sudden release of energy stored in the magnetic field of the active region on the Sun, but the trigger for flares is still uncertain. Currently, deep-learning-based solar flare prediction models have achieved good results and are widely recognized. However, these models focus more on data correlation rather than causality. An ideal flare prediction model should probe into the causes/triggers of solar flares, and diagnose the precursors of flares rather than just correlation analysis. To extract more informative precursors of flares from magnetograms, while suppressing the interference of confounding factors, a causal attention module is introduced to disentangle causal and confounder features from the input features. To address the problem of imbalanced positive and negative samples in the data set, an adaptive data set split mechanism is proposed. It divides the data set into several balanced subsets of positive and negative samples, and dynamically adjusts the subsets according to the model’s prediction results during the training process. The experimental results demonstrate that our proposed model achieves 4.08%, 8.38%, and 2.19% higher accuracy, true skill score, and area under the receiver operating characteristic curve than the baseline model. Additionally, the class-specific heatmaps by using the gradient-weighted class activation mapping method reveal that our proposed model generally focuses on the polarity inverse line of active regions, well in line with theoretical study.

Applications of atomic data to studies of the Sun

The Sun is a standard reference object for astrophysics and also a fascinating subject of study in its own right. X-ray and extreme ultraviolet movies of the Sun’s atmosphere show an extraordinary diversity of plasma phenomena, from barely visible bursts and jets to coronal mass ejections that impact a large portion of the solar surface. The processes that produce these phenomena, heat the corona and power the solar wind remain actively studied and accurate atomic data are essential for interpreting observations and making model predictions. For the Sun’s interior intense effort is focused on resolving the “solar problem,” (a discrepancy between solar interior models and helioseismology measurements) and atomic data are central to both element abundance measurements and interior physics such as opacity and nuclear reaction rates. In this article, topics within solar interior and solar atmosphere physics are discussed and the role of atomic data described. Areas of active research are highlighted and specific atomic data needs are identified.Graphical abstractAn image of a solar active region obtained with the 193 A channel of SDO/AIA, showing plasma at around 1.5 million degrees.

Helium Abundance of the Sun: A Spectroscopic Analysis

Determining the He/H ratio in cool stars presents a fundamental astrophysical challenge. While this ratio is established for hot O and B stars, its extrapolation to cool stars remains uncertain due to the absence of helium lines in their observed spectra. We address this knowledge gap by focusing on the Sun as a representative cool star. We conduct spectroscopic analyses of the observed solar photospheric lines by utilizing a combination of MgH molecular lines and neutral Mg atomic lines including yet another combination of CH and C2 molecular lines with neutral C atomic lines. Our spectroscopic analyses were further exploited by adopting solar model atmospheres constructed for distinct He/H ratios to determine the solar photospheric helium abundance. The helium abundance is determined by enforcing the fact that for an adopted model atmosphere with an appropriate He/H ratio, the derived Mg abundance from the neutral Mg atomic lines and that from the MgH molecular lines must be the same. The same holds for the C abundance derived from neutral C atomic lines and that from CH lines of the CH molecular band and C2 lines from the C2 Swan band. The estimated He/H ratio for the Sun is discussed based on the one-dimensional local thermodynamic equilibrium model atmosphere. The helium abundance (He/H = 0.091−0.014+0.019 ) obtained for the Sun serves as a critical reference point to characterize the He/H ratio of cool stars across the range in their effective temperature.

A novel hybrid algorithm based on improved marine predators algorithm and equilibrium optimizer for parameter extraction of solar photovoltaic models

In the field of solar photovoltaic (PV) systems, the accurate and reliable extraction of parameters from PV models is crucial for effective simulation, evaluation, and control. Although various optimization algorithms have been widely used for parameter extraction in solar PV systems, the accuracy and reliability of the parameters extracted by these methods usually fall short of the expected standards. To address these shortcomings, a novel hybrid algorithm that combines the improved marine predators algorithm (MPA) with the equilibrium optimizer (EO), named IMPAEO, is proposed. In the IMPAEO, an elite opposition-based learning strategy is introduced to expand the exploration range of the algorithm to enhance population diversity; an adaptive weight coefficient is employed to improve the updating rule of the MPA, facilitating a balance between global exploration and local exploitation; the EO operator is integrated to enhance the performance of the MPA in the local exploitation phase to improve the solution quality and convergence speed; and the linear population size reduction (LPSR) technique is introduced to dynamically reduce the population size and improve the search performance. The effectiveness of the proposed IMPAEO is validated using the CEC2017 benchmark test suite, which is also applied to extract parameters of the single diode model, the double diode model, and three different PV modules. Comparative analysis with other algorithms demonstrates that the IMPAEO exhibits significant advantages in terms of solution quality, convergence speed, and algorithm stability. Consequently, the proposed IMPAEO not only proves to be efficient and reliable in parameter extraction for solar PV models but also offers a promising alternative in this field.

Exploring the time variability of the solar wind using LOFAR pulsar data

High-precision pulsar timing is highly dependent on the precise and accurate modelling of any effects that can potentially impact the data. In particular, effects that contain stochastic elements contribute to some level of corruption and complexity in the analysis of pulsar-timing data. It has been shown that commonly used solar wind models do not accurately account for variability in the amplitude of the solar wind on both short and long timescales. In this study, we test and validate a new, cutting-edge solar wind modelling method included in the enterprise software suite (widely used for pulsar noise analysis) through extended simulations. We use it to investigate temporal variability in LOFAR data. Our model testing scheme in itself provides an invaluable asset for pulsar timing array (PTA) experiments. Since, improperly accounting for the solar wind signature in pulsar data can induce false-positive signals, it is of fundamental importance to include in any such investigations. We employed a Bayesian approach utilising a continuously varying Gaussian process to model the solar wind. It uses a spherical approximation that modulates the electron density. This method, which we refer to as a solar wind Gaussian process (SWGP), has been integrated into existing noise analysis software, specifically enterprise . Our Validation of this model was performed through simulations. We then conduct noise analysis on eight pulsars from the LOFAR dataset, with most pulsars having a time span of $ 11$ years encompassing one full solar activity cycle. Furthermore, we derived the electron densities from the dispersion measure values obtained by the SWGP model. Our analysis reveals a strong correlation between the electron density at 1 AU and the ecliptic latitude (ELAT) of the pulsar. Pulsars with $|ELAT|< 3^ circ $ exhibit significantly higher average electron densities. Furthermore, we observed distinct temporal patterns in the electron densities in different pulsars. In particular, pulsars within $|ELAT|< 3^ circ $ exhibit similar temporal variations, while the electron densities of those outside this range correlate with the solar activity cycle. Notably, some pulsars exhibit sensitivity to the solar wind up to $45^ circ $ away from the Sun in LOFAR data. The continuous variability in electron density offered in this model represents a substantial improvement over previous models, that assume a single value for piece-wise bins of time. This advancement holds promise for solar wind modelling in future International Pulsar Timing Array (IPTA) data combinations.

Measuring the Spot Variability of T Tauri Stars Using Near-infrared Atomic Fe and Molecular OH Lines

As part of the Young Exoplanets Spectroscopic Survey, this study explores the spot variability of 13 T Tauri Stars (TTSs) in the near-infrared H band, using spectra from the Immersion GRating INfrared Spectrometer. By analyzing effective temperature (T eff) sensitive lines of atomic Fe i at ∼1.56259 μm and ∼1.56362 μm, and molecular OH at ∼1.56310 and ∼1.56317 μm, we develop an empirical equivalent width ratio (EWR) relationship for T eff in the range of 3400–5000 K. This relationship allows for precise relative T eff estimates to within tens of Kelvin and demonstrates compatibility with solar metallicity target models. However, discrepancies between observational data and model predictions limit the extension of the T eff–EWR relationship to a broader parameter space. Our study reveals that both classical and weak-line TTSs can exhibit T eff variations exceeding 150 K over a span of 2 yr. The detection of a quarter-phase delay between the EWR and radial velocity phase curves in TTSs indicates spot-driven signals. A phase delay of 0.06 ± 0.13 for CI Tau, however, suggests additional dynamics, potentially caused by planetary interaction, inferred from a posited 1:1 commensurability between the rotation period and orbital period. Moreover, a positive correlation between T eff variation amplitude and stellar inclination angle supports the existence of high-latitude spots on TTSs, further enriching our understanding of stellar surface activity in young stars.

Testing the volume integrals of travel-time sensitivity kernels for flows

Context. Helioseismic inversions largely rely on sensitivity kernels, in which 3D spatial functions describe how the changes in the solar interior translate into the change in helioseismic observables. These sensitivity kernels in most cases come from the forward modelling that is used in the most advanced solar models. Aims. We aim to test the sensitivity kernels by comparing their volume integrals with measured values from helioseismic travel times. Methods. By manipulating the tracking rate, we mimicked the additional zonal velocity in the Dopplergram datacubes. These datacubes were then processed by a standard travel-time measurements pipeline. We investigated the dependence of the east-west travel time averaged over a box around the disc centre on the implanted tracking velocity. The slope of this dependence is directly proportional to the total volume integral of the sensitivity kernel that corresponds to the travel-time geometry that is used. Results. The agreement between measurements and models for travel times that are computed with a ridge filtering is very good to acceptable. The dependence we sought to determine indeed resembles a linear function, and its slope agrees with the expected volume integral from the forward-modelled sensitivity kernel. The agreement is poorer for the phase-speed filtered datacubes. The disagreement is particularly strong for the slowest phase speeds (filters td1–td4). For the higher phase speeds, our result indicates that the measured kernel integrals are systematically larger than expected from the forward modelling. We admit our testing procedure may not be appropriate for high phase speeds and higher radial modes.

Short-term solar eruptive activity prediction models based on machine learning approaches: A review

Short-term solar eruptive activity prediction models based on machine learning approaches: A review

Design-Optimization of Solar-Collector Tilt Angle for Annual-Maximum and Seasonal-Balanced Energy Received

Abstract A detailed research study was performed on the annual energy received by a fixed-tilt solar collector. A new clear-sky model of solar radiation was developed by combining existing clear-sky models. Then, this model, was used to perform optimizations with two different objectives. The first objective is to achieve the maximum-annual solar energy received by a solar panel, while the second is to balance its energy received so that it is more uniform over the course of a year. Each of these objectives have differing optimum tilt values, depending on the location and usage of a solar receiver system. The novelty of this investigation is the use of the aforementioned solar model with a unique algorithm for multiple optimizations for these two different objectives, which results in two different optimum tilt angles that are in turn different from the widely accepted rule-of-thumb that recommends an angle that is equal to the latitude. The solar radiation results obtained with the model used herein were experimentally confirmed through comparison with measured weather data for a number of locations with different sky clearness indexes as well as comparison with those of other studies found in the literature. A major contribution of this study are the plots and tables showing optimum tilt angles as estimated by the model as well as with empirical data. By selecting the appropriate tilt angle, a designer can now decide to either maximize annual electrical energy production or else to balance delivery of electricity over the year.

Chemical-Inspired Material Generation Algorithm (MGA) of Single- and Double-Diode Model Parameter Determination for Multi-Crystalline Silicon Solar Cells

The optimization of solar photovoltaic (PV) cells and modules is crucial for enhancing solar energy conversion efficiency, a significant barrier to the widespread adoption of solar energy. Accurate modeling and estimation of PV parameters are essential for the optimal design, control, and simulation of PV systems. Traditional optimization methods often suffer from limitations such as entrapment in local optima when addressing this complex problem. This study introduces the Material Generation Algorithm (MGA), inspired by the principles of material chemistry, to estimate PV parameters effectively. The MGA simulates the creation and stabilization of chemical compounds to explore and optimize the parameter space. The algorithm mimics the formation of ionic and covalent bonds to generate new candidate solutions and assesses their stability to ensure convergence to optimal parameters. The MGA is applied to estimate parameters for two different PV modules, RTC France and Kyocera KC200GT, considering their manufacturing technologies and solar cell models. The significant nature of the MGA in comparison to other algorithms is further demonstrated by experimental and statistical findings. A comparative analysis of the results indicates that the MGA outperforms the other optimization strategies that previous researchers have examined for parameter estimation of solar PV systems in terms of both effectiveness and robustness. Moreover, simulation results demonstrate that MGA enhances the electrical properties of PV systems by accurately identifying PV parameters under varying operating conditions of temperature and irradiance. In comparison to other reported methods, considering the Kyocera KC200GT module, the MGA consistently performs better in decreasing RMSE across a variety of weather situations; for SD and DD models, the percentage improvements vary from 8.07% to 90.29%.

A novel solar radiation forecasting model based on time series imaging and bidirectional long short‐term memory network

AbstractThe instability of solar energy is the biggest challenge to its successful integration with modern power grids, and accurate prediction of long‐term solar radiation can effectively solve this problem. In this study, we proposed a novel long‐term solar radiation prediction model based on time series imaging and bidirectional long short‐term memory network. First, inspired by the computer vision algorithm, the recursive graph algorithm is used to transform the one‐dimensional time series into two‐dimensional images, and then convolutional neural network is used to extract the features from the images, thus, the deeper features in the original solar radiation data can be mined. Second, to solve the problem of low accuracy of long‐term solar radiation prediction, a hybrid model BiLSTM‐Transformer is used to predict long‐term solar radiation. The hybrid prediction model can capture the long‐term dependencies, thereby further improving the accuracy of the prediction model. The experimental results show that the hybrid model proposed in this study is superior to other single models and hybrid models in long‐term solar radiation prediction accuracy. The accuracy and stability of the hybrid model are verified by many tests.

A new lightweight framework based on knowledge distillation for reducing the complexity of multi-modal solar irradiance prediction model

The inherent uncertainty of solar energy brings great difficulties to the grid connection and short-term energy planning and dispatching. Deep learning method makes it possible to predict the short-term solar energy with its powerful learning ability, but its complex model structure and huge trainable parameters bring great difficulties to the practical deployment. Therefore, this paper proposes a lightweight framework based on knowledge distillation strategy, which greatly reduces the complexity of multi-modal solar irradiance prediction model meanwhile ensuring an acceptable accuracy, facilitating the practical deployment. Firstly, a teacher model with multi-modal structure and good accuracy is built based on ResNet18-Informer. Then, the lightweight model is obtained by the proposed lightweight framework depending on the knowledge of teacher model. The comparisons of various models and the optimal settings of knowledge distillation are analyzed. Results show that the lightweight model can reduce the trainable parameters, inference time, and GPU memory by 97.7%, 52.5%, and 36.3%, respectively. The normalized root mean square error is reduced by 24.87% compared with the same structure model but without knowledge distillation, verifying the superiority of the proposed framework. The soft loss using the light loss with the ratio of 0.3 can obtain the best training results for the lightweight model. The structure with 3 residual blocks and 3 LSTM layers is proved to be the best for the lightweight model in the solar irradiance prediction task.

Estimation of Aerosol Characteristics from Broadband Solar Radiation Measurements Carried Out in Southern Algeria

Aerosols in the atmosphere significantly reduce the solar radiation reaching the Earth’s surface through scattering and absorption processes. Knowing their properties becomes essential when we are interested in measuring solar radiation at a given location on the ground. The commonly used parameters that characterize their effects are the Aerosol Optical Depth τ, the Angstrom exponent α, and the Angstrom coefficient β. One method for estimating these parameters is to fit ground-based measurements of clear-sky direct solar radiation using a model on which it depends. However, the choice of model depends on its suitability to the atmospheric conditions of the site considered. Eleven empirical solar radiation models depending on α and β were thus chosen and tested with solar radiation measurements recorded between 2005 and 2014 in Tamanrasset in southern Algeria. The results obtained were compared to measurements made with the AERONET solar photometer on the same site during the same period. Among the 11 models chosen, the best performing ones are REST2 and CPCR2. They proved to be the best suited to estimate β with approximately the same RMSE of 0.05 and a correlation coefficient R with respect to AERONET of 0.95. The results also highlighted good performances of these models for the estimation of τ with an RMSE of 0.05 and 0.04, and an R of 0.95 and 0.96, respectively. The values of α obtained from the fitting of these models were, however, less good, with R around 0.38. Additional treatments based on a Recurrent Neural Network (RNN) were necessary to improve its estimation. They provided promising results showing a significant improvement in α estimates with R reaching 0.7 when referring to AERONET data. Furthermore, this parameter made it possible to identify different types of aerosols in Tamanrasset such as the presence of maritime, dust, and mixed aerosols representing, respectively, 31.21%, 3.25%, and 65.54%, proportions calculated over the entire period studied. The seasonal analysis showed that maritime aerosols are predominant in the winter in Tamanrasset but decrease with the seasons to reach a minimum in the summer (JJA). Dust aerosols appear in February and persist mainly in the spring (MAM) and summer (JJA), then disappear in September. These results are also consistent with those obtained from AERONET.

A tree seed algorithm with multi-strategy for parameter estimation of solar photovoltaic models

Tree seed algorithm, which is one of the metaheuristics algorithms recently proposed for the solution of continuous optimization problems, has an effective algorithmic structure inspired by the relation between trees and seeds. At the same time, the use of two different solution generation mechanisms by depending on the control parameter in TSA aims to balance the exploration and exploitation capabilities of the algorithm. However, when the structure of the algorithm is examined in detail, it is seen that there are some disadvantages such as loss of population diversity and getting stuck in local minimums. To overcome these disadvantages in the basic algorithm, three different approaches (self-adaptive weighting mechanism, chaotic elite learning approach and experience-based learning method) were proposed to TSA under the name of multi-strategies in this study. The algorithm improved with these approaches is named as the multi-strategy-based tree seed algorithm (MS-TSA). MS-TSA was first tested on CEC2017 functions. Then MS-TSA was applied to the problems in the CEC2020 competition and compared with the results of the best performing algorithms in this competition. As a result of the comparisons, MS-TSA was found to be a competitive method on solving benchmark functions. Then, parameter estimation of single diode, double diode and photovoltaic module models using the input data of various solar panels was carried out by the MS-TSA. The results obtained with MS-TSA were compared with both the results of the basic TSA and the results of well-known algorithms in the literature. The results obtained are 9.8642E-04, 9.8356E-04, 2.4251E-03, 1.7534E-03 respectively. As a result of the comparative analysis, the lowest RMSE value was obtained by MS-TSA. In addition, comprehensive performance analyzes of the algorithms were made with the convergence curve, boxplots, current (I) - voltage (V) and power (P) - voltage (V) characteristic curves obtained according to the experimental results. As a result of the experiments and analyses, MS-TSA was found to be a more successful method than the compared algorithms in parameter estimation of PV models.

A Data-constrained Analysis for Joule Heating as a Solar Active Region Atmosphere Heating Mechanism. I. Sunspot Umbral Light Bridge

Understanding the mechanisms underlying the heating of the solar atmosphere is a fundamental problem in solar physics. The lower atmosphere of the Sun (i.e., photosphere and chromosphere) is composed of weakly ionized plasma. This results in anisotropic dissipation of electric currents by Coulomb and Cowling resistivities. Joule heating due to dissipation of currents perpendicular to the magnetic field by Cowling resistivity has been demonstrated to be the main mechanism for the heating of a sunspot umbral light bridge located in NOAA AR 12002 on 2014 March 13. Here, we focus on the same target region and demonstrate the importance of further constraining our Joule heating model using observational data in addition to magnetic field, namely plasma temperature calculated from the inversion of spectroscopic data obtained from the Interferometric BI-dimensional Spectrometer instrument of the ground-based Dunn Solar Telescope. As a parameter in our analysis, temperature is demonstrated to have the highest sensitivity after magnetic field. We show that the heating of the light bridge is a highly dynamic event that necessitates utilization of 3D spatially resolved observational data for temperature rather than a 1D temperature stratification based on theoretical/semiempirical solar atmosphere models. Our improved data-constrained analysis using spatially resolved temperatures shows that the entire light bridge is heated by the proposed mechanism, and yields heating rate values that are consistent with our previous study.

Helioseismic Properties of Dynamo Waves in the Variation of Solar Differential Rotation

Solar differential rotation exhibits a prominent feature: its cyclic variations over the solar cycle, referred to as zonal flows or torsional oscillations, are observed throughout the convection zone. Given the challenge of measuring magnetic fields in subsurface layers, understanding deep torsional oscillations becomes pivotal in deciphering the underlying solar dynamo mechanism. In this study, we address the critical question of identifying specific signatures within helioseismic frequency-splitting data associated with the torsional oscillations. To achieve this, a comprehensive forward modeling approach is employed to simulate the helioseismic data for a dynamo model that, to some extent, reproduces solar-cycle variations of magnetic fields and flows. We provide a comprehensive derivation of the forward modeling process utilizing generalized spherical harmonics, as it involves intricate algebraic computations. All estimated frequency-splitting coefficients from the model display an 11 yr periodicity. Using the simulated splitting coefficients and realistic noise, we show that it is possible to identify the dynamo wave signal present in the solar zonal flow from the tachocline to the solar surface. By analyzing observed data, we find similar dynamo wave patterns in the observational data from the Michelson Doppler Imager, Helioseismic Magnetic Imager, and Global Oscillation Network Group. This validates the earlier detection of dynamo waves and holds potential implications for the solar dynamo theory models.

A Markovian description of the multi-level source function and its application to the Lyman series in the Sun

Aims. We introduce a new method to calculate and interpret indirect transition rates populating atomic levels using Markov chain theory. Indirect transition rates are essential to evaluate interlocking in a multi-level source function, which quantifies all the processes that add and remove photons from a spectral line. A better understanding of the multi-level source function is central to interpret optically thick spectral line formation in stellar atmospheres, especially outside local thermodynamical equilibrium (LTE). Methods. We compute the level populations from a hydrogen model atom in statistical equilibrium, using the solar FALC model, a 1D static atmosphere. From the transition rates, we reconstruct the multi-level source function using our new method and compare it with existing methods to build the source function. We focus on the Lyman series lines and analyze the different contributions to the source functions and synthetic spectra. Results. Absorbing Markov chains can represent the level-ratio solution of the statistical equilibrium equation and can therefore be used to calculate the indirect transition rates between the upper and lower levels of an atomic transition. Our description of the multi-level source function allows a more physical interpretation of its individual terms, particularly a quantitative view of interlocking. For the Lyman lines in the FALC atmosphere, we find that interlocking becomes increasingly important with order in the series, with Ly-α showing very little, but Ly-β nearly 50% and Ly-γ about 60% contribution coming from interlocking. In some cases, this view seems opposed to the conventional wisdom that these lines are mostly scattering, and we discuss the reasons why. Conclusions. Our formalism to describe the multi-level source function is general and can provide more physical insight into the processes that set the line source function in a multi-level atom. The effects of interlocking for lines formed in the solar chromosphere can be more important than previously thought, and our method provides the basis for further exploration.

Proposal and Verification of Solar Carport Model using Only Renewable Energy

Proposal and Verification of Solar Carport Model using Only Renewable Energy

Voltage root mean square error calculation for solar cell parameter estimation: A novel g-function approach

The existing research on estimating solar cell parameters mainly focuses on minimizing the Root-Mean-Square Error (RMSE) between the estimated and measured current values of solar cells (referred to as the RMSEI). This involves using an analytical expression for current - I as a function of voltage - U (I = f1(U)) expressed through the Lambert W function. This paper introduces a new analytical solution for calculating the RMSE between measured and estimated solar cell voltages (referred to as the RMSEU). The formula is derived using the g-function or LogWright function, which provides a numerically applicable method for representing the analytical relation between solar cell voltage and current (U = f2(I)). Moreover, the paper presents the original formulas for calculating Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) for solar cell voltages expressed through the g-function. The paper also compares various published approaches and examines two well-known solar cells/modules, namely the RTC France solar cell and the SOLAREX MSX–60 PV solar module, in terms of the RMSEU and single-diode solar cell models. Additionally, a novel metaheuristic algorithm, known as the Chaotic Walrus Optimization Algorithm (Chaotic-WaOA), is proposed for solar cell parameter estimation. This algorithm is applied to both mentioned solar cells to determine their parameters in terms of minimal RMSEU. In order to verify the proposed research, experimental observations were conducted on a 60Wp monocrystalline solar module installed at the Faculty of Sciences and Mathematics in Niš, Serbia. This was done using a specialized PV-KLA apparatus under different outdoor conditions. The research was also verified with a Cadmium telluride solar module (CdTe75638) and a multi-crystalline silicon solar module (mSi0188). The investigation results demonstrate the accuracy, applicability, and numerical feasibility of the proposed RMSE expression and algorithm. Additionally, the study confirms the applicability of the g-function in invertible solar cell modeling.

Artificial Intelligence-Based Improvement of Empirical Methods for Accurate Global Solar Radiation Forecast: Development and Comparative Analysis

Artificial intelligence (AI) technology has expanded its potential in environmental and renewable energy applications, particularly in the use of artificial neural networks (ANNs) as the most widely used technique. To address the shortage of solar measurement in various places worldwide, several solar radiation methods have been developed to forecast global solar radiation (GSR). With this consideration, this study aims to develop temperature-based GSR models using a commonly utilized approach in machine learning techniques, ANNs, to predict GSR using just temperature data. It also compares the performance of these models to the commonly used empirical technique. Additionally, it develops precise GSR models for five new sites and the entire region, which currently lacks AI-based models despite the presence of proposed solar energy plants in the area. The study also examines the impact of varying lengths of validation datasets on solar radiation models’ prediction and accuracy, which has received little attention. Furthermore, it investigates different ANN architectures for GSR estimation and introduces a comprehensive comparative study. The findings indicate that the most advanced models of both methods accurately predict GSR, with coefficient of determination, R2, values ranging from 96% to 98%. Moreover, the local and general formulas of the empirical model exhibit comparable performance at non-coastal sites. Conversely, the local and general ANN-based models perform almost identically, with a high ability to forecast GSR in any location, even during the winter months. Additionally, ANN architectures with fewer neurons in their single hidden layer generally outperform those with more. Furthermore, the efficacy and precision of the models, particularly ANN-based ones, are minimally impacted by the size of the validation data sets. This study also reveals that the performance of the empirical models was significantly influenced by weather conditions such as clouds and rain, especially at coastal sites. In contrast, the ANN-based models were less impacted by such weather variations, with a performance that was approximately 7% better than the empirical ones at coastal sites. The best-developed models, particularly the ANN-based models, are thus highly recommended. They enable the precise and rapid forecast of GSR, which is useful in the design and performance evaluation of various solar applications, with the temperature data continuously and easily recorded for various purposes.