Obtaining precise solar radiation data is the first stage in determining the availability of solar energy. It is also regarded as one of the major inputs for a variety of solar applications. Due to the scarcity of solar radiation measurement data for many locations throughout the world, many solar radiation models are utilized to predict global solar radiation. Indeed, the most widely used AI technique is artificial neural networks (ANNs). Hitherto, while ANNs have been utilized in various studies to estimate global solar radiation (GSR), limited attention has been given to the architecture of ANN. Thus, this study aimed to: first, optimize the design of one of the faster and most used machine-learning (ML) algorithms, the ANN, to forecast GSR more accurately while saving computation power; second, optimize the number of neurons in the hidden layer to obtain the most significant ANN model for accurate GSR estimation, since it is still lacking; in addition to investigating the impact of varying the number of neurons in the hidden layer on the proficiency of the ANN-based model to predict GSR with high accuracy; and, finally, conduct a comparative study between the ANN and empirical techniques for estimating GSR. The results showed that the best ANN model and the empirical model provided an excellent estimation for the GSR, with a Coefficient of Determination $R^2$ greater than 0.98%. Additionally, ANN architectures with a smaller number of neurons in the single hidden layer (1–3 neurons) provided the best performance, with $R^2$ > 0.98%. Furthermore, the performance of the developed ANN models remained approximately stable and excellent when the number of hidden layer’s neurons was less than ten neurons ($R^2$ > 0.97%), as their performance was very close to each other. However, the ANN models experienced performance instability when the number of hidden layer’s neurons exceeded nine neurons. Furthermore, the performance comparison between the best ANN-based model and the empirical one revealed that both models performed well ($R^2$ > 0.98%). Moreover, while the relative error for the best ANN model slightly exceeded the range, ±10% in November and December, it remained within the range for the empirical model even in the winter months. Additionally, the obtained results of the best ANN model in this work were compared with the recent related work. While it had a good RMSE value of 0.8361 MJ/m² day⁻¹ within the ranges of previous work, its correlation coefficient (r) was the best one. Therefore, the developed models in this study can be utilized for accurate GSR forecasting. The accurate and efficient estimation of global solar radiation using both models can be valuable in designing and performance evaluation for different solar applications.

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.

Recently, solar energy has emerged as the most promising renewable energy source to meet the world’s energy demands. However, to harness the potential of solar energy, accurate data on solar radiation are crucial. It is considered the first step in assessing solar resources for various applications and achieving energy sustainability goals. Due to the unavailability of solar radiation measurements in many parts of the world, several models have been developed to predict global solar radiation (GSR) at these locations. Thus, this study aims to evaluate the proficiency of several GSR models at five new locations and determine the most suitable one for GSR prediction. The study has further developed solar radiation models for these new locations, as well as general ones for the entire region, which does not have any GSR models despite the existence of many planned solar energy facilities in this area. Additionally, the study investigates the effect of changing the length of the validation dataset on models’ performance and accuracy, as well as assesses the introduced models’ generalization capability. To achieve these objectives, the observed data of GSR for approximately 37 years at studied locations are used to develop and validate the proposed models. The study’s findings reveal that Model 1 provides the best performance at all locations, with accuracy, coefficient of determination ( $$R^{2}$$ R 2 ), ranging from 95 to 98%, except for the coastal location, where it is from 91 to 95%. The remaining performance indicators of the best models, such as RMSE, MABE, MAPE, and $$r$$ r , are good, and their values range from 0.7863 to 1.9097 (MJ m⁻² day⁻¹), from 0.6430 to 1.7060 (MJ m⁻² day⁻¹), from 3.4319 to 10.0890 (%), and from 0.9914 to 0.9981, respectively. The length of the validation dataset has a slight effect on the models’ performance, ranging from about 1% to 2%. Therefore, Model 1 is the recommended solar radiation model, which can provide precise and rapid estimates of global solar radiation. This approach could be used in the design and performance evaluation of many solar applications. The primary benefit of this approach in the current investigation is that temperature data are continuously and effortlessly recorded for various purposes.