Estimation Of Global Solar Radiation Research Articles

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.

Sunspot number-based neural network model for global solar radiation estimation in Ghardaïa

In this investigation, the estimation of global solar radiation was meticulously carried out within Ghardaïa city, a region situated in Southern Algeria, utilizing a sophisticated multilayer perceptron (MLP) neural network architecture. This research primarily concentrated on developing a predictive model based on a singular input parameter, specifically, the sunspot numbers, to forecast global solar radiation levels. The model's formulation was rooted in empirical data collected over an extensive period from 1984 to 2000, which was used for training the neural network. To assess the model's predictive accuracy and robustness, data from the years 2001 to 2004 were employed for validation purposes. The outcomes of this study were highly satisfactory, indicating that the MLP-based model possesses a significant predictive capability for Diffuse Global Solar Radiation (DGSR). This is substantiated by robust statistical metrics, including a normalized Root Mean Square Error (nRMSE) of 0.076, reflecting the model's accuracy in prediction, and a correlation coefficient (R) of 93.16%, denoting a strong correlation between the predicted and observed values. These results underscore the model's efficacy and potential application in accurately estimating global solar radiation in the specified region.

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.

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.

New independent daily global solar radiation estimation models based on the day number of the year

ABSTRACT Solar radiation data are necessary information for the analysis and prediction of certain environmental processes and the planning of related energy projects. Many models have been proposed in the literature for their estimation due to their unavailability in some locations, especially in developing countries. However, peoples are currently finding more accurate and simple models. One type of these models is the day of the year-based (DYB) model, which has as merit over others, the fact that it does not depend on any environmental parameter. In this work, nine existing DYB models are calibrated with Levenberg–Marquardt nonlinear curve fitting technique for the estimation of solar radiation in the regional cities of Cameroon. In addition, two new DYB models are proposed. The results show that one of the two proposed models (M11) best estimates the solar radiation in nine sites and is ranked second in one site with the root mean square error in the range [0.2365; 0.3039], the mean absolute percentage error in the range [0.1873; 0.2348], and the correlation coefficient in the range [0.8332; 0.9164]. The new model M11 is therefore recommended for estimating solar radiation of some sites of Cameroon and in regions with similar climate.

Estimation of Daily Global Solar Radiation with Different Sunshine-Based Models for Some Burundian Stations

Estimation of Daily Global Solar Radiation with Different Sunshine-Based Models for Some Burundian Stations

Evaluation and performance comparison of different models for global solar radiation forecasting: a case study on five cities

AbstractRecently, 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.

Modeling global solar radiation using ambient temperature

Accurate information about solar radiation is regarded as the most important stage in determining the availability of solar energy. As well, it is considered the principal input for various applications of solar energy. Because solar radiation measurements are unavailable at many locations throughout the world, many solar radiation models have been developed to predict global solar radiation. In this regard, this study objects to develop accurate and quick Global Solar Radiation (GSR) models for new locations, which currently lack an accurate model. Additionally, assess the performance of a recently introduced model, one of the best temperature-based models for GSR estimation, in these five new sites. Moreover, the proposed model's generalization capacity is investigated over the whole zone, the Suez Canal Zone, and a comparative analysis of its performance is presented. Models’ estimation is compared to the observed values, and the most common performance indicators are obtained to assess models’ performance. The findings indicate that the developed models in this study can predict global solar radiation accurately. Where the Models’ performance, both the local and general models, are larger than 95 % at all sites except for the local one at Port Said City (coastal site), it is 91 %. Additionally, the developed models have good RMSE and MABE values which range from 0.8 to 1.8 (MJ/m2 day−1) and 0.7 to 1.7 (MJ/m2 day−1), successively. Besides, they have excellent performance with coefficient of determination (R2) values ranging from 0.95 to 0.98, whereas prior research's ranges are 0.884–0.895 for R2 and 2.69–3.367 (MJ/m2 day−1) for RMSE. Therefore, the developed models in this study can be utilized for accurate GSR forecast, with their high applicability which may be achieved by combining them with various long- or short-term weather forecasting approaches, that are mostly used to reliably anticipate weather temperature. As well, these models' precise and speedy computation of GSR can also be employed in the design and performance evaluation of various solar applications.

DEVELOPMENT OF BACKPROPAGATION ALGORITHM FOR ESTIMATING SOLAR RADIATION: A CASE STUDY IN TURKEY

Solar energy has an important role in achieving the objective of replacing fossil fuels and reducing greenhouse gas emissions with significant potential. Engineers, architects, and farmers require accurate information about solar radiation in order to develop solar energy systems. It is a common practice for meteorological services around the world to measure the duration of sunshine and air temperature. Despite this, worldwide measurements of solar radiation are extremely rare and some information is lacking. It becomes vitally critical to estimate solar radiation at sites that have not own station. In the literature, a variety of models have been developed to estimate solar radiation. The artificial neural network (ANN) model is commonly used for the estimation of global solar radiation. In this study, a backpropagation algorithm for throughout the year is generated to estimate global solar radiation in Adana by using the meteorological data obtained from Turkish State Meteorological Services. ANN model was developed by using the data for 2014, 2015, and 2016 years with the MATLAB program. The data for 2017 is used for testing the model. A comparison between the developed model and real data is performed depending on the R2 value. As a result of the study, the R2 obtained by training the data was calculated as 0.9019. The R2 value derived from test data was calculated as 0.7277. In light of these results, it can be said that the estimation study was satisfactory.

Development of solar radiation prediction model using minimax algorithm

AbstractRecently, many studies are carried out on solar energy (SE) and global solar radiation (GSR) estimation. When the literature is examined, there are many GSR estimation models developed with different algorithms in different regions of the world. Although some of these models are seen to perform acceptable all over the world, the performance of some models differs according to their location. The Minimax algorithm is a decision‐making algorithm that maximizes the winning potential while minimizing the probability of losing. In this study, a new Angstorm model was developed for GSR with minimax algorithm. The most important aspect of this article in terms of originality is that the GSR estimation model is being developed with the minimax algorithm for the first time. The MATLAB program was used while developing the prediction model with the minimax algorithm. Different statistical error tests were used to evaluate the performance of the results. When these test results are evaluated in general, it is seen that the minimax algorithm gives acceptable results in GSR estimation.

Assessment of the high-resolution estimations of global and diffuse solar radiation using WRF-Solar

Compared with physical models, WRF-Solar, as an excellent numerical forecasting model, includes abundant novel cloud physical and dynamical processes, which enablesenable the high-frequency output of radiation components which are urgently needed by the solar energy industry. However, the popularisation of WRF-Solar in a wide range of applications, such as the estimation of diffuse radiation, suffers from unpredictable influences of cloud and aerosol optical property parameters. This study assessed the accuracy of the improved numerical weather prediction (WRF-Solar) model in simulating global and diffuse radiation. Aerosol optical properties at 550 nm, which were provided by a moderate resolution imaging spectroradiometer, were used as input to analyse the differences in accuracies obtained by the model with/without aerosol input. The sensitivity of WRF-Solar to aerosol and cloud optical properties and solar zenith angle (SZA) was analysed. The results show the superiority of WRF-Solar to WRF-Dudhia in terms of their root mean square error (RMSE) and mean absolute error (MAE). The coefficients of determination between WRF-Solar and WRF-Dudhia revealed no statistically significant difference, with values greater than 0.9 for the parent and nested domains. In addition, the relative RMSE (RRMSE%) reached 46.60%. The experiment on WRF-Solar and WRF-Dudhia revealed a negative bias for global radiation, but WRF-Solar attained a slightly lower RMSE and higher correlation coefficient than WRF-Dudhia. The WRF-Solar-simulated results on diffuse radiation under clear sky conditions were slightly poorer, with RMSE, RRMSE, mean percentage error and MAE of 181.93 W m−2, 170.52%, 93.04% and 138 W m−2, respectively. Based on Himawari-8 cloud data, statistical results on cloud optical thickness (COT) for cloudy days revealed that WRF-Solar overestimated diffuse radiation at COTs greater than 20. Moreover, when the aerosol optical depth was greater than or equal to 0.8, WRF-Solar also overestimated the diffuse radiation, with a mean difference of 58.57 W m−2. The errors of WRF-Solar simulations in global and diffuse radiation exhibited a significant dependence on the SZA. The dispersion degree of deviation increased gradually with the decrease in the SZA. Thus, WRF-Solar serves as an improved numerical tool that can provide high temporal and high-spatial-resolution solar radiation data for the prediction of photovoltaic power. Studies should explore the improvement of cumulus parameterisation schemes to enhance the accuracy of solar radiation component estimation and prediction under cloudy conditions.

A hybrid meteorological data simulation framework based on time-series generative adversarial network for global daily solar radiation estimation

Solar energy has received considerable attention worldwide because of its clean, safe and easily accessible characteristics. Accurate global solar radiation estimation is crucial for local energy potential assessments and deployment of photovoltaic power generation systems. However, the distribution of metrological data varies considerably between regions, and its quality considerably affects the accuracy of the estimation model. Motivated by these concerns, a metrological data estimation framework based on the time-series generative adversarial network (TimeGAN) and deep belief network (DBN) was proposed. First, the metrological data collected from 30 urban meteorological stations in China were analyzed through time-series aggregation. Next, a novel metrological data simulation model based on TimeGAN was proposed to enrich the quality of training samples. Furthermore, a generalized estimation model of daily solar radiation based on DBN was constructed. In the case study in China, TimeGAN outperforms other conventional generative adversarial networks in terms of data generation performance. The simulation results reveal that the TimeGAN-based simulation model can capture the inherent temporal characteristics of the original data, generate high-quality simulation data to reasonably improve the distribution of training data in various cities. In addition, the estimation results indicate that the proposed hybrid framework improve the accuracy of daily solar radiation estimation.

Estimation of daily global solar radiation based on different whitening applications using temperature in Mediterranean type greenhouses

The study aimed to estimate the daily global solar radiation (Rs) in Mediterranean-type greenhouses. Five different temperature-based Rs estimation models developed for open-field conditions were calibrated and validated in Mediterranean-type greenhouses in Almeria, Spain and Antalya, Türkiye, between August 26, 2013, and January 1, 2023, and between October 1, 2018, and 1 January 2023, respectively. Whitening applications were categorized according to greenhouse light transmissivity and classified as follows: without whitening or light-whitening, medium-whitening, and severe-whitening. Additionally, the best-performing model were compared with greenhouse plastic light transmissivity method. The estimation performance of the models was evaluated using the statistical indicators of the p-value of the slope, determination coefficient (R2), Nash–Sutcliffe model efficiency coefficient (NSE), root mean square error (RMSE), mean absolute error (MAE), relative error (RE), and Willmott Index (d). Compared with the other models, the Bristow and Campbell model showed a slightly higher performance in all whitening applications. Although the light transmissivity coefficient method performed slightly better than the temperature-based Rs estimation model, there was no statistical difference in the performances of the estimation models. Temperature-based estimation models offer a highly viable alternative for individuals who rely on the light transmittance approach to estimate Rs in greenhouses. This method can prove particularly useful in areas where measuring Rs outside the greenhouse is not possible or where partial time measurements cannot be taken owing to equipment malfunctions. All calibrated models can be used to estimate solar radiation using temperature data from various Mediterranean countries with similar climates and greenhouse cultivation.

Comparative Estimation of Global Solar Radiation over Two Nigerian Cities, Using Artificial Neural Network and Empirical Models

The estimation of solar radiation intensity has been a focus of many researchers due to the cost of setting up its actual measurements. While many of them employed empirical models, this study utilizes the artificial neural network for the analysis and estimation of global solar radiation over two Nigerian cities. The model developed using sunshine hours, temperatures and relative humidity were compared with the existing empirical models. Model performance indicators comparing the measured data and the computed data for the derived and selected models, using the same number of input meteorological parameters showed that ANN having average values of RMSE, MBE, and MPE of 0.0744 MJm-2day-1, -0.0020 MJm-2day-1, and -0.0043%, respectively, performed slightly better. When different number of input meteorological parameters were used, the ANN gave the following error indicators for RMSE, MBE, MPE of 0.0394MJm-2day-1, -0.0023MJm-2day-1 and -0.0144% respectively. Also, in the result of solar radiation in Abuja, using the same number of meteorological parameters, the model with the best performance in the estimation of solar radiation is the ANN model with average values of RMSE, MBE, MPE of 0.1301MJm-2day-1, 0.0053MJm-2day-1 and 0.0441% respectively. Hence, the models are versatile for predicting global solar radiation in locations in the same climatic zones as locations studied in this study, where direct measurements of solar radiation is scarce and widely separated but there is availability of commonly measured meteorological parameters such as sunshine duration, minimum temperature, maximum temperature and relative humidity.

Estimation of Global Solar Radiation using Angstrom and Gopinathan Model on Sunshine Hour and Temperature in Highland, Nepal

In developing countries like Nepal, the direct measurement of global solar radiation (GSR) is difficult. So, the estimation of GSR is carried out at Jumla (29°16′ N, 82° 11′ E and about 2347 m altitude) for the year 2015 and 2017 using regression technique to the meteorological parameters on Angstrom model, Gopinathan and Olomiyesan- Oyedum models. Among three models, Olomiyesan and Oyedum model is better than other models. Its empirical constants a=0.38, b=0.10 and c=0.09 are found. The values of statistical errors MBE, MPE and RMSE are smaller than other models. Similarly, the coefficient of determination (R2=0.89) is greater than other models. Finally, the finding empirical constants and meteorological parameters sunshine hour, and temperature are used to estimate the GSR for the year 2017. In addition to this, the annual average GSR for the year 2015 and 2017 are found to be 18.86 MJ/m2/day and 17.50 MJ/m2/day respectively. It is concluded that the finding empirical constants are used to estimate the GSR and solar energy at similar geographical location of Nepal.

Sunshine and Temperature Based Models for Estimating Global Solar Radiation in Maiduguri, Nigeria

Nine existing sunshine based models and three existing temperature based models were evaluated and compared to ascertain the most suitable models for estimating global solar radiation in Maiduguri; the most suitable sunshine based and most suitable temperature based models were also compared. The measured monthly average daily global solar radiation, sunshine hours, maximum and minimum temperature meteorological parameters during the period of thirty one years (1980 – 2010) was utilized and the evaluated models were tested statistically using validation indicators of coefficient of determination, Mean Bias Error, Root Mean Square Error, Mean Percentage Error, t – test, Nash – Sutcliffe Equation and Index of Agreement. The results indicated that the linear exponential sunshine based model and the logarithmic temperature based model were found more accurate for global solar radiation estimation in Maiduguri as compared to other evaluated models. Furthermore, the logarithmic temperature based model was found more accurate for estimating global solar radiation as compared to the linear exponential sunshine based model, and this was testified from the figure showing the comparison between the recommended sunshine based and temperature based models in which the recommended temperature based model depicts the best fitting with the measured global solar radiation data.

Comparison of Models to Evaluate Daily Available Energy with Photovoltaic Array at Sirinka, Ethiopia

Abstract: Background: Energy utilization in Ethiopia is mainly based on traditional biomass, which causes indoor air pollution, especially for women and children (causing acute lower respiratory infection) living in rural areas. Objective: This research was conducted to assess the performance of six empirical models for estimation of daily global solar radiation (DGSR) at Sirinka sites to evaluate daily available energy with photovoltaic (PV) array and energy available to the load (energy demand) and battery (energy-storage device). Materials and Methods: In this study, sunshine hours, minimum and maximum temperatures were obtained from Kombolcha Meteorological Agency. Ms-Excel 2010 was employed as a descriptive statistics tool and MATLAB 2013a was used as software to plot the analyzed data. Results: According to the statistical performance evaluation metrics, such as normalized mean bias error (NMBE) and normalized root mean square error (NRMSE), Angstrom Prescott model (AP), Louche model (LO) and Hargreaves model (H) performed best in order. The smallest daily mean power delivered to PV and the daily mean energy available to the load (energy demand) and battery (energy-storage device) were 288.11 W/m2 and 248.92 W/m2 sequenced and the largest mean power delivered to the PV and the mean energy available to the load and battery were 851.57 W/m2 and 735.76 W/m2 in order on April-10 for the study period of 2014–2018. Conclusion The result of this work showed high solar energy potential and therefore, rural electrification using PV system is possible in the study area and can reduce health problems due to indoor pollution in the future. Keywords: Daily global solar radiation, Empirical Model, environmentally friendly, Renewable Energy, Solar Power, Sunshine, Temperature, Photovoltaic.

Empirical model for the estimation of global solar radiation in the Aceh Besar Regency, Aceh

Solar energy plays an important role in the temperature distribution on the Earth's surface and essential energy that sustains life for human. In the calculation of solar energy, the limitation of solar radiation data is a major obstacle. Solar radiation data is very limited at some observation stations due to the costly procurement of measuring instruments and complicated techniques. Estimation of global solar radiation using empirical models is one way to overcome the limitations of global solar radiation data at various locations. This study built the Angstrom-Prescott model to estimate global solar radiation based on the sunshine duration’s parameter in Aceh Besar Regency. Two years (2019-2020) global solar radiation and sunshine duration data from Aceh Climatological Station were used to estimate global solar radiation in 2021. Data for 2021 was using to evaluated the equation H/H0=0.28+0.22 (n/N), the results showed good model accuracy with an RMSE value of 0.052 and a MAPE of 11% (good forecasting). The model equation has been reliable to calculate solar radiation in four locations in Aceh Besar Regency.

Estimation of Global Solar Radiation at Mid Hill Region, Okhaldhunga

Global solar radiation knowledge can be valuable in fulfilling energy demands and exploring numerous applications of solar technology. Since only a few meteorological stations are available and the installation of additional stations is complicated because the cost is very high, a comparative study of various models is essential for accurate global solar radiation prediction rather than measurement. In the present study, four models based on sunshine hour, temperature, and relative humidity for estimating daily global solar radiation were compared using the regression technique for Okhaldhunga (Lat.27.30810 N, Lon.86.50420 E, Alt.1720m.a.s.l.). These models were evaluated and compared based on four statistical error tests. The basis of analysis, model M4, which uses sunshine hour, maximum temperature, and relative humidity, was used for estimating daily global solar radiation with acceptable accuracy. The empirical constants obtained for M4 models were a=0.62, b=0.47, c= -0.003, and d= -0.003. The site's annual average global solar radiation was 4.16 kWh/m2 /day. The results of this study help estimate global solar radiation for similar geographical locations in Nepal.

Predicting surface solar radiation using a hybrid radiative Transfer–Machine learning model

Solar radiation is one of the cleanest sources of renewable energy, and it affects the carbon sink functions of terrestrial ecosystems. Although efforts have been made to establish solar radiation observation stations around the world, their coverage remains limited. Hence, the development of a wide variety of models and techniques is indispensable for obtaining effective solar radiation data. The aim of this study is to develop hybrid models with high computational speed and high accuracy to estimate global solar radiation (GSR) and quantify the uncertainty in GSR simulations caused by uncertainty in the measurements of atmospheric and surface parameters. The radiative transfer model (RTM) library for radiative transfer (LibRadtran) was coupled with six machine learning models: extreme gradient boosting (XGBoost), random forest (RF), multivariate adaptive regression splines (MARS), multilayer perceptron (MLP), deep neural networks (DNNs), and light gradient boosting machine (LightGBM). The estimated GSR was first compared to the inversion values of the GSR provided by the Aerosol Robotic Network (AERONET) and then validated using ground-based measurements at three locations in China from 2005 to 2018. The results showed that the RTM-RF is superior in terms of computational efficiency and performance, with a mean absolute errors (MAE) and coefficients of determination (R2) of 15.57 W m−2 and 0.98, respectively. Under clear sky conditions, aerosol optical depth (AOD) contributed the most to the accuracy of GSR estimates, with an average contribution of 57.95%. The measurement uncertainty due to the asymmetry factor, AOD, single-scattering albedo, and land surface albedo (LSA) can explain the differences in GSR between RTM estimates and GSR observations at the Lulin (20.33 vs. 20.91 W m−2), Wuhan (−1.40 vs. 14.58 W m−2), and Xianghe (7.28 vs. 14.32 W m−2) sites. Our study supports the use of physical models combined with machine learning models to estimate GSR and provides valuable scientific information for large-area solar radiation estimations.