Daily Global Solar Radiation Research Articles

ANN-based modelling and prediction of daily global solar irradiation using commonly measured meteorological parameters

The aim of this work is to develop an artificial neural network (ANN) based model for accurately predicting the daily global solar irradiation in the city of Fez. The potential of the developed model is verified and appraised through the local collected database for the period 2009-2015 from the radiometric station of the Faculty of Sciences and Technology of Fez. The obtained model is MLP with feed forward back-propagation algorithm containing three input parameters and a single hidden layer with nine neurons. Coefficient of determination R2, the mean absolute percentage error MAPE and the relative root mean square error RRMSE are respectively equal to 97.16%, 21.77% and 18.79%.

Estimation of the daily global solar radiation based on the Gaussian process regression methodology in the Saharan climate

Accurate estimation of solar radiation is the major concern in renewable energy applications. Over the past few years, a lot of machine learning paradigms have been proposed in order to improve the estimation performances, mostly based on artificial neural networks, fuzzy logic, support vector machine and adaptive neuro-fuzzy inference system. The aim of this work is the prediction of the daily global solar radiation, received on a horizontal surface through the Gaussian process regression (GPR) methodology. A case study of Ghardaia region (Algeria) has been used in order to validate the above methodology. In fact, several combinations have been tested; it was found that, GPR-model based on sunshine duration, minimum air temperature and relative humidity gives the best results in term of mean absolute bias error (MBE), root mean square error (RMSE), relative mean square error (rRMSE), and correlation coefficient (r) . The obtained values of these indicators are 0.67 MJ/m2, 1.15 MJ/m2, 5.2%, and 98.42%, respectively.

Estimation methods of global solar radiation, cell temperature and solar power forecasting: A review and case study in Eskişehir

Solar energy is the most important energy resource that has become an efficient solution to the world's energy challenges. Accurate knowledge of global solar radiation and cell temperature are required for solar photovoltaic power forecasting. In this paper, a comprehensive literature review of methods used for estimation of global solar radiation, cell temperature and solar power generation forecasting are presented. In addition, a comparative analysis is presented using the actual data, which is collected from a home placed in Anadolu University İki Eylül Campus in Eskişehir as a comprehensive case study. Within the scope of this analysis, hourly global solar radiation values on horizontal surface are estimated from the measured daily global solar radiation values by using eleven different models. By using the selected model, estimated hourly global solar radiation values for horizontal surface are converted to values for inclined surface. Cell temperature of the photovoltaic modules is estimated with various known models in the literature. Based on the estimated cell temperature and global solar radiation values on inclined surface, power generation values of the on-grid and off-grid systems are forecasted. Therefore, based on the statistical analysis, the most accurate models are recommended to be carried out in any location that has similar climatic conditions with the considered city.

Artificial Neural Network for Forecasting One Day Ahead of Global Solar Irradiance

Due to global warming, the world is seeking to use more renewable energy. In this study, we focus on solar energy, which has been receiving increased amounts of attention in the last few decades. The integration of solar energy into electricity networks requires reliable forecast information of solar resources enabling it to quantify the available energy and allowing it to optimally manage the transition between intermittent and conventional energies. Throughout our research, we investigated different forecasting techniques in order to find which one is appropriate for forecasting the daily global solar irradiance for the region of Rabat. The first-tested approach is linear modeling based on classical ARIMA-GARCH and exponential smoothing models. The second approach proposes non-linear modeling based on Artificial Neural Networks (ANNs) models. Numerous research has demonstrated the ability of ANNs to predict time series of weather data. In this study, we will examine a particular structure of ANNs, Multilayer Perceptron (MLP), which has been used the most among ANN structures in renewable energy and time series forecasting broadly. We used some statistical feature parameters to find the optimal structure of MLP in the univariate case and the multivariate case. The results showed that the MLP with exogenous variables performed better than the other models.

Solar radiation evaluation by t statistical analysis method for south and north region of Turkey

Energy demand for alternative resources increases day by day. The solar energy is one of the most preferred sustainable energy sources. Building a solar energy facility encounters the problem of alteration solar radiation. Literature shows that defining solar radiation can be done by regional models. Therefore, estimation of solar radiation potential for specified region which is Turkey should be done partially. Hence, the south and north region of Turkey is evaluated by means of solar radiation models on the horizontal surface for the clean air conditions. The examination is done under three stations of each region. As selected models represent monthly average of daily global solar radiation for specific stations, the responses of each model differ due to not only region but also each station of Turkey. In order to define the accuracy and performance of the models, t statistical examination is done for each station. The results meet the expectations by representing different models are suitable for particular regions. On the other hand, the solar radiation models on the horizontal surface for the clean air conditions are proposed to the literature for stations of north and south region of Turkey.

Comparison of artificial intelligence methods in estimation of daily global solar radiation

Abstract Assessment of solar potential over a location of interest is introduced as an important step for the successful planning of solar energy systems (photovoltaic or thermal). Due to the absence of meteorological stations and sophisticated solar sensors, solar data may be unavailable for every point of interest. Hence, empirical and intelligence methods are developed to estimate solar irradiance data. In this study, the idea of artificial intelligence methods is employed to predict the daily global solar radiation. The developed models are: group method of data handling (GMDH) type neural network, multilayer feed-forward neural network (MLFFNN), adaptive neuro-fuzzy inference system (ANFIS), ANFIS optimized with particle swarm optimization algorithm (ANFIS-PSO), ANFIS optimized with genetic algorithm (ANFIS-GA) and ANFIS optimized with ant colony (ANFIS-ACO). The data are collected from 12 stations in different climate zones of Iran. The input variables of the models are including month, day, average air temperature, maximum air temperature, minimum air temperature, air pressure, relative humidity, wind speed, top of atmosphere insolation, latitude and longitude. The results demonstrated that although the developed models can successfully predict the global horizontal irradiance, the GMDH model outperforms the other developed models. The values of root mean square error (RMSE), determination coefficient (R2) and mean square error (MSE) for the GMDH model were 0.2466 (kWh/m2/day), 0.9886 and 0.0608 (kWh/m2/day), respectively.

Solar Radiation Parameters for Assessing Temperature Distributions on Bridge Cross-Sections

Solar radiation is one of the most important factors influencing the temperature distribution on bridge girder cross-sections. The bridge temperature distribution can be estimated using estimation models that incorporate solar radiation data; however, such data could be cost- or time-prohibitive to obtain. A review of literature was carried out on estimation models for solar radiation parameters, including the global solar radiation, beam solar radiation and diffuse solar radiation. Solar radiation data from eight cities in Fujian Province in southeastern China were obtained on site. Solar radiation models applicable to Fujian, China were proposed and verified using the measured data. The linear Ångström–Page model (based on sunshine duration) can be used to estimate the daily global solar radiation. The Collares-Pereira and Rabl model and the Hottel model can be used to estimate the hourly global solar radiation and the beam solar radiation, respectively. Three bridges were chosen as case study, for which the temperature distribution on girder cross-sections were monitored on site. Finite element models (FEM) of cross-sections of bridge girders were implemented using the Midas program. The temperature–time curves obtained from FEM showed very close agreement with the measured values for summertime. Ignoring the solar radiation effect would result in lower and delayed temperature peaks. However, the influence of solar radiation on the temperature distribution in winter is negligible.

Analysis of different combinations of meteorological parameters in predicting the horizontal global solar radiation with ANN approach: A case study

Solar energy is a clean renewable energy source and availability of solar resources at a particular location depends on the local meteorological parameters. In the present study, prediction models using artificial neural networks (ANN) are developed by varying the meteorological parameters from one to six. A two year database of daily global solar radiation (GSR), daily minimum temperature (Tmin), daily maximum temperature (Tmax), difference of daily maximum and minimum temperature (DT), sunshine hours (S), theoretical sunshine hours (So) and extraterrestrial radiation (Ho) have been used to train the ANN. Six ANN models are developed (ANN-1 to ANN-6) with 32 possible combinations of inputs and are used to train the network to identify the best combination of inputs to estimate the monthly mean daily GSR accurately. All the models are validated and the performance of the models are analyzed by using the statistical tools.Out of the six ANN models with all the possible combination of input variables, ANN-2 and ANN-3 have given best prediction with the combinations of [DT, Ho] and [DT, Ho, So] respectively. The statistical tool Relative Root Mean Square Error (RRMSE) showed the least value of 3.96% with [DT, Ho] inputs. The ANN-1 trained with calculated approximate sunshine hours (Sa) has also shown high prediction accuracy. Sunshine based model and temperature based model are validated with ANN-1, ANN-2 and ANN-3 architectures. Results showed that the developed ANN models outperform the considered empirical models. The combination of [So, Ho] has produced excellent estimation, which are theoretical parameters and does not require any measured meteorological parameters. Superior performance is observed with less number of inputs which are readily available for any location.

Prospects of hybrid photovoltaic–diesel standalone system for six different climate locations in Indian state of Tamil Nadu

In developing countries like India, very limited observatory stations are available to collect daily global solar radiation (GSR). This paper focuses to develop a common model for daily GSR estimation and standalone hybrid power system (HPS) design in view of six different climate locations in India. A new strategy is applied for finding the daily GSR model in Madurai location in Tamil Nadu, India. It is based on new parameter, which is named as 'approximate-bright sunshine duration' from the real-time measurement. The remaining six locations are estimated from the new model developed. In addition, to improve the accuracy of estimation, the developed linear one input variable model is further modified as two input variable model, based on the latitudes of six different climatic locations. The performances of developed one-input and two input models are analyzed using statistical indicators. The result shows that the developed two input variable model has provided excellent estimations for Madurai and also other six Indian climatic locations of latitudes ranging from 9°N to 13°N. This work also projects a detailed analysis of HPS design based on three meteorological data inputs to cover the residential energy demand. The NASA data, predicted ANFIS data (estimated data) and the measurement data are used in the analysis for the HPS feasibility and size optimization. The analysis shows that the hybrid configurations are same for all the locations, whereas PV and converter sizes are varied in the range of 47–52 kW and 23–27 kW, respectively.

Hybrid models for global solar radiation prediction: a case study

ABSTRACTThis paper presents a comparison between different prediction models for solar radiation application. The present study assessed the performance of multi-layer perceptron (MLP) as well as boosted decision tree, and used a new combinition of these models with linear regression for the prediction of daily global solar irradiation (DGSR). The performance of the studied models was validated using a real dataset measured at the Applied Research Unit for Renewable Energies (URAER) situated in the south of Algeria. Different input combinations have been analysed in order to select the relevant input parameters for DGSR prediction. The results acheived show that the MLP model perfoms better than the others models in terms of statistical indicators: normalised root mean square error (0.033) and R2 (97.7%).

Comparison of Support Vector Machine and Extreme Gradient Boosting for predicting daily global solar radiation using temperature and precipitation in humid subtropical climates: A case study in China

The knowledge of global solar radiation (H) is a prerequisite for the use of renewable solar energy, but H measurements are always not available due to high costs and technical complexities. The present study proposes two machine learning algorithms, i.e. Support Vector Machine (SVM) and a novel simple tree-based ensemble method named Extreme Gradient Boosting (XGBoost), for accurate prediction of daily H using limited meteorological data. Daily H, maximum and minimum air temperatures (Tmax and Tmin), transformed precipitation (Pt, 1 for rainfall > 0 and 0 for rainfall = 0) and extra-terrestrial solar radiation (H0) during 1966–2000 and 2001–2015 from three radiation stations in humid subtropical China were used to train and test the models, respectively. Two combinations of input parameters, i.e. (i) only Tmax, Tmin and Ra, and (ii) complete data were considered for simulations. The proposed machine learning models were also compared with four well-known empirical models to evaluate their performances. The results suggest that the SVM and XGBoost models outperformed the selected empirical models. The performance of the machine learning models was improved by 5.9–12.2% for training phase and by 8.0–11.5% for testing phase in terms of RMSE when information of precipitation was further included. Compared with the SVM model, the XGBoost model generally showed better performance for training phase, and slightly weaker but comparable performance for testing phase in terms of accuracy. However, the XGBoost model was more stable with average increase of 6.3% in RMSE, compared to 10.5% for the SVM algorithm. Also, the XGBoost model (3.02 s and 0.05 s for training and testing phase, respectively) showed much higher computation speed than the SVM model (27.48 s and 4.13 s for training and testing phase, respectively). By jointly considering the prediction accuracy, model stability and computational efficiency, the XGBoost model is highly recommended to estimate daily H using commonly available temperature and precipitation data with excellent performance in humid subtropical climates.

New method for estimating daily global solar radiation over sloped topography in China

A new scheme for the estimation of daily global solar radiation over sloped topography in China is developed based on the Iqbal model C and MODIS cloud fraction. The effects of topography are determined using a digital elevation model. The scheme is tested using observations of solar radiation at 98 stations in China, and the results show that the mean absolute bias error is 1.51 MJ m−2 d−1 and the mean relative absolute bias error is 10.57%. Based on calculations using this scheme, the distribution of daily global solar radiation over slopes in China on four days in the middle of each season (15 January, 15 April, 15 July and 15 October 2003) at a spatial resolution of 1 km × 1 km are analyzed. To investigate the effects of topography on global solar radiation, the results determined in four mountains areas (Tianshan, Kunlun Mountains, Qinling, and Nanling) are discussed, and the typical characteristics of solar radiation over sloped surfaces revealed. In general, the new scheme can produce reasonable characteristics of solar radiation distribution at a high spatial resolution in mountain areas, which will be useful in analyses of mountain climate and planning for agricultural production.

Support vector regression methodology for estimating global solar radiation in Algeria

Accurate estimation of Daily Global Solar Radiation (DGSR) has been a major goal for solar energy applications. In this paper we show the possibility of developing a simple model based on the Support Vector Regression (SVM-R), which could be used to estimate DGSR on the horizontal surface in Algeria based only on sunshine ratio as input. The SVM model has been developed and tested using a data set recorded over three years (2005-2007). The data was collected at the Applied Research Unit for Renewable Energies (URAER) in Ghardaia city. The data collected between 2005-2006 are used to train the model while the 2007 data are used to test the performance of the selected model. The measured and the estimated values of DGSR were compared during the testing phase statistically using the Root Mean Square Error (RMSE), Relative Square Error (rRMSE), and correlation coefficient (\(r^{2}\)), which amount to 1.59(MJ/m2), 8.46 and 97,4%, respectively. The obtained results show that the SVM-R is highly qualified for DGSR estimation using only sunshine ratio.

Method and strategy for predicting daily global solar radiation using one and two input variables for Indian stations

Estimation of daily global solar radiation (GSR) for a particular location is a key parameter in modeling and designing any solar energy system. Very limited observatory stations are available to collect solar radiation data. This demands solar radiation estimation for new locations where there is no observatory station. In the present study, solar radiation models are developed for estimating daily GSR for sunny locations of India. Solar radiation data are collected for a period of three years from in situ measurements. Angstrom-Prescott linear correlation and nonlinear correlations such as quadratic, cubic, exponential, and power models are developed based on the bright sunshine duration. In this new approach, the “bright sunshine duration” is estimated by a new variable named “approximate bright sunshine duration.” The developed linear one input parameter model is applied to predict the daily GSR for five other Indian locations, namely, Allahabad, Bhopal, Chennai, Hyderabad, and Mumbai. The developed linear one input parameter model was further modified as a latitude dependent model to improve the accuracy of the model and named as the latitude model. The performance of all the models is analysed by using the statistical tools, namely, Relative Root Mean Square Error, Mean Percentage Error, and Mean Absolute Percentage Error. The latitude model resulted in fair to good estimations for Tiruchirappalli and the other five Indian locations of latitudes ranging from 10°N to 26°N. Further, other solar radiation models from the literature were considered to identify their applicability for the same selected locations. The models recommended for the selected Indian locations are presented. Out of all the universal models, the monthly mean daily GSR model of Page and out of all Indian models, the monthly mean daily GSR models of Katiyar and Mani showed satisfactory performance for the majority of selected Indian locations.

Evaluation of different sunshine-based models for predicting global solar radiation – case study: New Borg El-Arab city, Egypt

The unavailability of the solar radiation measurements for different locations around the world leads to develop various empirical models to estimate the global solar radiation. In this consider, this study aims to investigate the performance of different solar radiation models to predict the monthly average daily global solar radiation on a horizontal surface. To achieve this, the measured global solar radiation data for a case study location are used. The model predictions are compared with the measured data to introduce the most accurate model for estimating the global solar radiation. The performance of each model is evaluated based on the different statistical indicators. The results show that the Robaa model has the best performance among the other models. Consequently, it can be used for estimating global solar radiation on a horizontal surface in the location under consideration. The accurate estimations of the global solar radiation using this approach can be used in the design and evaluation of performance for different solar applications.

Evaluation and development of temperature-based empirical models for estimating daily global solar radiation in humid regions

In the present study, 14 existing temperature–based empirical models for daily horizontal global solar radiation estimation were comprehensively reviewed, and six new temperature–based models (P1P6) were developed for solar radiation estimation in humid regions. The accuracy and suitability of these models were further evaluated for the humid subtropical and tropical regions of China as a case study using meteorological data during 1966–2015 from 20 radiation stations. The results indicated that despite most studied models were capable of estimating daily global solar radiation with reasonable accuracy, the accuracy of the single temperature-based models was much improved when daily precipitation and relative humidity were included in the models. The newly proposed polynomial model P2 was the top-ranked single temperature-based model for most radiation stations and thus recommended at sites where only air temperatures were available. In terms of complex temperature-based models, the proposed models P3 and P5 by incorporating a proposed logarithmic form of precipitation and the actual value of relative humidity were superior to the existing complex models. Generally, the proposed temperature–based models can be applied for daily global solar radiation estimation with higher accuracy in humid subtropical and tropical regions of China and maybe elsewhere with similar climates.

Models for forecasting monthly mean daily global solar radiation from in-situ measurements: Application in Tropical Climate, India

Radiation availability varies with reference to the geometrical location. It is necessary to develop the models to predict the daily global solar radiation (GSR) for the locations where there is a huge potential of radiation exists and to improve the utilization of energy. This study aimed to develop the solar radiation models with readily available input parameters for Tiruchirappalli, which covers an annual mean daily global solar radiation of 4.50kWh/m2-day.Exponential model (EM), modified exponential model (MEM), power series model (PSM), Hargreaves and Samani model (HSM), Hargreaves-model (HGM) and Hunt models (HUM) which uses air temperature and extra-terrestrial radiation on a horizontal surface as input parameters were developed and validated. Dumas model which uses air temperature and maximum possible sunshine duration (So) as input parameters was also verified. Dumas model was modified as hybrid models to check for their applicability. The performance of the models was analyzed by using the statistical tools. Out of all the models, HSM and PSM had shown an accurate estimation of daily GSR (errors: 6.62% & 7.95% respectively) for the selected location. Dumas model's performance was also found satisfactory (error: 11.02%).

New combined models for estimating daily global solar radiation based on sunshine duration in humid regions: A case study in South China

In the present study, two single (T1 and T2) and eight combined (C1–C8) sunshine-based empirical models for estimating daily horizontal global solar radiation were compared, and four new combined sunshine-based models (P1–P4) were developed. The accuracy and applicability of these models were further evaluated in South China with humid subtropical and tropical climates influenced by the Asian monsoon as a case study. For this purpose, daily meteorological data during 1966–2000 from 20 radiation stations were used for model calibration and the data from 2001 to 2015 were used to validate the models. These models were evaluated and compared based on coefficient of determination (R2), root mean square error (RMSE), normalized root mean square error (NRMSE), mean bias error (MBE) and a global performance index (GPI). The single sunshine-based model T2 was recommended for Rs estimation with acceptable accuracy when using only sunshine duration data. In terms of combined sunshine-based models, the newly proposed model P3 was the optimum model for daily global solar radiation estimation when sunshine duration, air temperature and precipitation were available. The proposed complex model P4 was recommended for estimating daily global solar radiation with an excellent accuracy when vapor pressure deficit and relatively humidity were further available. Generally, the proposed combined models based on sunshine duration can be applied for daily global solar radiation estimation with high accuracy in humid regions of China and maybe elsewhere with similar climate.

Development and Analysis of a New Solar Radiation Atlas for Argentina from Ground-Based Measurements and CERES_SYN1deg data

Currently, quantifying global solar radiation at surface in Argentina is crucial for the development of projects related to solar energy, calculation of evapotranspiration and eco-sustainability architecture, among other environmental issues. In recent years, several models have been developed to estimate the solar energy resources by means of various techniques, e.g. satellite imaging, kriging, or Artificial Neural Networks. The use of satellite data allows for a better spatial representation, being of great relevance in areas with lack of terrain measurements. In this paper, we use the CERES_SYN1deg to develop a new Global Solar Radiation Atlas for Argentina. In this study, we developed maps of annual and monthly mean daily global solar radiation using CERES_SYN1deg data between 2000 and 2016. In order to validate the global solar radiation data provided by CERES_SYN1deg, they were compared with ground-based measurements in the time overlap of both instruments, in four monitoring sites of the SAVER-Net project and an additional site in Tandil, which belongs to the Remote Sensing Group of IHLLA. The maps show the spatial and temporal variation of global solar radiation in Argentina. Comparisons with ground-based pyranometers reveal relative differences of around 3% at a monthly scale for all sites, while the biases can be neglected. Therefore, it is possible to conclude that the maps could be very useful for different technical and scientific purposes, and the comparison with ground-based data demonstrates CERES_SYN1deg’s reliability.

Classification of solar radiation zones and general models for estimating the daily global solar radiation on horizontal surfaces in China

Classification of solar radiation zones constitutes the prerequisite for the establishment of regional daily global solar radiation (H) estimation general model. Current zone establishment methods are ordinary based on solar radiation observation stations (SROS) which present a sparse and non-uniform distribution. As a result, the possibility of misclassifications of the stations occurs in cases there are no records of radiation and far away from SROS. Therefore, by using k-means cluster and Support Vector Machine-Genetic Algorithm, a two-step radiation zoning method was proposed in this paper according to: (a) H, sunshine duration, temperature and relative humidity from 98 SROS and (b) sunshine duration, temperature and relative humidity from 562 stations without radiation. The method is capable to combine the SROS and the stations without radiation in the process of classification. Thus, these misclassifications have been effectively reduced and the accuracy of each classification has been significantly improved. Based on the method, five radiation zones have been identified. Concurrently, four sunshine-based models were obtained for each SROS and the analysis of statistical indexes indicated that the cubic models presented the best performances in each zone. According to the best site-specific models and radiation zones, the general models of regional H estimation were developed by introducing the geographical parameters, including latitude and altitude. The comparative results demonstrated that the general models proposed in this paper had better accuracies and can represent the general models for the H estimation of stations without radiation records in China.