Estimation Of Solar Radiation Research Articles

Estimation of daily global solar radiation using empirical and machine-learning methods: A case study of five Moroccan locations

Reliable solar radiation data are essential to study the feasibility and to determine the optimal size of solar plants in a particular location. However, the network of solar radiation measurement stations is very weak around the world. This has led to the development of a wide variety of models and techniques for estimating solar radiation from commonly measured meteorological variables (Ambient air temperature, relative humidity, wind velocity …). In this work, 22 empirical models, Artificial Neural Networks (ANN) techniques, and tree-based ensemble methods were tested in estimating the daily global solar radiation (GSR) in five Moroccan locations. Each database was partitioned into two sets: the training set and the validation set. The training set is used to calibrate the models, while the validation set is used to assess their credibility. The best-performing model at each station was selected based on three statistical indictors, the coefficient of correlation (R), the normalized mean absolute error (nMAE) and the normalized root mean square error (nRMSE). The results on the validation dataset revealed that the Random Forest method (R: 87.53–96.20%; nMAE: 5.84–11.81%; nRMSE: 7.85–15.33%) outperformed the all tested models in terms of accuracy. The other machine learning methods achieved generally good performance (R: 81.73–95.14%; nMAE: 5.88–13.86%; nRMSE: 8.22–18%). Among the empirical models (R: 56.58–93.46%; nMAE: 6.96–21.83%; nRMSE: 9.89–26.96%), the Temperature - Geographic factors model (TG1) (R: 72.38–93.46%; nMAE: 6.96–17.94%; nRMSE: 9.89–22.39%) can be recommended for estimating the GSR for the all considered stations.

Impact of global dimming/brightening on estimating Angström-Prescott parameters based on geographically weighted regression in China

Variations of solar radiation is related to energy planning and architectural designing. It is important to estimate solar radiation from other climate factors because of the limited measurement solar radiation data. In this study, temporospatial changes of solar radiation and sunshine hour are analysed for global dimming/brightening periods based on meteorological data in China. Results show that the trends of solar radiation and sunshine hour are both decreasing dramatically for 1961–1990 (the global dimming period) but levelling off for 1991–2010 (the global brightening period), which indicates that sunshine hour can be used the reliable indicator of solar radiation. Spatially, the change trends of solar radiation are consistent with sunshine hour in 36 stations during the period of 1961–1990, but they are consistent in just 27 stations during the period of 1991–2010. To assess the impact of changing solar radiation on Angstrom-Prescott parameters (a and b) in the equation of solar radiation estimation from sunshine hour, geographically weighted regression is developed to calculate a and b. There is not dramatic difference in the spatial patterns of a and b for the period of 1961–2010 and 1961–1990. However, there is significantly inconsistent changes in a and b during the global brightening period compared with the global dimming period, which suggests that global dimming/brightening has important impact on estimating solar radiation from sunshine hour, with dramatic variability among different regions.

Application of machine learning for solar radiation modeling

Solar radiation is an important parameter that affects the atmosphere-earth thermal balance and many water and soil processes such as evapotranspiration and plant growth. The modeling of the daily and monthly solar radiation by Gaussian process regression (GPR) with K-fold cross-validation model has been discussed recently. This study evaluated different neural models such as artificial neural network (ANN), support vector machine (SVM), adaptive network-based fuzzy inference system (ANFIS), and multiple linear regression (MLR) for estimating the global solar radiation (daily and monthly) with K-fold cross-validation method. For the appropriate comparison of the models, the randomized complete block (RCB) design applied in the training and test phases. Also, different data sets were evaluated by K-fold cross-validation in each model. The results showed that radial basis function (RBF) model has the lowest error for estimating the monthly and daily solar radiation. In this study, the result of RBF was compared with the GPR models. The conclusion indicated that RBF methodology can predict solar radiation with higher accuracy relative to the GPR model. The results of yearly solar radiation estimation (2009–2014) showed that the RBF model can estimate solar radiation with the MAPE and RMSE of 5.1% and 0.29, respectively. Also, the coefficient of correlation (R2) between actual and estimated values throughout the year is 98% and can be used by the engineers and other researchers for solar and thermal applications.

Estimation of solar radiation using modern methods

It is stated in the present study that extreme learning machines (ELM) will display a greater performance in solar radiation estimation compared to artificial neural networks (ANN). The data acquired from Karaman province during 2010–2018 were used for evaluating the performance of the suggested approach. It was put forth when results were compared that ELM has displayed a greater estimation performance. Moreover, ANN and ELM were tested with different activation functions in order to obtain the best estimation response. While the best estimation result for ANN was obtained with the tansig function as 0.9828, mean square error (MSE) was obtained as 0.000129. The best estimation result for ELM was obtained with the sin function as 0.991 and MSE was calculated as 0.000881. Additionally ELM, training time 0.295 s, test time 0.266 s, MSE time 0.558 s was obtained. ELM displayed a high estimation performance in a very short amount of time. The ELM achieved a root mean square error (RMSE) value of 0.0297. This algorithm has achieved high accuracy with minimal error. Confidence interval estimations were carried out for the acquired correlation coefficients and the results were compared. ELM estimation performance is better than ANN with 95% confidence interval.

Empirical model for estimation of global solar radiation at lowland region Biratnagar using satellite data

This study proposes to find the regression coefficient of the modified Angstrom type model for the estimation of global solar radiation (GSR) in lowland Biratnagar (Lat. 26.5º N, Long. 87.3º E and Alt. 72m) using relative sunshine duration and satellite data of GSR. Using the regression technique, the empirical constants 0.29 and 0.56 are found in the modified Angstrom model. Furthermore, Modified Angstrom model along with other linear models such as Glover and McCulloch model, Page model, Rietveld model, and Turton's model are statistically assessed to evaluate the significance of models. Statistical tests like MPE, MBE, RMSE, and CC reveal that all these models are statistically significant. These findings can be utilized for other locations with a high confidence level at the similar climatic locations of Nepal.
 BIBECHANA 18 (2021) 193-200

Global Horizontal and Direct Normal Solar Irradiance Modeling by the Machine Learning Methods Xgboost and Deep Neural Networks with CNN-LSTM Layers: A Case Study Using the GOES-16 Satellite Imagery

Restrictive legislations over the usage of fossil fuels are encouraging the research on clean and renewable energies. As a drawback, renewable energies are characterized by having random behavior, which hampers its integration on the current energy base system. Thus, solar irradiation estimation is important to a better assimilation of the renewable energies into the energy matrix. In this paper, two machine learning (ML) estimation models for global (GHI) and direct normal solar irradiance (DNI) are proposed: the first using XGBoost and the second using a hybrid convolutional neural network (CNN) with a long short-term memory (LSTM), CNN-LSTM. Both models use images from the GOES-16 satellite, taken from the city of Petrolina, Brazil. The results of the proposed models are analyzed against the reference models Copernicus Atmosphere Monitoring Service (CAMS), Solcast and the Physical Solar Model (PSM) provided by the National Solar Radiation Database (NSRDB). For GHI estimation, the PSM model reached the smallest RMSE, equal to 147.23 W/m^2, while the model CNN-LSTM was the best one to estimate DNI with a RMSE equal to 238.22 W/m^2. The proposed model managed to improve RMSE when compared against the benchmarking models by 2.89 % and 1.70 % for CNN-LSTM and XGBoost, respectively.

Solar radiation estimates at Meerut workplace

The research is concerned with the assessment and analysis of solar radiation in a workplace with longitude 77.706 (77°43' east) and latitude 28.984 (28°59' north) located in western Uttar Pradesh and the National Capital Region of India. Solar radiation energy which varies according to geographical location and weather conditions and also depends on the variation in components of solar radiation such as global and diffused components. Every year from June to September except the rainy season, the spreading components of solar radiation are more and there is a significant amount of solar energy available in this area. This work estimates the variation in solar components of complete solar radiation and the possible availability of solar radiation energy, may play an important role in reducing the electrical energy crisis in the Meerut region.

Estimation of Global Solar Radiation in Mubi, Nigeria

Application of solar energy system requires having knowledge about solar irradiation potential in different locations. This study therefore used artificial neural networks for predicting solar global radiation by using metrological data. There is no report about prediction of solar radiation potential for Mubi by using Artificial Neural Network (ANN) method. It is very encouraging to observe a very fine agreement between the measured and estimated values shown in study. The ANN Model is considered the best relation for estimating the global solar radiation intensity for Mubi region with an acceptable error. The MSE, RMSE, MBE, MABE and MAPE values are 0.930, 0.964, 0.3358 MJm−2day−1, 0.8175 MJm−2day−1 and 19.30%, respectively. The ANN models appear auspicious for estimating the Global Solar Radiation in the locations where there are no solar radiation measurement stations.

Simple method for the estimation of solar spectral radiation

In order to know the energy production of photovoltaic modules, it is necessary to have an idea of the spectral distribution of solar radiation, because each technology is characterized by a different spectral resolution. The objective of this work is to determine the irradiance in ultraviolet, visible, and infrared from the triangular shape of the global solar spectrum for any sky state. According to statistical indicators, the coefficient of determination R² varies between 97% and 99%.

Solar radiation forecasting by using deep neural networks in Eski̇şehi̇r

According to the World Economic Outlook (WEO), the global demand for energy is presum- ably going to be increased due to growing the world’s population up during the upcoming two decades. As a result of that, apprehensions about environmental effects, which appear as a re- sult of greenhouse gases are grown and cleaner energy technologies are developed. This clearly shows that extended growth of the worldwide market share of clean energy. Solar energy is considered as one of the fundamental types of renewable energy. For this reason, the need for a predictive model that effectively observes solar energy conversion with high performance becomes urgent. In this paper, classic empirical, artificial neural network (ANN), deep neural network (DNN), and time series models are applied, and their results are compared to each other to find the most accurate model for daily global solar radiation (DGSR) estimation. In addition, four regression models have been developed and applied for DGSR estimation. The obtained results are evaluated and compared by the root mean square error (RMSE), rela- tive root mean square error (rRMSE), mean absolute error (MAE), mean bias error (MBE), t-statistic, and coefficient of determination (R2). Finally, simulation results provided that the best result is found by the DNN model.

New Models for Estimation of Solar Radiation for Five Cities of Pakistan

New Models for Estimation of Solar Radiation for Five Cities of Pakistan

Global solar radiation prediction model with random forest algorithm

Global solar radiation estimation is crucial for regional climate assessment and crop growth. Therefore, studies on the prediction of solar radiation are emerging. With the availability of the public data on solar radiation, computerized models have been developed as well. These predictive models play significant role in determining the potentials of regions suitable for renewable energy generation required by engineering and agricultural activities. Herein a computerized model has been presented for estimating global solar radiation. The model utilizes random forest algorithm and reached predictive value of 93.9%.

Evaluation of Multiple Approaches to Estimate Daily Solar Radiation for Input to Crop Process Models

Daily solar radiation is a critical input for estimating plant growth and development, yet this variable is infrequently measured compared to other climate variables. This study evaluates the sensitivity of simulated maize and soybean production from the CERES-Maize and CROPGRO-Soybean modules of the Decision Support System for Agrotechnology Transfer (DSSAT) to daily solar radiation estimates obtained from traditional (stochastic, empirical, and mechanistic models) and non-traditional (satellite estimation, reanalysis datasets, and regional climate model simulations) approaches, using as an example radiation estimates for Hancock, Wisconsin, USA. When compared to observations, radiation estimates obtained from empirical and mechanistic models and a satellite-based dataset generally had smaller biases than other approaches. Daily solar radiation estimates from a reanalysis dataset and regional climate model simulations overestimate incoming daily solar radiation. When the radiation estimates were used as an input to CERES-Maize, no significant differences were found for maize yield obtained from the different radiation estimates compared to yield from observed radiation, even though differences were found in the daily values of leaf area index, crop evapotranspiration, and crop dry weight (biomass). In contrast, significant differences were found in simulated soybean yield from CROPGRO-Soybean for the majority of the radiation estimates.

Effective Estimation of Hourly Global Solar Radiation Using Machine Learning Algorithms

The precise estimation of solar radiation is of great importance in solar energy applications with respect to installation and capacity. In estimate modelling on selected target locations, various computer-based and experimental methods and techniques are employed. In the present study, the Multilayer Feed-Forward Neural Network (MFFNN), K -Nearest Neighbors ( K -NN), a Library for Support Vector Machines (LibSVM), and M5 rules algorithms, which are among the Machine Learning (ML) algorithms, were used to estimate the hourly average solar radiation of two geographic locations on the same latitude. The input variables that had the most impact on solar radiation were identified and grouped as a result of 29 different applications that were developed by using 6 different feature selection methods with Waikato Environment for Knowledge Analysis (WEKA) software. Estimation models were developed by using the selected data groups and all input variables for each target location. The results show that the estimations developed with the feature selection method were more successful for target locations, and the radiation potentials were similar. The performance of the estimation models was evaluated by comparing each model with different statistical indicators and with previous studies. According to the RMSE, MAE, R 2 , and SMAPE statistical scales, the results of the most successful estimation models that were developed with MFFNN were 0.0508-0.0536, 0.0341-0.0352, 0.9488-0.9656, and 7.77%-7.79%, respectively.

Estimation of Solar Radiation on Tilted Surface by Using Regression Analysis at Different Locations in India

This study is to find the regression model for estimation of monthly meanhourly global solar radiation on tilted surface at different locations of India.This study is quite precious due to lack of solar radiation data availability onthe tilted surface. Firstly, we have selected some locations having differentclimatic conditions such as New Delhi, Mumbai, Kolkata, Lucknow andJaipur to find the solar radiation on tilted surface using Liu and Jordan model,HDKR model and Perez model. The mean values of these models are plottedalong with the daytime. Based on regression techniques, four empiricalmodels are developed which are tested to compute the solar radiation ontilted surface for three new stations Ahmadabad, Bangalore and Chennai.The estimated solar radiation by these four developed models are comparedwith the estimated values using existing models Lie & Jordan, HDKR andPerez based on mean bias error (MBE) and root mean square error (RMSE).It has been found that developed Model-3 has minimum error and the valuesestimated this model is comparable to existing models. The maximum values of RMSE in Model-3 in tested stations are 2.01% with Liu and Jordan, 2.63%with HDKR and 2.10% with Perez. Similarly, maximum values of MBE are−1.79% with Liu and Jordan, −2.27% with HDKR and −1.89% with Perez.Now the Model-3 finally selected to determine the solar radiation on Bhopal,Bhubneshwar, Dehradun, Guwahati and Trivendrum (Thiruvananthapuram).

Semiparametric Semivariogram Modeling with a Scaling Criterion for Node Spacing: A Case Study of Solar Radiation Distribution in Thailand

Geostatistical interpolation methods, sometimes referred to as kriging, have been proven effective and efficient for the estimation of target quantity at ungauged sites. The merit of the kriging approach relies heavily on the semivariograms in which the parametric functions are prevalently used. In this work, we explore the semiparametric semivariogram where no close-form semivariogram is required. By additionally enforcing the monotonicity condition in order to suppress the presence of spurious oscillation, a scaling of the nodes of the semiparametric kriging is proposed. To this end, the solar radiation estimates across extensive but unmeasured regions in Thailand using three different semivariogram models are undertaken. A cross validation analysis is carried out in order to justify the performance of each approach. The best results are achieved by the semiparametric model with an improvement of around 7–13% compared to those obtained from the parametric semivariograms.

Possibilities to Estimate Daily Solar Radiation on 2-Axis Tracking Plane Using a Model Based on Temperature Amplitude

Significant errors may occur when estimating daily solar radiation in central Europe using empirical models based on air temperature (especially in the winter months). The first goal of this article is to investigate to what extent it is possible to increase the accuracy of the Hargreaves and Samani model, by using the calibration dedicated for each month. We also corrected the temperature amplitudes by narrowing the daily intervals from which the minimum and maximum values were taken. The second goal of this article is to compare the precision of the daily solar radiation estimation on the horizontal plane and on the 2-axis tracking plane. The database comprises the series of parallel measurements on both planes over a period of 10 years. We considered two procedures, direct and indirect, for the 2-axis tracking plane. The second procedure, dubbed “the first estimate horizontal than calculate tracking” is based on the strong relationship between daily solar radiations on both planes. The direct procedure allows for a slightly more accurate estimation. The estimation of daily radiation on the 2-axis tracking plane reflects the measured values worse than estimation on the horizontal plane. We discovered that the increase of estimation errors on the 2-axis tracking plane, compared to the horizontal one, is proportional to the increase in the coefficient of variability of the daily solar radiations.

Effect of Asynchronous Data Processing on Solar Irradiance and Clearness Index Estimation by Sky Imagery

Among the different available technologies to estimate and forecast solar irradiance, total sky imagers are a suitable option for short-term horizons (less than 30 min). For this purpose, cameras are able to capture images at high speed (less than a second), depending on hardware and configuration, while solarimetric measurements are restricted to the characteristic response time of pyranometers, that varies from 15–60 s, depending on its quality. This difference in sampling time may lead to asynchronous data logging. As a result, two data sets with different sampling time are obtained. A close look into specialized literature showed a lack of evidence on data synchronization of reported results. Thus, the existence of asynchronous data is a plausible assumption. The main objective of this study is to investigate the effect of considering two different approaches in which data can be treated for the estimation of the solar irradiance and clearness index. For this purpose, two data sets are available: solar irradiance with a sampling time of 1 min, and sky images taken at 5 s intervals. Image acquisition system is based on a fisheye camera and a cost-effective and portable device. In one approach solar irradiance is interpolated to image time, and in the second approach, extracted image data is interpolated to solar irradiance time. For each case, solar irradiance and clearness index estimation is based on a multiple linear regression with two different number of features extracted from sky image processing. To evaluate these models, the mean bias error, root mean square error and the mean absolute percentage error were used. Obtained results shows that using different approaches to process data of measurements may lead to errors of 6.2% for the solar irradiance and 3% for the clearness index.

A comprehensive study on modeling of photovoltaic arrays and calculation of photovoltaic potential using digital elevation model

ABSTRACT In solar energy applications, a complete knowledge and detailed analysis of the solar radiation potential is a prerequisite. In this paper, two different photovoltaic (PV) potential analysis approaches are compared. The first approach is based on a simulation of PV array and six different solar radiation estimation models. The outputs of these models are compared with hourly ground measured data by using statistical error methods. The most suitable solar radiation model which gives more accurate results is used to calculate the PV potential of Engineering Faculty of Eskişehir Technical University. The PV panel simulation is implemented in Matlab/Simulink besides that a suitable algorithm is selected to calculate the possible amount of PV panel energy generation based on the hourly measured wind speed (WS), air pressure, global solar radiation (GSR), and temperature values. In the second approach, building roof surface PV potential of the selected area is calculated and the available roof area for PV installation is determined by using Aeronautical Geographical Information System (ArcGIS) software. When the performances of the methods are compared, these methods obtained satisfactory results. Furthermore, it is clear that the algorithms and effective factors of the two methods are different. The results of ArcGIS outperform the PV array simulation based on Badescu model by the reason of its suitability of selecting optimal site for PV installation.

Estimation of Incident Solar Radiation on the Walls of a Habitat for the Typical Day of April in the Ouagadougou Region in Burkina Faso

His work aims to develop a mathematical model of incident solar radiation on all the walls of a sloping roof habitat for a typical climate in the Ouagadougou region. Subsequently, we set up a program for the calculations of the essential parameters of illumination and the various components of solar radiation under the Fortran Programming Language and to plot our curves using the Origin software. This work allowed us to estimate the amount of solar flux that each wall of a habitat receives during a day, to know the importance of the orientation of the main facade of the habitat to the south and to find out how often the roof is exposed to solar radiation.