Hourly Global Solar Radiation Research Articles

A Novel Non-Linear Model Based on Bootstrapped Aggregated Support Vector Machine for the Prediction of Hourly Global Solar Radiation

A Novel Non-Linear Model Based on Bootstrapped Aggregated Support Vector Machine for the Prediction of Hourly Global Solar Radiation

Using artificial intelligence for global solar radiation modeling from meteorological variables

Long-term quantification of solar energy variables at ground level is not easily achievable in many locations. In order to overcome this limitation, use of artificial intelligence such as the application of machine learning methods is commonly used for solar irradiance prediction.In this context, this study proposes the implementation of artificial neural networks as deep learning and the XGBoost algorithm as a machine learning method for modeling the hourly global solar radiation for a humid climate such as the Rabat region. For this purpose, hourly meteorological data from the city of Rabat in Morocco are chosen in order of importance using the random forests method, for training and testing the models, namely date and time, sunshine duration, temperature, relative humidity, wind speed/direction and pressure. Subsequently, models are selected after the validation phase for testing, whose performance is evaluated using relevant statistical indicators. As a result, we retain 2 ANN and 1 XGBoost models which are eventually very close in terms of performance with a coefficient of determination value equal to 98% and 97% respectively. However, statistical indicators have proven to be effective in assessing the accuracy and fidelity of each model.Ultimately, the intent of the modeling in terms of accuracy, simplicity or fidelity is a crucial factor in the selection of the model algorithm to adopt.

Calibration of GFS Solar Irradiation Forecasts: A Case Study in Romania

Models based on Numerical Weather Prediction (NWP) are widely used for the day-ahead forecast of solar resources. This study is focused on the calibration of the hourly global solar irradiation forecasts provided by the Global Forecast System (GFS), a model from the NWP class. Since the evaluation of GFS raw forecasts sometimes shows a high degree of uncertainty (the relative error exceeding 100%), a procedure for reducing the errors is needed as a prerequisite for engineering applications. In this study, a deep analysis of the error sources in relation to the state of the atmosphere is reported. Of special note is the use of sky imagery in the identification process. Generally, it has been found that the largest errors are determined by the underestimation of cloud coverage. For calibration, a new ensemble forecast is proposed. It combines two machine learning approaches, Support Vector Regression and Multi-Layer Perceptron. In contrast to a typical calibration, the objective function is constructed based on the absolute error instead of the traditional root mean squared error. In terms of normalized root mean squared error, the calibration roughly reduces the uncertainty in hourly global solar irradiation by 16%. The study was conducted with high-quality ground-measured data from the Solar Platform of the West University of Timisoara, Romania. To ensure high accessibility, all the parameters required to run the proposed calibration procedures are provided.

Simplified Method for Predicting Hourly Global Solar Radiation Using Extraterrestrial Radiation and Limited Weather Forecast Parameters

Solar radiation has important impacts on buildings such as for cooling/heating load forecasting, energy consumption forecasting, and multi-energy complementary optimization. Two types of solar radiation data are commonly used in buildings: radiation data in typical meteorological years and measured radiation data from meteorological stations, both of which are types of historical data. However, it is difficult to predict the hourly global solar radiation, which affects the application of relevant prediction models in practical engineering. Most existing methods for predicting hourly global solar radiation have issues such as difficulty in obtaining input parameters or complex data processing, which limits their practical engineering applications. This study proposed a simplified method to accurately predict the hourly horizontal solar radiation using extraterrestrial solar radiation, weather types, cloud cover, air temperature, relative humidity, and time as the input parameters. The back-propagation network, support vector machine, and light gradient boosting machine (LightGBM) models were used to establish the prediction model, and Shapley additive explanations were used to analyze the relationship between the input variables and the prediction results to simplify the structure of the prediction model. Taking Lanzhou New District in Gansu Province as an example, the results showed that the LightGBM model performed the best, with the root mean square error of 126.1 W/m2. Shapley additive explanations analysis showed that weather type was not a significant factor in the LightGBM model. Therefore, the weather type was removed from the LightGBM model and the root mean square error was 135.2 W/m2. The results showed that extra-terrestrial radiation and limited weather forecast parameters can be used to predict hourly global solar radiation with satisfactory prediction results.

Prediction of Hourly Global Solar Radiation: Comparison of Neural Networks / Bootstrap Aggregating

Prediction of Hourly Global Solar Radiation: Comparison of Neural Networks / Bootstrap Aggregating

Global Solar Radiation Modelling using an Artificial Neural Network for Kazaure, Jigawa State, Nigeria

This research presents an algorithm based on Artificial Neural Networks (ANN), for estimating monthly mean daily and hourly values of solar global radiation. To effectively investigate solar energy consumption and estimate solar renewable energy resources, the Hourly Global Solar Radiation measurements are necessary. In order to predict monthly average daily global sun irradiance on a horizontal area of Kazaure- Nigeria, this study creates a model utilizing ANN to solve the problem of solar energy distribution. Five empirical correlations are developed using the data from 42 months to aid in the prediction of the solar energy distribution pattern. The software is constructed around the Multilayer Perceptron under categorized tabs, with Multilayer perception in neural network Toolbox in MATLAB 9.7 version as a feed forward ANN that maps sets of input data into a set of suitable output. It differs from conventional linear perception by employing three or more layers of neurons (nodes) with nonlinear activation functions. It is also more effective than perceptrons in identifying input that is not linearly separable by a linear hyper-plane. Results obtained utilizing the suggested structure reveals good agreement between the calculated and measured levels of global solar irradiation. The ANN model is shown to be superior when compared to empirical models, due to negligible noise margin.

Long-Term Variations of Global Solar Radiation and Its Potential Effects at Dome C (Antarctica).

An empirical model to predict hourly global solar irradiance under all-sky conditions as a function of absorbing and scattering factors has been applied at the Dome C station in the Antarctic, using measured solar radiation and meteorological variables. The calculated hourly global solar irradiance agrees well with measurements at the ground in 2008–2011 (the model development period) and at the top of the atmosphere (TOA). This model is applied to compute global solar irradiance at the ground and its extinction in the atmosphere caused by absorbing and scattering substances during the 2006–2016 period. A sensitivity study shows that the responses of global solar irradiance to changes in water vapor and scattering factors (expressed by water vapor pressure and S/G, respectively; S and G are diffuse and global solar irradiance, respectively) are nonlinear and negative, and that global solar irradiance is more sensitive to changes in scattering than to changes in water vapor. Applying this empirical model, the albedos at the TOA and the surface in 2006–2016 are estimated and found to agree with the satellite-based retrievals. During 2006–2016, the annual mean observed and estimated global solar exposures decreased by 0.05% and 0.09%, respectively, and the diffuse exposure increased by 0.68% per year, associated with the yearly increase of the S/G ratio by 0.57% and the water vapor pressure by 1.46%. The annual mean air temperature increased by about 1.80 °C over the ten years, and agrees with the warming trends for all of Antarctica. The annual averages were 316.49 Wm−2 for the calculated global solar radiation, 0.332 for S/G, −46.23 °C for the air temperature and 0.10 hPa for the water vapor pressure. The annual mean losses of solar exposure due to absorbing and scattering substances and the total loss were 4.02, 0.19 and 4.21 MJ m−2, respectively. The annual mean absorbing loss was much larger than the scattering loss; their contributions to the total loss were 95.49% and 4.51%, respectively, indicating that absorbing substances are dominant and play essential roles. The annual absorbing, scattering and total losses increased by 0.01%, 0.39% and 0.28% per year, respectively. The estimated and satellite-retrieved annual albedos increased at the surface. The mechanisms of air-temperature change at two pole sites, as well as a mid-latitude site, are discussed.

Long-Term Variations of Global Solar Radiation and Atmospheric Constituents at Sodankylä in the Arctic

An empirical model of global solar irradiance (EMGSI) under all sky conditions was developed by using solar radiation and meteorological parameters at Sodankylä. The calculated hourly global solar irradiance is in agreement with that observed at the ground during 2008–2011 and at the top of the atmosphere (TOA). This model is used to calculate the global solar irradiance at the ground and its attenuation in the atmosphere due to absorbing and scattering substances in 2000–2018. The sensitivity test indicates that the responses of global solar irradiance to changes in water vapor and scattering factors are nonlinear and negative, and global solar irradiance is more sensitive to changes in scattering (expressed by the scattering factor S/G, S and G are diffuse and global solar radiation, respectively) than to changes in water vapor. Using this empirical model, we calculated the albedos at the TOA and the surface, which are in agreement with the satellite-retrieved values. A good relationship between S/G and aerosol optical depth (AOD) was determined and used to estimate AOD in 2000–2018. An empirical model for estimation of tropospheric NO2 vertical column density (VCD) was also developed and used to calculate tropospheric NO2 VCD in 2000–2018. During 2000–2018, the estimated global solar irradiance decreased by 0.92%, and diffuse irradiance increased by 1.28% per year, which is ascribed to the increases of S/G (1.73%) and water vapor (0.43%). Annual surface air temperature increases by 0.07 °C per year. Annual mean loss of global solar irradiance caused by absorbing and scattering substances and total loss are 1.94, 1.17 and 3.11 MJ m−2, respectively. Annual mean losses of absorbing and scattering global solar irradiance show negative and positive trends, respectively, and the annual total loss increases by 0.24% per year. Annual mean losses due to absorption were much larger than those due to scattering. The calculated albedos at the TOA are smaller than at the surface. The calculated and satellite-retrieved annual albedos decrease at the TOA and increase at the surface. During 2000–2018, annual means of the AOD and the tropospheric NO2 VCD increased by 8.23% and 0.03% per year, respectively.

Estimation of the Hourly Global Solar Irradiation on the Tilted and Oriented Plane of Photovoltaic Solar Panels Applied to Greenhouse Production

Agrometeorological stations have horizontal solar irradiation data available, but the design and simulation of photovoltaic (PV) systems require data about the solar panel (inclined and/or oriented). Greenhouses for agricultural production, outside the large protected production areas, are usually off-grid; thus, the solar irradiation variable on the panel plane is critical for an optimal PV design. Modeling of solar radiation components (beam, diffuse, and ground-reflected) is carried out by calculating the extraterrestrial solar radiation, solar height, angle of incidence, and diffuse solar radiation. In this study, the modeling was done using Simulink-MATLAB blocks to facilitate its application, using the day of the year, the time of day, and the hourly horizontal global solar irradiation as input variables. The rest of the parameters (i.e., inclination, orientation, solar constant, albedo, latitude, and longitude) were fixed in each block. The results obtained using anisotropic models of diffuse solar irradiation of the sky in the region of Castile and León (Spain) showed improvements over the results obtained with isotropic models. This work enables the precise estimation of solar irradiation on a solar panel flexibly, for particular places, and with the best models for each of the components of solar radiation.

A profile-free non-parametric approach towards generation of synthetic hourly global solar irradiation data from daily totals

Solar radiation is an essential input in the design and operation of many engineering systems. However, access to high-resolution data (hourly or sub-hourly) is usually limited, especially in developing countries, either due to its unavailability or expensive costs. A novel data-driven approach is proposed to predict the hourly global irradiation profiles from the cheaper and more likely available records of daily global irradiation. The proposed approach is based on a prior categorization of hourly observations using the K-means clustering algorithm, followed by non-parametric function approximation using the multi-layered perceptron artificial neural network. This approach is applied to measured data (130,000 data points) at six locations in the North African Sahara, and the developed models are benchmarked against all existing parametric models in the literature. The artificial neural network-based models outperformed all existing models, with maximum and minimum coefficients of determination of 0.960 and 0.930, respectively. The non-parametric models also captured the true asymmetric profiles of hourly irradiation with enhanced distributions of the residuals. Hence, the suggested models can be used to generate synthetic hourly data for multiple applications, most notably for building energy simulations and scheduling the operation of power generation systems.

A Deep Learning Algorithm for Solar Radiation Time Series Forecasting: A Case Study of El Kelaa Des Sraghna City

Nowadays, the studies that address solar radiation (SR) forecasting tend to focus on the implementation of conventional techniques. This provides good results, but researchers should focus on the creation of new methodologies that help us in going further and boost the prediction accuracy of SR data. The prime aim of this research study is to propose an efficient deep learning (DL) algorithm that can handle nonlinearities and dynamic behaviors of the meteorological data, and generate accurate real-time forecasting of hourly global solar radiation (GSR) data of the city of El Kelaa des Sraghna (32°2’53”N 7°24’30”W), Morocco. The proposed DL algorithm integrates the dynamic model named Elman neural network with a new input configuration-based autoregressive process in order to learn from the seasonal patterns of the historical SR measurements, and the actual measurements of air temperature. The attained performance proves the reliability and the accuracy of the proposed model to forecast the hourly GSR time series in case of missing values detection or pyranometer damage. Hence, electrical power engineers can adopt this forecasting tool to improve the integration of solar power resources into the power grid system.

Binding data mining and expert knowledge for one-day-ahead prediction of hourly global solar radiation

A new methodology to predict one-day-ahead hourly solar global radiation is proposed in this paper. This information is very useful to address many real problems; for instance, energy-market decision making is one of the contexts where that information is essential to ensure the correct integration of grid-connected photovoltaic solar systems. The developed methodology is based on the contribution of different experts to obtain improved data-driven models when included in the data mining process. The modelling phase, when models are induced and new patterns can be identified, is the one that most benefits from that expert knowledge. In this case, it is achieved by combining clustering, regression and classification methods that exploit meteorological data (directly measured or predicted by weather services). The developed models have been embedded in a prediction system that offers reliable forecasts on next-day hourly global solar radiation. As a result of the automatic learning process including the knowledge of different experts, 14 different types of day were identified based on the shape of hourly solar radiation throughout a day. The conventional definitions of types of days, that usually consider 4 options, are updated with this new proposal. The next-day prediction of hourly global radiation is obtained in two phases: in the first one, the next-day type is obtained from among the 14 possible types of day; in the second one, values of hourly global radiation are obtained using the centroid of the predicted type of day and extraterrestrial solar radiation. The relative root mean square error of the prediction model is less than 20%, meaning a significant reduction compared to previous models. Moreover, the proposed models can be recognized in the context of eXplainable Artificial Intelligence.

A Proposed Model to Forecast Hourly Global Solar Irradiation Based on Satellite Derived Data, Deep Learning and Machine Learning Approaches

A Proposed Model to Forecast Hourly Global Solar Irradiation Based on Satellite Derived Data, Deep Learning and Machine Learning Approaches

Hourly global solar forecasting models based on a supervised machine learning algorithm and time series principle

In this paper, predictive models of hourly global solar radiation (HGSR) at one-hour step ahead have been developed by adopting a new methodology. It consists of the association of a supervised machine learning algorithm named Support Vector Machine (SVM) with time series principle. A dataset of HGSR, collected at Ghardaïa city in the south of Algeria, has been used. In order to evaluate the accuracy of the developed models, we have used the Kolmogorov–Smirnov Integral (KSI) in addition to the conventional metrics (RMSE, NRMSE, NMBE, MAPE, and R). The results showed that accurate forecasts of HGSR at one-hour step ahead have been obtained by considering only its previous values at two-step forward. The predictive performances of the selected SVM model have been compared with those of some models available in the literature. This work proved the ability of the investigated machine learning algorithm with time series principle in predicting HGSR with good accuracy.

A new artificial multi-neural approach to estimate the hourly global solar radiation in a semi-arid climate site

In this study, we propose a new hybrid machine-learning algorithm called artificial multi-neural approach, for developing a multi-input single-output (MISO) model to estimate the hourly global solar radiation (HGSR) time series in Agdal site (latitude 31° 37′ N, longitude 08° 01′ W, elevation 466 m), Marrakesh, Morocco. To achieve this goal, three training algorithms (scaled conjugate gradient, Levenberg-Marquardt, and resilient backpropagation) are selected to train the developed model, using some accessible hourly meteorological variables recorded during 7 years (from 2008 to 2014) as exogenous inputs. Furthermore, a pertinence determination test (PDT) is used to point out the most pertinent inputs for the accurate HGSR estimation. The best configuration of the multi-neural model includes five known input parameters as the best scenario, i.e., acquisition hour, air temperature, relative humidity, wind speed, and precipitation. Some statistical indicators are used to evaluate the performance of the developed model. The obtained results demonstrate the reliability and the precision of our proposed machine-learning algorithm, representing a good solution for estimating solar radiation time series needed to design and manage solar energy systems compared to the present standards.

A deep learning algorithm to estimate hourly global solar radiation from geostationary satellite data

To apply deep learning technique for estimating hourly global solar radiation (GSR) from geostationary satellite observations, a hybrid deep network is proposed, relying on convolutional neural network (CNN) to extract spatial pattern from satellite imagery, multi-layer perceptron (MLP) to link the abstract patterns and additional time/location information to target hourly GSR. Its representative advantage lies in the ability to characterize changeable cloud morphology and simulate complex non-linear relationships. The deep network is trained using ground measured GSR values at 90 Chinese radiation stations in 2008 as well as the radiative transfer model simulation at the top of Mt. Everest which serves as constraints of extrapolation for high elevation regions. The extensibility of trained network is validated at 5 independent stations in 2008, yielding an overall coefficient of determination (R2) of 0.82, and at all stations in 2007 along with an R2 of 0.88. Comparative experiments confirm that the combination of spatial pattern and point information can lead to more accurate estimation of hourly GSR, achieving a minimum root mean square error (RMSE) of 84.18 W/m2 (0.30 MJ/m2), 1.92 MJ/m2 and 1.08 MJ/m2 in hourly, daily total and monthly total scales, respectively. Moreover, the deep network is capable of mapping spatially continuous hourly GSR which reflects the regional differences and reproduce the diurnal cycles of solar radiation properly.

Improvement of the hourly global solar model and solar radiation for air-conditioning design in China

Hourly global solar radiation is a significant parameter for building thermal simulation and building cooling design. Most of the solar radiation models focus on the monthly or daily values, but very few studies have been carried out on the hourly values. The Zhang and Huang model was developed for estimating the hourly solar radiation based on the dry-bulb temperature difference, relative humidity and cloud cover. In this study, a new model is proposed for all-sky conditions based on the hourly/daily radiation ratio retrieved from the Zhang and Huang model and the measured daily solar radiation dataset. Compared with the Zhang and Huang model, the proposed model is more accurate (the average of R2, 0.90) and the average of RMSE is improved by 29.54%. The latest hourly solar radiation dataset for 17 locations in China from 2006 to 2016 is subsequently established using the proposed model, which can be used to generate the latest Typical Meteorological Year (TMY). As one of the significant applications to buildings, two kinds of hourly solar radiation datasets (the frequency levels of 95% and 97.5%) for 17 locations are also developed, which are usable for air-conditioning design.

An Advanced Simple Method for Generating Synthetic Average Instant Hourly Solar Energy

The main objective of this study is to generate accurate synthetic hourly solar radiation data by using an easily accessible open source data. In this regard, a new approach is proposed for estimation of synthetic hourly global solar radiation during the day by utilizing only annual solar energy data. First time in literature, a model has been developed for prediction hourly and daily solar radiation based on annual solar energy parameter in this study. Parameters of the model were generated and tested for Turkey and one of them was presented as a case study within this paper. Long term measured hourly horizontal solar irradiance data from a network of Turkish meteorological stations was used to calibrate the model function. The predictions are compared with the solar data available in literature for Turkey. The advanced simple new model is utilized in open source computer program and has the potential to be adapted to other countries.

Algoritmo teórico para la estimación de la radiación solar global para una atmósfera despejada

En este trabajo se desarrolló un algoritmo para la estimación de la radiación solar global, para una atmósfera despejada, mediante el modelamiento de las ecuaciones teóricas de transferencia radiativa de onda corta (0,3 µm - 2,8 µm). Se determinaron las transmitancias espectrales de la radiación solar directa, debidas a la dispersión de Rayleigh y Mie. Mediante el software Propiedades Ópticas de Aerosoles y Nubes (Optical Properties of Aerosols and Clouds, OPAC) se calculó la Profundidad Óptica de los Aerosoles (Aerosol Optical Depth, AOD) para cuatro diferentes tipos de atmósferas, indispensable para la determinación del índice de turbidez de Ångström. Así mismo, se calcularon la transmitancias espectrales por absorción de la radiación solar directa, incluyendo aerosoles, vapor de agua, ozono y aire seco (mezcla de gases). El contenido de O3 se obtuvo de los datos diarios existentes en base de datos de la NASA. Para la componente difusa de la radiación solar, se dedujo una nueva expresión para el cálculo de la fracción de la radiación solar dispersada por los aerosoles hacia la superficie terrestre. La comparación estadística entre los resultados obtenidos con el algoritmo desarrollado, los datos medidos de la radiación global (estación Potsdam - Alemania) y los resultados de otros tres modelos radiativos, entre 2012 y 2014, muestra que el nuevo modelo permite calcular los valores horarios de la radiación solar global con suficiente precisión. © 2018. Acad. Colomb. Cienc. Ex. Fis. Nat.

Evaluation of a cloud cover based model for estimation of hourly global solar radiation in Western Canada

ABSTRACTCloud cover based solar radiation models are relatively simple and convenient as the models require the input of cloud cover data which are mostly available from the meteorological stations. In this study, the performance of a cloud cover based solar radiation model (Kasten–Czeplak model) with original or locally fitted coefficients was evaluated for estimating the hourly global solar radiation for four different locations in Western Canada. The average value of R2, mean bias error, and root mean square error are 0.69, −61.6, and 157.9 W m−2, respectively, for the model with original coefficients, whereas 0.82, 4.4, 107.1 W m−2 with locally fitted coefficients. Results show that the Kasten–Czeplak model with locally fitted coefficients satisfactorily estimated the hourly solar radiation of four different locations in Western Canada. Also, the results indicate that the model with original coefficients has very limited accuracy under intermediate cloud cover conditions.