Search

Precise solar radiation forecasting can provide great benefits and solutions for smart grid distribution and electricity management. However, its non-stationary behavior and randomness render its estimation very difficult. In this respect, a new hybrid learning approach is proposed for multi-hour global solar radiation forecasting, relying on Convolutional Neural Network (CNN), Nonparametric Gaussian Process Regression (GPR), Least Support Vector Machine (LS-SVM), and Extreme Learning Machine (ELM) as essence predictors. Then compressive sensing technique is applied to perform a hybridization scheme of the model's output. Hourly global solar radiation data from two sites in Algeria with different climate conditions are used to evaluate the full potential of the integrated model, with stationarity checks with an advanced clear sky model (MecClear model). Different comparative simulations show the superiority of the proposed pipeline in forecasting hourly global solar radiation data for multi-hour ahead compared to the stand-alone model. Experimental results show that the proposed hybridization methodology can effectively improve the prediction accuracy and outperforms benchmarking models during all the forecasting horizons. [ABSTRACT FROM AUTHOR]

This article proposes a new hybrid least squares-support vector machine and artificial bee colony algorithm (ABC-LS-SVM) for multi-hour ahead forecasting of global solar radiation (GHI) data. The framework performs on training the least squares-support vector machine (LS-SVM) model by means of the ABC algorithm using the measured data. ABC is developed for free parameters optimization for the LS-SVM model in a search space so as to boost the forecasting performance. The developed ABC-LS-SVM approach is verified on an hourly scale on a database of five years of measurements. The measured data were collected from 2013 to 2017 at the Applied Research Unit for Renewable Energy (URAER) in Ghardaia, south of Algeria. Several combinations of input data have been tested to model the desired output. Forecasting results of 12 h ahead GHI with the ABC-LS-SVM model led to the root-mean-square error (RMSE) equal to 116.22 Wh/m², Correlation coefficient r = 94.3%. With the classical LS-SVM, the RMSE error equals to 117.73 Wh/m² and correlation coefficient r = 92.42%; for cuckoo search algorithm combined with LS-SVM, the RMSE = 116.89 Wh/m² and r = 93.78%. The results achieved reveal that the proposed hybridization scheme provides a more accurate performance compared to cuckoo search-LS-SVM and the stand-alone LS-SVM. [ABSTRACT FROM AUTHOR]

Accurate and effective forecasting of short-term global solar radiation is critical for the development of photovoltaic systems, particularly for integration into existing grid systems. However, its non-stationary characteristics caused by climatic factors make its estimation extremely challenging. In this regard, a newly designed learning technique for multi-hour global solar radiation forecasting is proposed based on a time-varying filter-empirical mode decomposition (TVF-EMD), feature selection, and extreme learning machine (ELM) as an essence regression. The proposed hybridization strategy consists of three main steps for understanding the fundamental behavioral aspects of hourly global solar radiation data. The first phase employs the TVF-EMD algorithm to deal with the variability of global solar radiation data by separating it into a series of more stable and constant subseries. Then, the feature selection step is employed to evaluate and identify distinctive features set from the whole decomposed subseries by means of the RReliefF algorithm. The selected feature sets are used to train and optimize our forecasting extreme learning machine model, then the tuned ELM model is used to assess the forecasting accuracy. The proposed TVF-EMD-RF-ELM model is evaluated and validated in different regions in Algeria with various climate conditions. The forecasting findings of the TVF-EMD algorithm demonstrate high accuracy compared to the recent version of empirical mode decomposition CEEMDAN. Overall forecasting periods, the TVF-EMD-RF-ELM model produces an error less than 8.3% in terms of normalized root mean square error nRMSE in all studied regions. • A detailed review of different decomposition techniques for global solar radiation forecasting. • An improved artificial intelligence model for multi-hour solar radiation forecasting has been developed. • A regional examination is established, with three sites. • The proposed hybrid model is validated against several models. • The suggested model has been validated for its ability to forecast hourly global solar radiation. [ABSTRACT FROM AUTHOR]

In this paper, a new combined model coupling the linear autoregressive moving average (ARMA) model and the nonlinear artificial neural network (ANN) model has been proposed in order to estimate the daily global solar radiation. The main feature of this approach lies in the fact that has the strength to capture the advantages containing in both models. The combined method have been developed and tested using global solar radiation data recorded during two years (2012–2013) for two different climate sites in Algeria. The obtained results showed an improvement of the combined model over ARMA and ANN models in term of mean absolute error (MPE) of about 18.1% and 2.7%, for the first site, of about 27.26% and 1.39% for the second site. Moreover, compared to the ARMA and ANN models, a decrease in the RMSE values of about 17.1% and 3.59% compared to the ARMA and ANN models has been observed. [ABSTRACT FROM AUTHOR]