AI Data Driven Approach-Based Endogenous Inputs for Global Solar Radiation Forecasting

Abstract

Controlling the random nature of renewable energy sources such as solar radiation at ground, allows electric grid operators to better integrate it. In this paper, an intelligent datadriven model based on artificial neural network with autoregressive input sequence is developed to forecast the global solar radiation (GSR) time series on a half hour resolution in the site of Agdal, Marrakesh, Morocco. The database that is used to create this model was divided into two subsets. The first subset is used for training the proposed model on the data measured during the year 2008 by adopting three efficient optimizers (levenbergmarquardt, resilient backpropagation, and scaled conjugate gradient). The second subset is used for testing the efficiency and the robustness of the developed model to generate accurate predictions during the next six years (from 2009 to 2014). The obtained results demonstrate the accuracy and the stability of the proposed data-driven model to perform prediction in case of GSR measurements intermittence or sensor damage.

New Model of Photovoltaic System Adapted by a Digital MPPT Control and Radiation Predictions Using Deep Learning

“…Jallal et al [33] suggested a Deep Neural Network capable of handling the non-linearity and dynamic behavior of meteorological data and providing accurate real-time predictions of hourly global solar radiation. The neural network used hourly data on global solar radiation and meteorological parameters based on the METEONORM data sets of the city of El Kelaa des Sraghna, Morocco.…”

Multi-Target Learning Algorithm for Solar Radiation Components Forecasting Based on the Desired Tilt Angle of a Solar Energy System

“…In the present work, eight training optimizers (Resilient back Propagation (Rp), One step secant (OSS), Levenberg-Marquardt (LM) Algorithm, Fletcher-Reeves algorithm (Cgf), Polak-Ribiere algorithm (Cgp), Powell-Beale algorithm (Cgb), gradient descent (Gd) algorithm and scaled conjugate gradient algorithm (Scg)) are applied to estimate the ENN model parameters [6,17,18], with the mean square error (MSE) that is used as a loss function. More details about the training process of artificial neural networks are presented in Ref.…”

“…The ENN model performance was evaluated using the MSE and the correlation coefficient (R) metrics [19], which are computed by Eq. ( 5) and (6). Subsequently, the influence of the Elman ANN's feedback links order (TDL) on the forecasting accuracy is investigated by changing its value between 1 and 48.…”

“…Jallal et al developed a new machine learning based on the hybridization of deep neural networks with the multimodel approach to predict accurately the hourly global solar radiation (GSR) time series for Marrakech, Morocco, using the most pertinent exogenous inputs measured during seven years (2008)(2009)(2010)(2011)(2012)(2013)(2014) [5]. The same authors have developed an effective AI-data driven approach to forecast the global solar radiation measurements at each half-hour based only on endogenous inputs [6]. They have also developed an autoregressive moving average model based on Box-Jenkins approach to enhance the accuracy of daily global solar radiation prediction [7].…”

The Implementation of Deep Reinforcement Learning in E-Learning and Distance Learning: Remote Practical Work

“…Machine learning is one of the most important branches of artificial intelligence, which is used in a range of fields and disciplines [2][3][4], including the field of education that we will focus on [5].…”