Direct Normal Irradiance (DNI) is a key variable for solar resource assessment and the design of concentrating solar power plants, yet it is less frequently measured than Global Hor-izontal Irradiance (GHI) due to high costs and operational constraints. To address this gap, a novel two-stage deep learning framework is proposed that first forecasts GHI and then reconstructs DNI through an advanced separation model, eliminating the reliance on dedicated DNI measurements. The framework leverages bidirectional recurrent neural networks (BiLSTM/BiGRU) to capture temporal dependencies, combined with a wrapper-based Mutual Information (WMI) feature selection method to optimally integrate diverse radiometric and atmospheric quantities. Validated on 15-minute data representing arid, dust-prone, tropical, and temperate climates, the WMI-DL consistently and significantly outperformed four conventional empirical separation methods (Erbs, DISC, DIRINT, Engerer4) across RMSD, MAD, MBD, and R² metrics. At Gobabeb and Desert Rock arid sites, RMSD is reduced to 10.4 W/m² and 20.7 W/m², respectively, with R² of 96.6% and 90.5%, while at Tamanrasset, under arid and frequently dusty conditions, RMSD reaches 18.5 W/m² with an R² of 93.7%, markedly outperforming empirical models. Even in cloudy temperate climates, the model achieves RMSD values of 30.5 W/m² in Bermuda and 36.3 W/m² in Palaiseau, with R² ≈ 91%, demonstrating robustness across diverse atmospheric conditions. While direct DNI forecasting with BiGRU achieves slightly higher accuracy, the WMI-DL framework provides a cost-effective, adaptable, and robust solution for high-fidelity DNI estimation, outperforming both conventional separation methods and ECMWF reanalysis benchmarks in regions lacking direct measurements.

This study proposes a novel hybrid model for wind speed forecasting (WSF) based on a three-stage framework comprising decomposition, feature selection, and forecasting. The proposed approach employs Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) to decompose wind speed time series into Intrinsic Mode Functions (IMFs). A distinctive contribution of this study is the use of sample entropy as a feature selection mechanism to identify the most relevant Intrinsic Mode Functions (IMFs). The selected IMFs are then integrated through a classification-based fusion technique, significantly enhancing forecasting accuracy and distinguishing this approach from conventional methods. Two distinct forecasting approaches are evaluated using multiple performance metrics, including RMSE, MAE, MAPE, and R². The first approach applies the fusion technique directly to the original wind speed time series, while the second incorporates ICEEMDAN to decompose the time series. Experimental validation using two real-world datasets from Algeria demonstrates the superiority of the proposed hybrid model over individual forecasting models, yielding significant improvements in prediction accuracy, robustness, and stability. These findings underscore the effectiveness of the three-stage framework, offering a reliable and efficient solution for short-term wind speed forecasting, with potential applications in renewable energy management and grid optimization.

This study presents a novel framework for the optimal design of an off-grid residential energy system, applied to the hyper-arid region of Tamanrasset, Algeria. The proposed hybrid renewable energy system (HRES) integrates photovoltaic panels, wind turbines, battery storage, and diesel generators. A key innovation is the integration of a green energy forecasting module within a multi-objective techno-economic optimization process. Various machine learning and deep learning models suited for time series prediction were evaluated, and the best-performing models for each meteorological parameter were selected which enables precise long-term hourly forecasts, improving system design and operational efficiency compared to conventional methods based on historical averages. The optimization targets three objectives: minimizing the Levelized Cost of Energy (LCOE), reducing the Loss of Power Supply Probability (LPSP), and maximizing the Reliability Factor (RF). Using the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm, the system achieves an LCOE of $0.05433/kWh, an LPSP of 3.1 %, and an RF of 98 %, indicating a strong balance between cost and reliability. Energy contributions are 47 % from solar PV, 35 % from batteries, 12 % from wind, and 6 % from diesel. Comparison with HOMER Pro simulations confirms the superior economic performance of the MOPSO-based configuration. Sensitivity analyses underscore the critical role of forecast accuracy in HRES performance, while environmental assessments show an 80 % reduction in CO₂ emissions compared to diesel-only systems. The integrated forecasting module serves as a valuable decision-support tool for rural electrification, particularly in resource-constrained and climate-challenged regions.

In this paper, Response Surface Methodology (RSM) with optimization is proposed to determine the optimal annual energy output of a small-scale residential bifacial PV system, considering constraints input parameters such as albedo type, module height, tilt angle, and spacing. This could be done by quantifying the relationship between the variable input parameters and the corresponding output parameter where RSM is integrated with PVsyst simulation tool for the analysis. The proposed Box-Behnken design (BBD) optimization required 29 runs for data acquisition and modelling the response surface and Design-Expert software was used to design the experiments and randomize the runs. Regression model was developed and its adequacy was verified to predict the output value at nearly all conditions. The optimization study shows that the system produces the maximum yearly energy when the ground albedo is 0.71, module height is 1.68 m, tilt angle is 19.42°, and module spacing is 7.42 m. These characteristics were found using the meteorological conditions in Tamanrasset, Algeria.

Given the significant importance of renewable or alternative energies today, extensive research is being conducted to enhance the efficiency and reduce the costs of utilizing these energy sources. Among these studies, solar energy forecasting plays a crucial role in achieving these objectives. Accurate forecasting can optimize energy yield, improve grid management, and facilitate the integration of solar power into existing energy systems, ultimately contributing to more reliable and cost-effective renewable energy solutions. This contribution investigates how data volume influences forecasting accuracy. In particular, the impacts on forecasting accuracy of varying forecast horizons and optimal data splitting for training and testing phases are examined. Additionally, the effects on forecasting Global Horizontal Irradiance (GHI) of clustering the data into 3, 5, or 7 groups using the K-means algorithm are investigated. Five different predictive models are employed— multi layer perceptron (MLP), support vector regression (SVR), random forest (RF), convolutional neural network (CNN), and long short-term memory (LSTM)— alongside the newly proposed kResCLSTM hybrid method. Using GHI observations at an arid site in southern Algeria, it is found that a 10-year time series is optimal, along with a 60%-40% split in it for the training vs. testing periods.

The first objective of this study is to estimate the potential of solar irradiance in Algeria using an artificial intelligence approach, namely, convolutional neural networks (CNNs). The second objective is to visualize the different solar components DNI, DHI, and GHI as spatial maps. Data from 267 locations for 2005–2022, obtained from the NREL database, subdivided into training and testing data are used to build the different forecasting models. Testing data are not used when training CNNs to provide an indication of the performance at unknown locations. CNN models with 9 input variables average temperature, relative humidity, sunshine duration, wind speed, precipitation, longitude, latitude, altitude, and month were used to estimate the monthly values of GHI, DNI, and DHI. Statistical error analysis is conducted using mean absolute error (MAE), Mean Absolute Percentage Error (MAPE), Mean Bias Error (MBE) and coefficient of determination R². This study shows that CNNs can be a better solution to estimate solar irradiance data. Seasonal solar mapping was developed in a GIS environment, representing locations and values of solar irradiance. These maps offer valuable insights into solar energy resources, aiding in the implementation of solar energy systems.

Sustainable energy sources like solar and wind speed provide an economically efficient source of energy. Prediction of the output of renewable energy plays a crucial role in shaping decisions concerning electrical system operation and management. Forecasting precision in renewable energy output is essential to ensuring the reliability and stability of the grid, as well as for mitigating risks and minimizing costs within the energy market and power systems. Various statistical techniques were developed to predict solar radiation and wind speed for this purpose and there are two types approaches commonly used: Deep learning and artificial Neural network (ANN). This work propose the used of three statistical methods based in Elman Recurrent Neural Network (ERNN), Long Short Term Memory (LSTM) and Support Vector Machine (SVM) to forecast the output data in different forecasting horizons. Four evaluation deferent metric are used: Forecast skill (FS), Root mean square error (RMSE), Mean Absolute Error (MAE) and coefficient of determination ( R²). These metrics confirm the robustness and accuracy of the LSTM model, validated by its RMSE, MAE, FS, and R² values for both sites. These performances demonstrate the effectiveness of LSTM in capturing temporal patterns, with significant implications for weather forecasting and renewable energy applications.

One primary cause of degradation for materials exposed to the outdoors is ultraviolet (UV) light. To estimate their rate of degradation and determine their longevity (service life), truthful quantification of the UV irradiation incident on their surfaces is needed. In this paper, several neural models are proposed to estimate three UV components based on the broadband global horizontal irradiance and various atmospheric constituents. Using spectral results obtained with the SMARTS model for a large variety of atmospheric conditions and sun positions, three groups of experiments are carried out: the first one is designed to estimate the total UV irradiance at the surface (280–400 nm), whereas the second and third groups aim at estimating its UV-A and UV-B fractions, respectively. The results generally show a good prediction performance, particularly for the first two groups.

Nowadays, the integration of solar power into the electrical grids is vital to increase energy efficiency and profitability. Effective usage of the instable solar production of photovoltaic (PV) systems necessitates trustworthy forecasting information. Actually, this addition can gives an ameliorated service quality if the solar radiation variation can be forecasted accurately. In this paper, we propose a new forecasting approach that integrates Autoregressive Moving Average (ARMA) and Genetic algorithms (GA) to make benefit of both of them in order to forecast Global Horizontal Irradiance (GHI) component. The proposed approach is compared with the standard ARMA model. The experimental results show that, the proposed approach outperforms the classical ARMA models in terms of mean absolute percentage error (MAPE), root mean squared error (RMSE) coefficient of determination (R)² and the normalized mean squared error (NMSE).

This contribution proposes a novel solar time series forecasting approach based on multimodel statistical ensembles to predict global horizontal irradiance (GHI) in short-term horizons (up to 1 hour ahead). The goal of the proposed methodology is to exploit the diversity of a set of dissimilar predictors with the  purpose of increasing the accuracy of the forecasting process. The performance of a specific multimodel ensemble forecast showing an improved forecast skill is demonstrated and compared to a variety of individual single models. The proposed system can be applied in two distinct ways. The first one incorporates the forecasts acquired from the different forecasting models constituting the ensemble via a linear combination (combination-based). The other one consists of a novel methodology that delivers as output the forecast provided by the specific model (involved in the ensemble) that delivers the maximum precision in the zone of the variable space connected with the considered GHI time series (selection-based approach). This forecasting model is issued from an appropriate division of the variable space. The efficiency of the proposed methodology has been evaluated using high-quality measurements carried out at 1min intervals at four radiometric sites representing widely different radiative climates (Arid, Temperate, Tropical, and High Albedo). The obtained results emphasize that, at all sites, the proposed multi-model ensemble is able to increase the accuracy of the forecasting process using the different combination approaches, with a significant performance improvement when using the classification strategy.

Nowadays, the real life constraints necessitates
controlling modern machines using human intervention
by means of sensorial organs. The voice is one of the human
senses that can control/monitor modern interfaces.
In this context, Automatic Speech Recognition is principally
used to convert natural voice into computer text as
well as to perform an action based on the instructions
given by the human. In this paper, we propose a general
framework for Arabic speech recognition that uses Long
Short-Term Memory (LSTM) and Neural Network (Multi-
Layer Perceptron: MLP) classifier to cope with the nonuniform
sequence length of the speech utterances issued
fromboth feature extraction techniques, (1)Mel Frequency
Cepstral Coefficients MFCC (static and dynamic features),
(2) the Filter Banks (FB) coefficients. The neural architecture
can recognize the isolated Arabic speech via classification
technique. The proposed system involves, first, extracting
pertinent features from the natural speech signal
using MFCC (static and dynamic features) and FB. Next,
the extracted features are padded in order to deal with the
non-uniformity of the sequences length. Then, a deep architecture
represented by a recurrent LSTM or GRU (Gated
Recurrent Unit) architectures are used to encode the sequences
ofMFCC/FB features as a fixed size vector that will
be introduced to a Multi-Layer Perceptron network (MLP)
to perform the classification (recognition). The proposed
system is assessed using two different databases, the first
one concerns the spoken digit recognition where a comparison
with other related works in the literature is performed,
whereas the second one contains the spoken TV
commands. The obtained results show the superiority of
the proposed approach.

Accurate solar irradiance forecasts are now key to successfully integrate the (variable) production from large solar energy systems into the electricity grid. This paper describes a wrapper forecasting methodology for irradiance time series that combines mutual information and an Extreme Learning Machine (ELM), with application to short forecast horizons between 5-min and 3-hour ahead. The method is referred to as Wrapper Mutual Information Methodology (WMIM). To evaluate the proposed approach, its performance is compared to that of three dimensionality reduction scenarios: full space (latest 50 variables), partial space (latest 5 variables), and the usual Principal Component Analysis (PCA). Based on measured irradiance data from two arid sites (Madina and Tamanrasset), the present results reveal that the reduction of the historical input space increases the forecasting performance of global solar radiation. In the case of Madina and forecast horizons from 5-min to 30-min ahead, the WMIM forecasts have a better coefficient of determination (R² between 0.927 and 0.967) than those using the next best performing strategy, PCA (R² between 0.921 and 0.959). The Mean Absolute Percentage Error (MAP) is also better for WMIM [7.4–10.77] than for PCA [8.4–11.55]. In the case of Tamanrasset and forecasting horizons from 1-hour to 3-hours ahead, the WMIM forecasts have an R² between 0.883 and 0.957, slightly better than the next best performing strategy (PCA) (R² between 0.873 and 0.910). The Normalized Mean Squared Error (NMSE) is similarly better for WMIM [0.048–0.128] than for PCA [0.105–0.130]. It is also found that the ELM technique is considerably more computationally efficient than the more conventional Multi Layer Perceptron (MLP). It is concluded that the proposed mutual information-based variable selection method has the potential to outperform various other proposed techniques in terms of prediction performance.

The use of wind energy is progressively utilized to produce electrical energy. Wind energy is related to the variation of some atmospheric variables such as wind direction, wind speed, air density and atmospheric pressure. Recently, numerous methods base on Artificial Intelligence techniques to forecast wind speed have been proposed in the literature. In this paper a new artificial intelligence approach for wind speed time series forecasting is proposed, it is composed from two blocs: The first one is based on the use of a deep architecture. The Autoencoder which is a type of deep neural networks, utilized generally for Denoising, is employed to reduce the wind speed input dimensionality. In the second bloc of the proposed methodology, the Elman neural network is employed to forecast future values of wind series, it is a kind of recurrent neural networks that are very sensitive to historical variations. To evaluate our approach we used the following error indicators: Root Mean Square Error (RMSE),Mean Absolute Bias Error (MABE), Mean Absolute Percentage Error (MAPE)and the coefficient of determination (R ² ). The obtained results are compared with those of the Extreme Learning Machine method.