Belkacem Draoui

Efficient water management is essential for sustainable agricultural production, particularly in arid and semi-arid regions where water resources are limited. Machine-learning-based irrigation systems can support automated pump-operation decisions using environmental and soil-related sensor data. However, most previous studies have focused on individual machine-learning models, while the application of stacked ensembles to binary irrigation pump-status prediction remains relatively limited. This study proposes a stacking-based framework for predicting irrigation pump operation in an ON/OFF classification setting. The framework uses environmental and soil-related variables, including soil moisture, temperature, humidity, and a numerical time-related feature. Random forest (RF), extreme gradient boosting (XGBoost), and multilayer perceptron models were trained as base learners, and their out-of-fold (OOF) predictions were combined using a logistic-regression meta-learner. The models were evaluated on a held-out test set using accuracy, precision, recall, specificity, and F1-score. The individual models achieved accuracies ranging from 97.34% to 99.95%, with XGBoost providing the best individual performance. The proposed stacking ensemble achieved 99.97% accuracy, 99.94% precision, 100.00% recall, 99.94% specificity, and a 99.97% F1-score. Compared with XGBoost, the ensemble further refined predictive performance, improving accuracy by 0.02 percentage points and F1-score by 0.01 percentage points while achieving complete elimination of false negatives (100.00% recall). These results demonstrate the potential of stacked ensemble learning to improve binary pump-operation prediction and support data-driven irrigation management in water-limited environments.

Against an unprecedented water crisis in our country, balancing water supply and demand is necessary for a secure and sustainable water supply. This challenge requires systems capable of delivering the necessary quantities while conserving resources. Numerous research initiatives focus on addressing water distribution challenges with the help of smart water systems to optimize network operations and minimize water demand. Based on these advancements, this paper proposes a new smart water distribution system for southwest of Algeria. The system integrates the Internet of Things (IoT), information and communication technologies, and smart technologies to address critical attributes for enhancing efficiency. To achieve the efficient management of two-way flows (both water and data) based on water demand and its availability, two innovative architectures have been proposed, using various measurements of water quantity and quality parameters. Algorithms to automate and optimize water distribution are also proposed. According to obtained results, performance has improved, with an accuracy rate of over 98%. These results establish the suggested system as a strong option for intelligent and sustainable water resource management by demonstrating its efficacy and durability.

Reference evapotranspiration (ET0) is a critical parameter for water resource management and irrigation scheduling. Accurate estimation of ET0 has challenged scientists over the years due to its high sensitivity to climatic variations. Classical methods for estimating ET0 mainly rely on empirical models with a significant number of parameters, which has hampered their use in many cases. Regarding its importance and strong relationship with global food security, this topic has attracted the attention of many researchers. The development of simple models with a low number of parameters or taking advantage of artificial intelligence algorithms has been the aim of different researchers, as it is in this paper, where two approaches are implemented to estimate reference evapotranspiration. The first one is based on the use of classical system identification models, namely linear and nonlinear AutoRegressive models with eXogenous variables (ARX and nonlinear ARX). For the second approach, AI-supported system identification models are used, in which neural networks’ performances are used to develop multilayer and deep neural network models for nonlinear system identification. The four models show a high accuracy, with a system fitting exceeded 98%.

The technical landfill center TLC in Bechar receives every day an enormous quantity of waste estimated at 120 tons, of which organic matter represents more than 60%. The latter is a beneficial substrate for the production of green energy called biogas. Biogas is a renewable energy source, which can be produced by the methanization process which consists of the decomposition of organic matter in the absence of oxygen, also called anaerobic digestion. The present work consists in the valorization of the biogas produced from household waste of the technical landfill center TLC of the city of Bechar, Algeria. The results obtained in this study are to characterize the physicochemical parameters of the experimental analysis. In particular, the daily quantity of biogas (methanisable waste) and that of the electrical energy produced. The application of this work can fill an energy deficit compared to the energy demand in the summer period (air conditioning).

In energy applications, convection plays a dominant role in the transport of energy for the proper design in order to achieve higher heat transfer rates. Where the fluid motion is simply induced by density gradients. This study is concerned with transient natural convection in an isosceles-superimposed triangular enclosure. The inclination of the inclined partitions walls was θ =45°. Critical Rayleigh numbers have been obtained: Ra=106, especially for the fluid crushed between the two triangles where we can visualize the critical convection transaction between the air and water. The results in quantitative measurement of average Nusselt number change and qualitative visualization of streamline and isotherms are examined. The air and water streamlines and the temperature fields were obtained for the Rayleigh numbers 103 ≤ Ra ≤ 108 for both cases studied. Besides, the highest heat transfer performance is observed during the natural convection case of the water fluid, and the lowest heat is transferred when the fluid is dilatant. The triangular shape can greatly enhance the heat transfer from the bottom to the top at first in a smooth and harmonious manner, and then it turns into a more complex and disorderly shape when the water reaches high temperatures.

La climatisation électrique est très énergivore et l’offre ne peut plus satisfaire à la demande accrue surtout dans les saisons chaudes. L’utilisation des systèmes de rafraîchissement par géothermie constitue une bonne alternative notamment dans la région du sud de l’Algérie où la période estivale est très longue et très chaude. Dans ce travail, nous présentons une étude effectuée sur un échangeur géothermique à air destiné au rafraîchissement des bâtiments. Les résultats obtenus sont très encourageants et nous indiquent que cette technique bon marché peut couvrir une bonne partie de nos besoins en matière de climatisations domestiques. Cette étude nous a permis également d’examiner l’influence des paramètres de l’échangeur sur son efficacité moyenne quotidienne et sur la température de l’air refroidi en sortie de l’échangeur.

La serre tunnel agricole classique est largement répandue dans les pays méditerranéens, malgré les insuffisances qu’elle présente, notamment la surchauffe pendant le jour et le refroidissement nocturne intense, qui parfois se traduit par l’inversion thermique interne. La serre chapelle habillée en verre est relativement plus performante, mais son évolution reste lente en raison de son coût d’investissement et son amortissement. Le comportement énergétique d’une serre a fait l’objet de plusieurs études et concerne globalement la nuit et afin de contribuer en une meilleure gestion climatique nous nous sommes proposé d’entreprendre une étude et analyse thermique de la paroi en fonction d’une modélisation. Lors de ce travail, nous avons mis au point un quotient nommé Bibi, qui permet de caractériser un matériau de couverture. Cet état d’évolution thermique est fonction du degré d’étanchéité de la couverture et de ses caractéristiques physiques, pour être transparente aux rayonnements solaires, absorbante et réfléchissante des rayonnements infrarouges émis par le sol, d’où l’effet de piégeage des radiations solaires autrement appelé ‘effet de serre’. Nous proposons dans cet article, la modélisation et l’analyse du comportement thermique de la paroi en polyéthylène ‘PE’ de la serre tunnel expérimentale.

The effect of drying temperature and the application of ultrasound on drying kinetics of Lawsonia inermis (henna) leaves and the extraction of lawsone from the dried samples was addressed. Indeed, henna leaves were dried with and without the application of ultrasound (21.7 kHz, 30.8 kW/m³) at 40, 50 and 60 °C with a constant air velocity (1 m/s). As expected, both the increase of temperature and the application of ultrasound decreased the drying time and increased the rate of extraction of the lawsone. The values of the effective diffusion coefficients obtained were used to quantify this influence showing the value increases with higher drying temperature and the application of ultrasound. Moreover, the influence of temperature was quantified by the estimation of the activation energy from an Arrhenius-type equation (46.25 kJ/mol in the case of drying without ultrasound application and 44.06 kJ/mol in the case of ultrasonically-assisted drying). Regarding the influence of studied variables on lawsone extraction yield, the higher is the temperature, the lower is the yield, probably linked with lawsone degradation reaction due to thermal treatment. On the contrary, the application of ultrasound improved the extraction yield mainly at the lower drying temperature tested of 40 °C.

High accuracy trajectory tracking with constrained inputs is challenging topic in greenhouse indoor environment control. This is due to nonlinearities and inputs/outputs coupling present in physical model of greenhouse. The objective of this study is about the problem of identification and control of a Multi-Input Multi-Output (MIMO) greenhouse process. Proportional Integral (PI) and sliding mode controllers (SMC) are used in conjunction so that estimated outputs tracks desired trajectories with good performance as near as possible in spite of coupled dynamics and outside disturbances climate. To reflect the practical aspect, the constraints of the process and control elements were taken into account .The results demonstrate the feasibility of PISMC control method. Simulation has been done in the environment of Matlab and Simulink; it shows that combined controllers are capable to manage successfully the microclimate of greenhouse.