Institutional repository of university M'Hamed Bougara Boumerdes
Not a member yet
13163 research outputs found
Sort by
Seasonal Forecasting of Global Horizontal Irradiance for Grid-Connected PV Plants: A Combined CNN-BiGRU Approach
No full textThe quest for environmental sustainability in power systems necessitates the incorporation of renewable energy sources into the grid infrastructure. Among these renewable sources, solar energy has risen to prominence due to its widespread availability. However, the variable nature of solar irradiance poses challenges in operational and control aspects of its integration. A potential solution lies in predictions of global horizontal irradiance (GHI). This study introduces an ensemble deep learning-based forecasting approach, leveraging a Convolutional Neural Network and Bidirectional Gated Recurrent Unit (CNN-BiGRU). The efficacy of this approach is evaluated against three ensemble models: The Convolutional Neural Network Bidirectional Long Short Term Memory (CNN-BiLSTM), Convolutional Neural Network Gated Recurrent Unit (CNN-GRU), the Convolutional Neural Network Long Short Term Memory (CNN-LSTM). The comparative analysis is centered on seasonal GHI forecasting in Alice Springs, Australia, with a 1-hour time horizon. Four metrics are employed to gauge the accuracy of the models: coefficient of determination (R2), mean absolute error (MAE), normalised root mean square error (nRMSE), and root mean square error (RMSE). The findings reveal that the proposed ensemble bidirectional model outperforms its counterparts in all seasons. Specifically, in terms of seasonal forecasting, the CNN-BiGRU model achieves a maximum nRMSE of 0.0955, indicating its superior performance
Detection de maladies des plantes en utilisant les reseaux de neurones approfondies
No full text65 p. : ill. ; 30 cmL'un des principaux défis auxquels sont confrontés les professionnels agricoles est la découverte des maladies des plantes, un aspect crucial pour assurer des cultures saines et abondantes et la conservation des plantes. Notre projet vise à résoudre ce problème en utilisant l'intelligence artificielle et des techniques d'apprentissage profond pour développer un système d'analyse des feuilles précis en utilisant des réseaux de neurones artificiels. Ces modèles ont une capacité exceptionnelle à identifier des grains satisfaisants, permettant des mesures préventives efficaces pour prévenir la perte de cultures et de plantes. Nous visons à fournir aux agriculteurs et aux paysans une solution fiable et proactive pour faciliter la détection des maladies des plantes. Le projet représente des progrès importants en agriculture, offrant de nouvelles possibilités d'améliorer la surveillance et le traitement des maladie
Lp-theory for elliptic systems without divergence constraint and with Navier-type boundary condition
No full textIn this paper, we study a linear elliptic system involving Navier-type boundary conditions or tangential components in a three-dimensional bounded domain, possibly multiply connected and with boundary possibly nonconnected. We establish an appropriate Inf-Sup condition and then we prove the existence and uniqueness of the generalized solutions in (Formula presented.) -theory. We also investigate the case of strong solutions
Conception et réalisation d’un système intelligent de gestion d’éclairage public solaire intégrant des technologies avancées
No full text73 p. : ill. ; 30 cmCe Projet de fin d'études vise à développer un système d'éclairage public intelligent en utilisant les microcontrôleurs Raspberry Pi et ESP8266, avec un tableau de bord basé sur Node-RED pour une surveillance et un contrôle complet. Le système intègre un système photovoltaïque (PV) autonome pour alimenter l'éclairage, assurant ainsi durabilité et efficacité énergétique. Des capacités de surveillance avancées sont mises en oeuvre pour observer et analyser divers paramètres du système PV, garantissant des performances optimales. Le projet se concentre non seulement sur la réduction de la consommation d'énergie, mais aussi sur le maintien de la qualité de l'éclairage et de la sécurité des usagers. La solution développée est conçue pour être autonome, durable et capable de suivre les performances en temps réel, offrant ainsi une alternative viable et efficace aux systèmes d'éclairage public traditionnels
Commande d'un pendule inversé monté sur un chariot - Application de l'optimisation par essaim particulaire
No full text85 p. : ill. ; 30 cmCe mémoire explore la commande optimale du pendule inversé en utilisant des techniques avancées de régulation et d'optimisation. D'abord, il introduit le pendule inversé comme un modèle clé pour les systèmes non linéaires, soulignant son importance en robotique et en automatisation. Ensuite, il se concentre sur la conception et la mise en œuvre d'un régulateur linéaire quadratique (LQR) pour stabiliser le pendule, démontrant l'efficacité de cette approche pour minimiser une fonction de coût quadratique. Enfin, il présente l'optimisation par essaim particulaire (OEP) pour affiner les paramètres du LQR, illustrant comment cette technique bio- inspirée améliore significativement la performance et la stabilité du système. L'étude conclut que l'OEP est une méthode puissante pour optimiser les régulateurs dans des systèmes dynamiques complexes, avec des applications potentielles dans divers domaines technologique
Numerical analysis of gray gas radiation effects on heat and mass transfer in an annular cavity
No full textThis study deals with a numerical investigation of coupled double diffusive natural-convection with thermal radiation in an annular cavity containing a gray gas mixture. The black vertical cylindrical walls are maintained at different temperatures and concentrations to create cooperating flows. The finite volume method (using the SIMPLER algorithm) is used to solve the governing equations and the discrete ordinate method (with S8 quadrature) to treat the radiative aspect of the problem. A parametric study illustrating the influence of the optical thickness and the ratio of buoyancy forces, on the flow field and heat and mass transfer for Reyleigh number equal to 5 ∙ 106 and aspect ratio equal to 1, is performed. The numerical results show that gas radiation modifies the flow structure and the distribution of temperature and concentration in the cavity. The effect of permutation of boundary conditions, between the vertical walls, on heat and mass transfer is also considered. The thermal radiation reduces the total heat transfer in the annular space regardless of the configuration of the boundary condition
Comparative interactions analysis of «APC» and «PID» controllers in Distillation column at Skikda «GL1K» complex
No full text114 p. : ill. ; 30 cmThe Objective of this thesis is to evaluate the performance of Honeywell's Advanced Process Control (APC) system in multivariable processes and to compare it with traditional PID controllers. This study was conducted at the GL1K complex in Skikda, which specializes in the liquefaction of natural gas. We examined the effectiveness of APC in managing the complex interactions within the process, focusing on key performance indicators such as stability, efficiency, and product quality. The results demonstrated that APC significantly improves process performance and reduces variability compared to conventional PID controllers. This research highlights the advantages of advanced control strategies in optimizing industrial operations and ensuring superior control over multivariable systems
Comparative analysis on heat transfer, between a steady and oscillating jet in a cavity
No full textThis paper numerically investigates the cooling of a heated rectangular cavity by a cold slot jet. The study aims to examine the effect of the jet location inside the cavity (Lf and Lh) and Reynolds number on heat transfer, using URANS turbulence modelling. Different flow behaviours, including oscillatory and steady flows, are generated depending on the jet location inside the cavity. The study identifies and discusses the optimal jet locations for achieving optimal cavity cooling. The results indicate that the lateral placement of the jet has a negligible effect on heat transfer across all cavity walls. Additionally, oscillatory flow consistently expands the heat exchange zone along all three walls, resulting in a wider effective exchange area compared to steady flow conditions. The study proposes optimised jet positions within the cavity for specific wall cooling requirements. By considering the optimal combination of jet height and impinging distance, the cooling performance can be optimised