Browsing by Author "Lamamra, Kheireddine"

Now showing 1 - 7 of 7
  • No Thumbnail Available
    Item
    Box and jenkins nonlinear system modelling using RBF neural networks designed by NSGAII
    (Springer, 2015) Lamamra, Kheireddine; Belarbi, Khaled; Boukhtini, Souaad
    In this work, we use radial basis function neural network for modeling nonlinear systems. Generally, the main problem in artificial neural network is often to find a better structure. The choice of the architecture of artificial neural network for a given problem has long been a problem. Developments show that it is often possible to find architecture of artificial neural network that greatly improves the results obtained with conventional methods. We propose in this work a method based on No Sorting Genetic Algorithm II (NSGA II) to determine the best parameters of a radial basis function neural network. The NSGAII should provide the best connection weights between the hidden layer and output layer, find the parameters of the radial function of neurons in the hidden layer and the optimal number of neurons in the hidden layers and thus ensure learning necessary. Two functions are optimized by NSGAII: the number of neurons in the hidden layer of the radial basis function neural network, and the error which is the difference between desired input and the output of the radial basis function neural network. This method is applied to modeling Box and Jenkins system. The obtained results are very satisfactory.
  • No Thumbnail Available
    Item
    Business intelligence platform for projects related to the internet of things
    (University of Oum El Bouaghi, 2021) Bezzaz, Soumia; Benmerzoug, Djamel; Lamamra, Kheireddine
    The exponential growth in the number of connected objects is evidence of a particu-larly impactful trend in the digital world, driven by the uses of the general public. This innovative technological trend will also have a strong impact on companies, bringing changes to the management of the large amount of data that comes from connected objects. To ensure better management of this data, we have proposed a business intelligence platform called tGRC (Governance, Risk Management and Compliance related to IoT) as a suitable solution for this area. We believe that this platform is much more important for connected objects because of the heterogeneity of objects, the more or less controlled access to their data and the way their data can be used. Moreover, it is essential that a significant part of the tasks is effectively automated in order to guarantee homogeneous operations when making decisions in compliance with the rules and controlling the risks.
  • No Thumbnail Available
    Item
    Comparison of tow methods P&O and fuzzy logic controller for MPPT in the photovoltaic system
    (University of Oum El Bouaghi, 2021) Khelaifa, Fethi; Lamamra, Kheireddine; Toumi, Djaafar
    An important consideration in the use of the PV system is to operate the system near the maximum power point to increase the output efficiency of the PV panel. Maximum power point tracking (MPPT) techniques are used in PV applications to extract the maximum power that the PV panel can produce, which depends on different atmospheric conditions, namely solar insolation and temperature. In this paper, a new maximum power point tracker using fuzzy set theory is proposed to improve the energy conversion efficiency. A fuzzy algorithm based on linguistic rules describing the operator control strategy is applied to the boost converter control so that the maximum power point tracker linguistic variables implemented in the fuzzy logic controller were appropriately selected to modulate the DC-DC converter. The fuzzy logic controller offers advanced features such as fast response, good oscillation performance the results obtained are compared with the P&O technology.
  • No Thumbnail Available
    Item
    Contribution on modeling and control of a wind power system based on fuzzy
    (University of Oum El Bouaghi, 2021) Baladji, Djalal; Lamamra, Kheireddine
    In this work, we study the modeling and control of a wind energy conversion chain based on a Dual Power Asynchronous Machine. First, the mathematical model of the whole system under study is developed in d-q Park coordinates. Then, the proposed control strategy is designed based on a fuzzy logic control technique that gives good results. Finally, the results of the different simulations of the whole conversion chain, performed under MATLAB/Simulink environment, are presented.
  • No Thumbnail Available
    Item
    Contribution to motion planning and control of robotic systems
    (University of Oum El Bouaghi, 2024) Ghediri, Akram; Lamamra, Kheireddine
    This study presents a novel design for adaptive PID gain tuning in the context of robot manipulators’ PID Computed-Torque control. The approach incorporates Deep Deterministic Policy Gradient (DDPG) reinforcement learning to account for unmodeled dynamics and external disturbances. The main objective is to dynamically compute the outer-loop PID controller gains, ensuring minimal trajectory tracking errors while effectively rejecting disturbances and maintaining stable closed-loop dynamics. To implement the control scheme, a comprehensive understanding of the robot’s dynamics is necessary. Therefore, the study develops both the kinematic and dynamic equations for an n-link serial manipulator. The UR5e robot manipulator model is utilized, and the dynamic and kinematic parameters provided by the manufacturer and related works are considered to ensure accuracy. Simulation results demonstrate the effectiveness and robustness of the proposed approach. It successfully handles bounded internal and external disturbances, showcasing its ability to adapt and mitigate their effects. Additionally, the adaptive gain tuning offers improved trajectory tracking performance compared to conventional PID controllers. Overall, this work contributes to the advancement of control strategies for robot manipulators by integrating adaptive PID gain tuning with DDPG reinforcement learning. By addressing unmodeled dynamics and external disturbances, the proposed approach enhances the stability and tracking capabilities of robot manipulator systems. Cette étude présente une conception novatrice pour l'ajustement adaptatif des gains PID dans le contexte du contrôle PID de la commande de couple des manipulateurs robotiques. L'approche intègre l'apprentissage par renforcement Deep Deterministic Policy Gradient (DDPG) pour prendre en compte les dynamiques non modélisées et les perturbations externes. L'objectif principal est de calculer dynamiquement les gains du régulateur PID en boucle externe, garantissant des erreurs minimales de suivi de trajectoire tout en rejetant efficacement les perturbations et en maintenant une dynamique en boucle fermée stable. Pour mettre en œuvre le schéma de contrôle, une compréhension approfondie de la dynamique du robot est nécessaire. Par conséquent, l'étude développe à la fois les équations cinématiques et dynamiques pour un manipulateur série à n maillons. Le modèle de manipulateur robotique UR5e est utilisé, et les paramètres dynamiques et cinématiques fournis par le fabricant et les travaux connexes sont pris en compte pour assurer la précision. Les résultats de la simulation démontrent l'efficacité et la robustesse de l'approche proposée. Elle gère avec succès les perturbations internes et externes bornées, illustrant sa capacité à s'adapter et à atténuer leurs effets. De plus, l'ajustement adaptatif des gains offre une meilleure performance de suivi de trajectoire par rapport aux régulateurs PID classiques. Dans l'ensemble, ce travail contribue à l'avancement des stratégies de contrôle pour les manipulateurs robotiques en intégrant l'ajustement adaptatif des gains PID avec l'apprentissage par renforcement DDPG. En traitant les dynamiques non modélisées et les perturbations externes, l'approche proposée améliore la stabilité et les capacités de suivi des systèmes de manipulateurs robotiques.
  • No Thumbnail Available
    Item
    Linear quadratic-based computed-torque control of an actuated pendulum with friction
    (University of Oum El Bouaghi, 2021) Ghediri, Akram; Lamamra, Kheireddine; Ait Kaki, Abdelaziz
    In this paper, the trajectory tracking control of an actuated pendulum with friction using LQ-based computed-torque controller is studied. The control scheme consists of feedback linearization loop to cancel nonlinear dynamics and LQ as outer-loop controller. The dynamic model is developed using Euler-Lagrange formulation and the stability behavior is analyzed for both open and closed-loop system. Simulation result show that LQ-based computed-torque controller is effective in terms of computing optimal feedback gains that minimize the position and velocity tracking errors.
  • No Thumbnail Available
    Item
    Numerical study and performance analysis of carbone nanotube field effect transistors
    (Opatija, Croatia, 2013) Rechem, Djamil; Benkara, Salima; Lamamra, Kheireddine
    As transistors are scaled down to nanometers, the theory and structure of nanometers devices such as carbon nanotubes field effect transistors (CNTFET) are being extensively studied. Self consistent solution of the Poisson and Schrödinger equations is performed using the nonequilibrium Green’s function (NEGF) formalism to investigate the impact of nanotube diameter, gate oxide thickness and high-k gate dielectric permittivity parameters on the coaxially gate, n-type CNTFET. Our results show that the nanotube diameter and gate oxide thickness influences the ION/IOFF current ratio, the drain induced barrier lowering (DIBL), the subthreshold slop as well as transconductance and drain conductance. Furthermore, in this work we focus on the impact of high-k gate dielectric permittivity on the performance of CNTFETs. Using high-k dielectric is caused by the enhancement in device characteristics. A good agreement with numerical simulation results is obtained.