NOURI Benaidja

BARKATI Hamza

Modélisation floue d’un système Non linéaire

This paper discusses the application of the method of reference model to determine parameters of nonlinear systems using three society-inspired heuristics. Softcomputing techniques used in this paper are harmonic search HS, teaching-learning algorithm TLA and cultural Algorithm CA. Furthermore, hybridization of CA with artificial immune system AIS is investigated in order to obtain high performance meta-heuristic CA-AIS. The algorithm is thus called CA-AIS to estimate the parameters of Tanagi-Sugeno fuzzy model of nonlinear systems. The CA-AIS comprises two main different phases. First, the belief space is built considering situational knowledge and domain knowledge only, the event of dramatic drop of objective function is recorded and exploration-exploitation balance is preserved. Second, the historical knowledge is added to the belief space to exploit the information collected at the previous phase and enhance the intensification capability. Finally, the algorithms are examined by parameter identification of the fuzzy model of Narendra-Parthasarathy nonlinear system. Simulation results show that the proposed algorithm is more efficient than typical genetic algorithm and genetic algorithm hybridized with simulated annealing GA-SA.

Citation

Nouri BENAIDJA , ,(2025-11-20), Optimisation of the fuzzy model of non linear systems by society-inspired heuristics,International Conference on Recent Advances in Mathematics and Informatics,Sousse, Tunisie

A new semi-empirical formula has been developed to calculate the (n, 3 He) reaction cross-section at a neutron energy of 14.6 MeV, with the aim of improving prediction accuracy by incorporating the Q-value and asymmetry parameters. The formula is based on the exciton model, which describes the pre-equilibrium reaction mechanism. Experimental data from the literature were used to validate the formula and optimize its free parameters. The results show that the proposed formula provides a better fit to experimental data compared to previous models, especially at incident energies above 15 MeV. The model has also been extended to cover a wider energy range and validated using additional experimental data. Keywords — (n,t) function excitaion;; fusion reaction; 3

Citation

Samra Nehaoua , Asma Ghezal , Nouri BENAIDJA , ,(2025-06-25), Improved semi-empirical formula for (n, 3 He) reaction cross section at 14.6 MeV,17th International Conference on Nuclear Structure Properties (NSP2025),,Turkey

The new formula enhances the accuracy of (n, t) reaction cross-section predictions at 14.6 MeV, offering a reliable tool for theoretical and applied nuclear physics. Future work will focus on extending the model to a broader energy range and validating it with additional experimental datasets.

Citation

Samra Nehaoua , Asma Ghezal , Nouri BENAIDJA , ,(2025-04-13), Novel semi empirical systematic for (n,t) cross section at 14,6 MeV,4th International Conference on Radiations And Applications ICRAA4,Algeria

Background/Purpose: A semi empirical formula for (n,d) nuclear reaction cross section is proposed. The systematic was derived using analytical expressions for the calculation of the deuterium emission spectrum in (n,d) nuclear reactions. The pre-equilibrium exciton model and the evaporation model are used to describe the deuterium emission. The experimental data are used to adjust free parameters in systematic formula at 14-15 MeV. The systematics gives closer results to experimental data compared with formulas earlier studies. Materials & Methods: The pre-equilibrium exciton model and the evaporation model are used to describe the deuterium emission; previous studies have ignored pre equilibrium mechanism. Results: The proposed semi empirical systematic describe better (n,d) function excitation and gives least values of statistical parameters which proof the quality of our formula. [1,2]. Conclusion: The systematics gives closer results to experimental data compared with formulas earlier studies. The proposed semi empirical systematic describe better (n,d) function excitation and gives least values of statistical parameters which proof the quality of our formula.

Citation

Samra Nehaoua , Asma Ghezal , Nouri BENAIDJA , ,(2025-04-13), Semi empirical formula for (n,d) reaction cross section at 14-15 MeV,4th International Conference on Radiations And Applications ICRAA4,Algeria

In this paper, a novel evolutionary optimisation algorithm is proposed by hybridizing the cultural algorithm (CA) and artificial immune system (AIS). However, the hypermutation scheme is inserted at the end of the influence step. The algorithm is thus called CAAIS to estimate the parameters of Tanagi-Sugeno fuzzy model of nonlinear systems. The CAAIS comprises two main different phases. First, the belief space is built considering situational knowledge and domain knowledge only, the event of dramatic drop of objective function is recorded and exploration-exploitation balance is preserved. Second, the historical knowledge is added to the belief space to exploit the information collected at the previous phase and enhance the intensification capability. Finally, the CAAIS algorithm is examined by parameter identification of the the fuzzy model of Narendra-Parthasarathy nonlinear system. Simulation results show that the proposed algorithm is more efficient than typical genetic algorithm and genetic algorithm hybridized with simulated annealing GASA.

Citation

Nouri BENAIDJA , ,(2024-11-18), Optimisation of the fuzzy model of non linear systems by hybrid cultural algorithm,The International Congress on Geomatics Applications and Geographic Artificial Intelligence,Hammamet, Tunisie

Ce polycopié de rappel de cours et de travaux pratiques regroupe un certain nombre de notions de base sur l’algorithmique ainsi que l’application de ces notions en s’appuyant sur le langage Pascal. Ces travaux pratiques d’algorithmique, destinés particulièrement aux étudiants de première année Sciences de technologie (ST), font assurés pendant pratiquement une dizaine d’années par notre équipe de formation, affiliée à la Faculté de Sciences de technologie (ST) (Université de M’sila). Sans être exhaustif ce polycopié est en conformité avec le programme proposé par la tutelle. Le polycopié est organisé en deux grandes parties : informatique 1 et informatique 2. La première partie englobe trois chapitres. L’introduction porte sur l’environnement matériel et logiciel d’un ordinateur. En particulier les différentes fonctionnalités des systèmes d’exploitation sont illustrées en s’appuyant sur Windows 7® IDE disponible au niveau de la faculté. Puis une illustration de l’environnement de programmation Free Pascal IDE disponible au niveau de la faculté. Ce chapitre est conclu par un TP préliminaire permettant à l’étudiant de se familiariser, tout d’abord, avec les différentes instructions de base de programmation Pascal. Le deuxième chapitre regroupe l’essentiel du cours d’algorithmique, à savoir, le langage algorithmique qui couvre toutes les instructions de base constituant un algorithme et les structures conditionnelles et itératives. Le troisième chapitre est consacré à des problèmes plus complexes, à savoir les vecteurs et les matrices. La deuxième partie est divisée en deux grands volets. Le premier chapitre illustre l'aspect modulaire du langage algorithmique, à savoir les fonctions et procédures. Ce qui fournit une possibilité de structurer encore davantage l’algorithme en créant de nouveaux ordres, définis indépendamment de l’algorithme principal, utilisables dans le corps de l’algorithme. La notion d'enregistrement offre la possibilité d'étendre les types de base vers des ensembles de données hétérogènes. Ces structures peuvent être stockées dans un fichier. Les enregistrements et fichiers font l'objet d'illustration dans le deuxième chapitre. Chaque chapitre est conclu par un ensemble d'exercices traités, en séance de TP, par les étudiants de première année ST. Des exercices supplémentaires sont ajoutés à certaines sections jugées fondamentales. Ces séries d’exercices couvrent toutes les sections présentées en cours. Pour ce faire, j’ai fait appel au langage Pascal. Ce dernier est un langage facile à manipuler par l’étudiant débutant et possède une structure très proche de celle du langage naturel.

Citation

NouriBENAIDJA , ,(2020-10-11); TP Informatique I et II,University of M'sila,

Softcomputing techniques are receiving attention as optimisation techniques for many industrial applications. Although these techniques eliminate the need for derivatives computation, they require much work to adjust their parameters at the stage of research and development. Issues such as speed, stability, and parameters convergence remain much to be investigated. This paper discusses the application of the method of reference model to determine parameters of asynchronous machines using two optimisation techniques. Softcomputing techniques used in this paper are evolutionary strategy and the chemotaxis algorithm. Identification results using the two techniques are presented and compared with respect to the conventional simplex technique of Nelder and Mead. Discussion about the chemotaxis algorithm as the most promising optimisation technique is presented, giving its advantages and disadvantages.

Citation

Nouri BENAIDJA , , (2009-09-09), Softcomputing identification techniques of asynchronous machine parameters: Evolutionary strategy and chemotaxis algorithm, Turkish Journal of Electrical Engineering & Computer Sciences, Vol:17, Issue:1, pages:69-85, The Scientific and Technological Research Council of Turkey

Four evolutionary techniques (scatter search, evolutionary programming, ant colony, and particle swarm algorithms) were used for off-line identification of three phase asynchronous machine parameters. Optimization techniques were then tested on two distinct machines. In order to evaluate how much good the achieved machines parameters obtained, experimental and simulation input-output behaviors are presented for each method. The performances in term of objective function and convergence time prove the effectiveness of this class of optimization methods.

Citation

Nouri BENAIDJA , abdelrazak rahmouni , Nabil Khenfer, , (2006-01-01), Identification of asynchronous machine parameters by evolutionary techniques, Electric Power Components and Systems, Vol:34, Issue:12, pages:1359-1376, Taylor and Francis Ltd

The present study describes a comparative study of genetic algorithm and neighbourhood techniques used for off-line identification of three phase asynchronous machine parameters. All the methods are then tested on two distinct machines and the influence of the start time of identification, with reference to the time of startup of machine, is studied. Results show the superiority of genetic algorithm for low values of the start time of identification.

Citation

Nouri BENAIDJA , abdelrazak rahmouni , Nabil Khenfer, , (2006-01-01), Heuristics for Asynchronous Machine Identification:Genetic Algorithm and Neighbourhood Techniques, Asian Journal of Information Technology, Vol:5, Issue:8, pages:854 - 859, Medwell Journals

Résumé Cet article présente l’identification off-line des paramètres de la machine asynchrone à partir de la connaissance des quantités externes mesurables : courants et tensions statoriques et vitesse de rotation. Le problème d’identification est considéré comme un problème d’optimisation et résolu par la méthode basée sur les algorithmes génétiques. La commande indirecte par orientation du flux nécessite la connaissance des paramètres de la machine et par conséquent des résultats de simulation de la commande utilisant les paramètres trouvés lors de l’identification sont développées.

Citation

Nouri BENAIDJA , ,(2001-06-09), Identification de la Machine Asynchrone par les Algorithmes Génétiques,Conférence Internationale sur la Productique,Alger, Algerie