المستودع الرقمي · Dépôt institutionnel

جامعة عمّار ثليجي – الأغواط

Université Amar Telidji de Laghouat

Archive ouverte de la recherche, des thèses et des cours pédagogiques
الأرشيف المفتوح للبحوث والمذكرات والدروس

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 12

Recent Submissions

  • Item type:Item,
    Simulation d'un Système de Surveillance Cardiaque avec Détection d'Anomalies par EdgeAI
    (UNIVERSITÉ AMAR THELIDJI DE LAGHOUAT FACULTE DU TECHNOLOGIE DEPARTEMENT ELECTRONIQUE, 2026) Ketel Zehour; Leila Amal Vilbois
    This memoir aims to simulate an intelligent cardiac monitoring system based on Edge AI technologies for the early detection of cardiac anomalies in real time. Three approaches were used for the classification and detection of abnormal cases: Random Forest, Deep ECGNet, and CCFC-CLTC. Their performance was compared in terms of accuracy, efficiency, and execution speed on edge devices. The obtained results demonstrate the potential of integrating Edge AI into cardiac monitoring systems to improve early anomaly detection and reduce response time.
  • Item type:Item,
    Étude de l’Activité Antioxydante des Extraits Phénoliques de quelques Plantes Médicinales Locales (Peganum harmala, Haloxylon scoparium et Tamarix aphylla)
    (Université Amar Thelidji- Laghouat FACULTE : TECHNOLOGIE DEPARTEMENT : GÉNIE DES PROCÉDÉS, 2026) DALIA Ikram; DENNAKA Oum Kalthoum; GUENANE Hadjira
    The aim of this study was to evaluate the antioxidant activity of hydroethanolic extracts of Haloxylon scoparium, Peganum harmala, and Tamarix aphylla through the determination of total polyphenol and flavonoid contents. The results showed that Haloxylon scoparium exhibited the highest levels of polyphenols (10.127 ± 0.066 mg GAE/g DM) and flavonoids (9.609 ± 0.310 mg RE/g DM), followed by Tamarix aphylla, whereas Peganum harmala showed the lowest values. Antioxidant activity was evaluated using DPPH, phosphomolybdenum, and FRAP assays. Haloxylon scoparium showed the strongest activity in both the DPPH assay (IC₅₀ = 0.067 ± 0.0018 mg/mL) and the phosphomolybdenum assay (0.063 ± 0.001 M), while Peganum harmala exhibited the highest reducing power in the FRAP assay (0.111 ± 0.005 M).
  • Item type:Item,
    Modeling & Intelligent Control of a Multi-DoF Robotic Arm
    (University of Amar Telidji - Laghouat Faculty of Technology Department of Electrical Engineering, 2026) CHERAIF Souad; MOKRANI Lakhdar
    This dissertation addresses the dynamic modeling and high-speed trajectory tracking control of a 6-DOF PUMA 560 robotic manipulator under severe dynamic stress and parametric uncertainty. Classical PID controllers, being structurally unaware of nonlinear multi-body dynamics, exhibit serious tracking distortion (16.8 mm error) and actuator saturation under an unmodeled 5 kg payload. To overcome these limitations, two opposing strategies are investigated: Model-free Mamdani Fuzzy Logic Control (FLC) offers robustness but lacks deterministic precision, while model-based Computed Torque Control (CTC) delivers perfect nominal precision (0.05 mm error) but fails catastrophically under parametric variations. The main contribution is a hybrid CTC-FLC architecture that synergistically integrates exact nonlinear cancellation from the dynamic model with intelligent uncertainty rejection from the fuzzy compensator. Under sustained high-frequency helical maneuvers and an unmodeled 5 kg payload, the hybrid controller entirely eliminates steady-state spatial shift and actuator chattering, achieving a remarkable mean Cartesian error of only 0.1 mm. This work provides empirical evidence that absolute spatial accuracy in unpredictable industrial environments requires the synergistic fusion of deterministic physical modeling and artificial intelligence.
  • Item type:Item,
    Fault diagnosis of induction machine using AI based on ANFIS method
    (UNIVERSITÉ AMAR TELIDJI LAGHOUAT Faculty of technology DEPARTMENT: Electrotechnics, 2026) BOUZIDI Ilyes; CHAOUCHE Brahim; BESSEDIK Sid Ahmed
    The induction machine is widely used in industry due to its simplicity, robustness, and low cost. It also requires minimal maintenance and provides high performance. However, during operation, it may be subjected to internal or external stresses. These stresses can lead to faults and operational failures. Such failures often cause costly production shutdowns. They may also result in significant material and financial losses. Therefore, it is essential to develop efficient and fast diagnostic methods. This thesis investigates the use of the ANFIS neuro-fuzzy approach for stator fault diagnosis. This method enables early fault detection and accurate fault localization. As a result, it improves machine reliability and reduces industrial losses.
  • Item type:Item,
    Etude de l’inhibition de la corrosion d’un acier au carbone X70 par trois composés chimiques dans un milieu chloruré
    (Université Amar Thelidji- Laghouat FACULTE : TECHNOLOGIE DEPARTEMENT : GÉNIE DES PROCÉDÉS, 2026) GUERMIT Mebarek Abde raoufe; SOFRANI Maamar Abdelbaset; BOUTASSOUNA Nassima
    This study investigates the corrosion inhibition performance of three organic compounds on API 5L X70 carbon steel in 0.5% NaCl solution. The inhibitors tested were sulfanilamide, pyrrolidinone, and hydroxylamine. Their inhibition efficiency was evaluated using weight loss measurements and potentiodynamic polarization techniques. The results revealed that hydroxylamine exhibited the highest inhibition efficiency, reaching 98.92% at 150 ppm, followed by pyrrolidinone (97.28% at 200 ppm) and sulfanilamide (95.92% at 150 ppm). All three inhibitors acted as mixed-type inhibitors with predominant cathodic effect. Their performance was found to be strongly dependent on molecular structure and the presence of heteroatoms (N and O), which facilitate adsorption on the steel surface. These findings suggest that these organic compounds, particularly hydroxylamine, are promising eco-friendly inhibitors for protecting pipelines in chloride-containing environments.