Mémoires de Master
URI permanent de cette collection
Parcourir par
Dépôts récents
1 - 5 sur 17
- ItemRealisation d`une machine CNC laser « FOCUS CNC »(2024-06-30)Ce projet de fin d'études porte sur la conception et la réalisation d'une machine CNC laser «FOCUS CNC » à faible coût, dotée d’un system de surveillance à distance et d’un capteur de flamme. Nous avons conçu, imprimé et assemblé les pièces nécessaires, puis paramétré la machine en effectuant divers tests sur différents matériaux. En parallèle, nous avons mené une étude économique approfondie du marché algérien pour évaluer la faisabilité et les opportunités de lancement de notre machine.
- ItemProtocol for UV visible spectrophotometric calibration check(2024-06-30)This project outlines the protocol for calibration check of a UV-Visible spectrophotometer Shimadzu UV-3600 in Laboratory of Thin Films and Materials for Electronics. The calibration process involved preparing a series of coloured solutions with known absorbance spectra to serve as calibration standards. The spectrophotometer's performance was evaluated by measuring the absorbance of these solutions across the UV-Visible spectrum. Theobtained results demonstrated the instrument's accuracy and linearity, confirming its reliability for precise spectrophotometric analysis. This protocol ensures consistent performance of the spectrophotometer, essential for high-quality quantitative measurements in various applications.
- ItemAn IoT Low-Cost Fertilizer for Crop Recommendation using Machine Learning(2024-06-27)Agriculture plays a crucial role in sustaining the world's growing population, yet farmers face challenges in optimizing crop production and resource management. This thesis introduces an innovative IoT device empowered by machine learning (ML) to monitor soil nutrients and offer precise crop recommendations. The device integrates sensors such as, an NPK sensor and MAX485 TTL to gather real-time data on soil composition, humidity, temperature, rainfall, ph, and nutrient levels. This data is then transmitted to a server via protocol. ML algorithms analyse the collected data to generate tailored recommendations, including optimal crop choices, suitable fertilizer types, and application rates based on crop needs and soil conditions. Field experiments validate the system's efficacy compared to traditional methods, demonstrating its ability to boost crop productivity, optimize resource allocation, and promote sustainable agricultural practices for enhanced food security.
- ItemAn IoT-based deep learning frame work for early assessment of pneumonia.(2024)The COVID-19 pandemic has highlighted the importance of remote patient monitoring for the well-being of both patients and healthcare providers. Pneumonia, a leading cause of morbidity and mortality in children, requires prompt diagnosis and treatment due to its impact on the small air sacs in the lungs. Chest X-rays are a common diagnostic tool for detecting pneumonia. This study emphasizes the development of a support tool to assist healthcare professionals in enhancing the accuracy and efficiency of pneumonia diagnosis through advanced deep learning approaches. The best model accuracy of VGGNet of 95.89% confirms that the model is robust however MobileNet can perform well in real-world situations.
- ItemMulti objective optimization for DC-DC power converter with EMC consideration(2024-07-01)This research project proposes a comprehensive multi-objective optimization approach for DCDC power converters, emphasizing the resolution of Electromagnetic Compatibility (EMC) challenges. The methodology includes the application of the Design of Experiments (DOE) method alongside traditional objectives such as efficiency, analysis of EMI emissions under optimized conditions, and PCB implementation. Case studies will evaluate the proposed approach specifically in electric vehicle applications. The anticipated outcomes aim to guide engineers in developing high-performing DC-DC power converters compliant with EMC standards, contributing to advanced design practices aligned with contemporary power electronics priorities