Teses de Mestrado ESTiG
Permanent URI for this community
Browse
Browsing Teses de Mestrado ESTiG by Sustainable Development Goals (SDG) "15:Proteger a Vida Terrestre"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Disease detection and mapping in olive groves using UAVs and deep learning for precision agriculturePublication . Morais, Maurício Herche Fófano de; Lima, José; Santos, Murillo Ferreira dos; Mendes, João CarlosThis dissertation presents the implementation and validation of a cost-effective, Unmanned Aerial Vehicle (UAV) based system for automated detection and spatial mapping of olive knot disease in olive groves. Addressing the need for accessible and efficient plant disease monitoring in Precision Agriculture (PA), the proposed methodology leverages existing UAV imagery capabilities with lightweight Deep Learning (DL) models, specifically the You Only Look Once (YOLO) object detection architecture, to enable scalable and accurate detection. The academic contributions presented in this work have resulted in two peer-reviewed publications related to the dissertation topic, as detailed at the end of this document. An annotated dataset of UAV-acquired images was compiled, and several state-of-the-art YOLO object detection models were trained and evaluated under identical conditions. The best-performing model achieved a strong F1-score, demonstrating good results in detecting olive knot disease and accurately mapping its spatial distribution within the plantation. The workflow integrates spatial cross-referencing of detections with UAV flight path data and proximity analysis, enabling the assignment of disease detections to individual trees. An interactive map interface, developed using the Folium Python library, provides visualization of the disease distribution and supports practical grove management. The experimental results indicate that cost-effective UAVs and lightweight DL models can be effectively combined for plant disease detection and spatial analysis, offering a robust and scalable approach for real-world agricultural applications. Limitations regarding early symptom detection and image quality are discussed, and directions for future work are proposed.
- Separation of methane and hydrogen in a 3D-printed porous carbon monolithPublication . Junqueira, Matheus Raphael Diniz; Silva, José A.C.; Dumont, Marcello RosaThe efficient separation of methane (CH4) and hydrogen (H2) is a key challenge in industrial processes such as steam methane reforming (SMR), which is the primary technology used for hydrogen production; however, the generated gas contains impurities such as carbon monoxide (CO), carbon dioxide (CO2), and unreacted methane, which must be removed to ensure high-purity H2. The separation of these gases is commonly performed through pressure swing adsorption (PSA) using adsorbent materials such as zeolites (e.g., zeolite 13X) and activated carbons. However, these adsorbents often present challenges, including high pressure drop and limited control over pore structure. In this context, this study investigates a 3D-printed porous carbon monolith with tetragonal cubic centred unit cells, designed to maximize CH4 selectivity over H2 while reducing pressure drop due to its highly controllable structural design. The material was characterized using fixed-bed adsorption experiments analyzed via flow gas chromatography, including single-component (H2 and CH4) and binary adsorption (CH4/H2 mixtures) at 303 K, 313 K, and 343 K, with pressures up to 30 bar. Adsorption equilibrium modelling was conducted using the Dual-Site Langmuir (DSL) isotherm, accurately describing an experimental 76/24 (% vol.) CH4/H2 mixture, reinforcing the material’s selectivity, closely matching the values predicted by the isotherm from single component experiments. These findings highlight the potential of 3D-printed porous carbon monoliths for selective CH4 separation in PSA processes applied to SMR, offering a promising alternative with lower pressure drop and greater structural control compared to conventional adsorbents. the gas-monolith interactions. Results showed that H2 adsorption was negligible under all tested conditions. For CH4, the maximum adsorption experimental capacity was 3.25 mol.kg−1 at 303 K and 30 bar. Equilibrium isotherms confirmed material heterogeneity, with two distinct adsorption sites, each with its own adsorption capacity. The isosteric heat of adsorption ranged from 17.5 to 17.1 kJ.mol−1, indicating a moderate physisorption mechanism. In binary adsorption experiments at 303 K and 5 bar, CH4 adsorption reached 2.10 mol.kg−1 for an experimental 76/24 (% vol.) CH4/H2 mixture, reinforcing the material’s selectivity, closely matching the values predicted by the isotherm from single component experiments. These findings highlight the potential of 3D-printed porous carbon monoliths for selective CH4 separation in PSA processes applied to SMR, offering a promising alternative with lower pressure drop and greater structural control compared to conventional adsorbents.