http://repositorio.ufc.br/handle/riufc/464 2026-06-10T05:50:11Z 2026-06-10T05:50:11Z Barbosa, Klysmann Gladson Ferreira http://repositorio.ufc.br/handle/riufc/86617 2026-06-05T00:48:01Z 2026-01-01T00:00:00Z

Título: Controle de posição da junta de manipulador cilíndrico acionado por motor de indução trifásico com transmissão flexível e otimização conjunta de sintonia Autor(es): Barbosa, Klysmann Gladson Ferreira Abstract: This work investigates the position control of the base rotational joint of a cylindrical manipulator driven by a three-phase induction motor with a flexible mechanical transmission, with emphasis on the systematic tuning of controllers. The electromechanical system is modeled in an integrated manner, including the three-phase induction motor under Indirect Field-Oriented Control (IFOC), the manipulator dynamics, and the elastic motor-load coupling represented by a two-mass model derived from the Euler-Lagrange formulation. The study focuses on the influence of flexible mechanical coupling on control performance and on the systematic determination of controller parameters using metaheuristic and probabilistic optimization methods. Different cascaded control architectures combining P and PI controllers with Adaptive Sliding Mode Control (ASMC) are evaluated, both in sensored operation and in sensorless configuration using a Sliding Mode Observer (SMO). Additionally, the influence of measurements obtained from a non-ideal incremental encoder on the quality of state feedback is considered. The results show that transmission elasticity introduces resonant modes that significantly degrade performance when conventional controller parameters are used. Systematic optimization-based tuning improves the trade-off between response speed, accuracy, and robustness, and allows the identification of operating regions in which PID controllers achieve performance comparable to robust control strategies, as well as conditions in which ASMC provides significant advantages under parametric uncertainties and external disturbances. Tipo: Dissertação

2026-01-01T00:00:00Z Landim, Pedro Lino Azevedo http://repositorio.ufc.br/handle/riufc/86616 2026-06-04T23:52:20Z 2025-01-01T00:00:00Z

Título: Plataforma web de anotações interativas para segmentação de neuroimagens por ressonância magnética Autor(es): Landim, Pedro Lino Azevedo Abstract: The growing volume and complexity of medical imaging, especially magnetic resonance imaging, have increasingly overwhelmed healthcare professionals and highlighted the need for computer-aided diagnostic tools. In this context, this work presents the development of an interactive web platform for the segmentation of magnetic resonance neuroimages, integrating artificial intelligence techniques and accessible visualization. The tool was designed with a modular architecture composed of a Flutter Web interface, a Flask-based API, and deep learning models implemented in PyTorch. The system performs automatic segmentation using the U-Net architecture and enables dataset enhancement through mask generation using the Flood Fill algorithm, validated by a CNN. The interface provides features for image upload, region of interest marking, result visualization, and retraining of the model with customized data. Load and stress tests were conducted to assess system performance, along with both quantitative and qualitative analyses of the segmentations. The U-Net segmentation model achieved a mean Dice Score of 90.22% , while the CNN for mask validation obtained 97.49% accuracy. These results, combined with the API’s high success rate under stress, demonstrate the feasibility of applying the platform in clinical and research environments, highlighting its flexibility, adaptability, and integration with medical workflows. Tipo: Dissertação

2025-01-01T00:00:00Z SOARES FILHO, DÁRIO CONCEIÇÃO http://repositorio.ufc.br/handle/riufc/86332 2026-05-17T22:47:52Z 2025-01-01T00:00:00Z

Título: Análise de viabilidade técnico-econômica aplicada à energia eólica offshore Autor(es): SOARES FILHO, DÁRIO CONCEIÇÃO Abstract: The global energy transition and the need to decarbonize power systems have fostered the expan sion of renewable sources, with offshore wind standing out for its high technical potential and strategic relevance. Brazil, particularly the Northeast coast, offers exceptional wind conditions and a broad continental shelf, with potential exceeding 700 GW. This dissertation develops and applies an integrated techno-economic assessment model for offshore wind generation, with a case study in Pecém, Ceará. The methodology combines statistical wind modeling through the Weibull distribution, integration with the power curve of a 10 MW NREL/IEA turbine, calculation of the FCE and AEP, and explicit consideration of transmission losses. The economic evaluation covers LCOE, Net VPL, TIR, and PBd, based on realistic CAPEX, OPEX, and tariff assumptions. Uncertainty analysis is carried out through sensitivity tests and Monte Carlo simula tion. Results indicate FCE between 62% and 66% and AEP between 2,716 and 2,891 GWh/year, ensuring international competitiveness. LCOE ranges from 68 to 117 US$/MWh (P5–P95), with averages of 78–100 US$/MWh, while discounted PBd spans 6–12 years, averaging 7–9.5 years. Probabilistic analysis shows that over 80% of simulations yield LCOE below 100 US$/MWh and Payback around 9 years, confirming project attractiveness even under uncertainty. Offshore wind in Pecém is thus a viable and strategic option for expanding Brazil’s power matrix and integrating into emerging chains such as green hydrogen, providing technical input for policies and methodological advances by unifying deterministic and stochastic analyses. Tipo: Dissertação

2025-01-01T00:00:00Z Assis, Dionizio Porfírio de http://repositorio.ufc.br/handle/riufc/86105 2026-05-01T21:05:04Z 2025-01-01T00:00:00Z

Título: Monitoramento ambiental e operacional de plantas fotovoltaicas em solo e flutuante com sistemas IoT utilizando comunicação LoRa Autor(es): Assis, Dionizio Porfírio de Abstract: The advancement of renewable energy sources, especially solar energy, requires precise and accessible monitoring systems capable of ensuring the efficient operation of photovoltaic (PV) plants. In this context, this dissertation presents the implementation and validation of a Photovoltaic Monitoring System (PMS) applied to ground-mounted and floating photovoltaic (FPV) plants, conceived as a complete, accessible, and modular solution based on open, well- documented technologies widely used in the IoT development community. The proposed architecture integrates minute-level data acquisition, point-to-point LoRa communication, cloud synchronization through the ThingSpeak platform, and a web system with a MySQL database for data storage, monitoring, and export. The system was deployed in the experimental environment of the Alternative Energies Laboratory (LEA) of the Federal University of Ceará (UFC), known as LEA 3, simultaneously instrumenting a tank with floating modules, a reference tank, and a ground-mounted installation. Variables monitored included irradiance, ambient temperature, relative humidity, wind speed, and temperatures on the modules and in the tank water. The adoption of the acknowledgment (ACK) reception protocol increased the packet delivery rate from values below 50% to 95.49%, remaining at 91.76% throughout 2024. Cross-validation with INMET data confirmed the accuracy of the measurements, with a mean error of 2.12% for irradiation, 1.68% for ambient temperature, and 7.49% for relative humidity. Thermal analyses indicated homogeneity between the center and edge sensors of the modules, with an average difference of less than 1 °C, and similar behavior between ground-mounted and floating modules. For the tanks, it was observed that the reservoir with floating coverage kept the water cooler at maximum temperatures and showed lower daily thermal variability, with an average difference of 1.34 °C compared to the reference tank and a standard deviation of 0.16 °C versus 0.67 °C in the uncovered tank. The work contributes technically to the advancement of PV and FPV monitoring in Brazil by providing an open, reproducible, and experimentally validated IoT architecture that integrates LoRa communication, a web platform, and metrological validation with official data. It is concluded that the proposed system meets the objectives of monitoring, storing, and providing environmental and operational data reliably, establishing an alternative for intelligent monitoring of PV plants. Tipo: Dissertação

2025-01-01T00:00:00Z