EMC - Escola de Engenharia Elétrica, Mecânica e de Computação

URI Permanente desta comunidade

http://200.137.215.59//handle/ri/587

A Escola de Engenharia Elétrica, Mecânica e de Computação (EMC), da Universidade Federal de Goiás, oferece Graduação em: Engenharia Elétrica, Engenharia Mecânica e Engenharia da Computação. Além de Especialização em: Telecomunicações; Instalações Elétricas; e, Gestão de Negócios.

Navegar

Agora exibindo 1 - 20 de 52

A inspeção técnica veicular e seu impacto no comércio de veículos usados
(Universidade Federal de Goiás, 2023-08-14) Libório, João Pedro Neri; Oliveira, Ademyr Gonçalves de; Oliveira , Ademyr Gonçalves de; Figueiredo, Kleber Mendes de; Franco, André Luiz Carneiro
The present work aims to present the Vehicle Technical Inspection (ITV) alongwith all its technical standards regarding vehicle safety and emission of gases, anddemonstrate how the lack of standardization and regulation by the State leads to aprecarious aging of the Brazilian vehicle fleet, resulting in incalculable damagesdirectly and indirectly to the population. According to the Federal Highway Police, asignificant portion of vehicular accidents occurring in the country today are caused bymechanical defects and lack of maintenance. This explanation illustrates how thistype of accident ends up burdening the Brazilian public healthcare system throughhospitalizations and medical procedures. Additionally, the project also aims toanalyze how the ITV can impact the trade and maintenance of used vehicles in thecountry, highlighting the lack of reliability perceived by a considerable portion of thepopulation, including dealers, intermediaries, mechanical workshops, and the finalcustomer, regarding second-hand automobiles, as it is notorious that a large numberof vehicles with numerous mechanical and structural defects are licensed and solddaily. Lastly, it also seeks to understand and compare how the ITV is practiced inother developed countries, including its standards and obligations, as well as itsperiodic implementation at workstations, in order to explore possible new solutionsand improvements for Brazilian regulations and standards, by adapting successfulexperiences from other global contexts.
Adaptação e implementação da framework NIST em redes domésticas
(Universidade Federal de Goiás, 2024-01-31) Soares, Luan Weba; Pinheiro Júnior, Carlos Galvão; Pinheiro Júnior, Carlos Galvão; Oliveira, Marco Antonio Assfalk de; Alves, João Miguel Estevão
This study addresses the implementation and feasibility of the National Institute of Standards and Technology Cybersecurity Framework (NIST CSF). We explore the challenges brought by the rise of remote work in residential environments, both of which contribute to a more complex security landscape. Using the Raspberry Pi equipped with security tools such as SIEM and EDR, and vulnerability scanning techniques, the study proposes a comprehensive set of security practices based on the NIST CSF. It aims to provide a clear, low-cost roadmap for effectively modeling cybersecurity measures, protecting data integrity and privacy on home networks. Finally, we arrived at the result that the framework is adaptable, but requires a lot of intervention from specialists, to the point that support by them rather than the end user is more appropriate.
Adequação de uma ferramenta off-line para a programação de um robô manipulador dedicado ao processo de manufatura aditiva por deposição a arco
(Universidade Federal de Goiás, 2023-11-08) Matos, Giovana Andrade de; Ferreira Filho, Demostenes; Ferreira Filho, Demostenes; Fonsceca, João Paulo da Silva; Lopes, Gustavo de Castro
The Wire Arc Additive Manufacturing (WAAM) process has great potential for the manufacture of metallic parts. However, although the procedures are similar to those used in polymer 3D printing, which is already established as an industrial process, WAAM still faces several challenges to become mature enough to be applied on a large scale in industry. One of the ways to carry out the WAAM process is through manipulator robots, which play a significant role in this technique. Deposition trajectories can be generated in two main ways: by "teach" mode or by "Offline" programming. This project adapted a commercial off-line programming tool for the application of the WAAM process through the development of a virtual workstation identical to the one available at the Laboratory of Materials and Manufacturing Processes, programming of the robotic cell using the RoboDK software and the Teach programming mode, and the manufacture of a part with a complex shape (airfoil NACA 04) by the WAAM process.
Alocação de recursos em redes sem fio utilizando algoritmos baseados em aprendizagem de máquina
(Universidade Federal de Goiás, 2024-02-01) Lopes, Adriano Ferreira; Silva, Jean Lucas Barbosa; Vieira, Flávio Henrique Teles; Vieira, Flávio Henrique Teles; Cardoso, Alisson Assis; Carneiro, Daniel Porto Queiroz
With the increasing rise in the number of mobile devices and intelligent IoT devices, wireless networks have become increasingly complex, autonomous, and heterogeneous in terms of the types of network architectures they incorporate. Within these networks, efficiently allocating and managing resources for users is a significant problem to be solved. In this context, deep reinforcement learning techniques are expected to be among the main technologies used to achieve global optimization in dynamic resource allocation. This paper presents a proposal for resource allocation in wireless networks, termed Cross-Entropy Reinforcement Learning, in order to maximize energy efficiency and meet users’ quality of service (QoS) requirements. The approach considers a multi-user, multi-objective communication system based on CP-OFDM (Cyclic-Prefix Orthogonal Frequency Division Multiplexing) technology using reinforcement learning methods, specifically a Deep Q-Network (DQN) associated with the CrossEntropy algorithm to obtain an optimal resource allocation policy. The implementation encompasses system parameters such as bandwidth, modulation, number of users, and average packet size, while detailing the structure of resource elements, subcarrier distribution, and transmission capacity in different modulation modes within the LTE (Long Term Evolution) system frame. Simulation results demonstrate that the proposed method exhibits good convergence characteristics and performs better than the traditional DQN approach without cross-entropy.
Análise da inversão do fluxo de potência em instalação elétrica industrial com geração de energia solar fotovoltaica
(Universidade Federal de Goiás, 2024-12-03) Ferreira, Gabriel Fernandes; Belchior, Fernando Nunes; Belchior, Fernando Nunes; Oliveira, Antônio Melo de; Castro, Marcelo Stehling de; Pires Junior, Kairo Hitler De Almeida Faria
The increasing integration of distributed generation, particularly solar photovoltaic energy, in industrial installations has revealed critical challenges for the stability and quality of electrical systems. Phenomena such as power flow inversion, harmonic distortions, current and voltage imbalances, and power factor variations add complexity to electrical operations, demanding advanced monitoring and control techniques. These challenges are further exacerbated in industrial contexts, where energy demand is highly dynamic and subject to abrupt changes, intensifying the interactions between local generation and the electrical grid. Such dynamics can significantly impact system reliability, compromising equipment efficiency, reducing lifespan, and increasing operational costs. Moreover, energy quality distortions can cause premature wear of critical components, transformer overloads, and disruptions to plant operations, directly affecting productivity. These consequences highlight the need for solutions that align electrical operations with regulatory compliance, ensuring system stability and efficiency. The main motivation for this study emerged from an incident reported by the maintenance team at Volga Engenharia, where an overload in the entry transformer tripped its protections, disconnecting it from the system and preventing catastrophic equipment failure. This critical event highlighted the need to deeply understand the phenomena associated with distributed generation, particularly power flow inversion and its impacts on the electrical system. In this context, the opportunity arose to align this academic research project with the challenges faced by the company, focusing on practical and technical solutions. Throughout the development of this work, the scope of the analysis expanded to address additional issues related to energy quality and specific operational conditions of the plant. Data obtained during monitoring provided a detailed view of the effects of abrupt power flow inversion and its implications on other critical parameters, such as current imbalances and harmonic distortions. This analysis reinforces the importance of studies like this one, aiming not only to understand the challenges posed by distributed generation but also to propose solutions that ensure efficiency, safety, and sustainability for industrial systems.
Análise de faturamento para consumidores do grupo A: estudos de caso e propostas de otimização
(Universidade Federal de Goiás, 2024-11-12) Correia, Matheus Vicente; Belchior, Fernando Nunes; Belchior, Fernando Nunes; Oliveira, Antônio Melo de; Castro, Marcelo Stehling de; Santos, Josephy Dias
This study presents a detailed analysis of billing for Group A consumers, focusing on case studies and optimization proposals. The research addresses the current tariff models in Brazil and their application in calculating electricity consumption and demand, considering the importance of the electricity bill history as an analysis tool. Additionally, concepts such as power factor and the use of capacitor banks for reactive power correction are explored, aiming at reducing penalties and improving energy efficiency. Based on case studies, solutions are proposed for optimizing consumption and contracted demand of the analyzed consumer units to promote cost reduction and improve energy management. It concludes that adopting best practices in demand sizing and efficient use of reactive energy can result in financial benefits for both energy distributors and consumers. This work contributes to SDG 7 – Affordable and Clean Energy by promoting energy efficiency practices and consumption optimization, which aim at both cost reduction and the conscious use of electricity, positively impacting the sector's sustainability and fostering the development of solutions that minimize the waste of natural resources.
Análise de sistemas de ventilação e ar-condicionado utilizando técnicas de Dinâmica dos Fluidos Computacional com software livre
(Universidade Federal de Goiás, 2023-08-14) Gratão, Gustavo Gonçalves; Mariano, Felipe Pamplona; Mariano, Felipe Pamplona; Nascimento, Andreia Aoyagui; Vilela, Carlos Alberto de Almeida
In this work, the importance of indoor air quality for people's health was emphasized, highlighting the relevance of an adequately ventilated and air-conditioned environment for the well-being and comfort of occupants. Analyzes of different ventilation and air-conditioning systems were carried out using computational fluid dynamics techniques with free, open source and free software. Simulations of internal air flows in an enclosure were conducted using OpenFoam, while the data obtained were processed using Paraview. The computational experiments results were promising, with the model employed providing satisfactory results. These results demonstrate the potential application of the methods used in other analyzes and provide valuable information for the optimization of ventilation and air conditioning systems in different environments.
Análise do funcionamento das funções 51, 27, 59 e 87 em diversas condições de curto-circuito de uma linha de transmissão de alta tensão utilizando o software ATPDraw
(Universidade Federal de Goiás, 2024-12-12) Souza, Alex de Almeida Rodrigues de; Campos, Laura Ferreira; Almeida, Maria Leonor Silva de; Almeida, Maria Leonor Silva de; Kopcak, Igor; Belchior, Fernando Nunes
The present study investigates, through the implementation in the ATPDraw software, the behavior of four functions commonly used in the protection of high-voltage transmission lines: overcurrent (51), overvoltage (59), undervoltage (27), and differential (87) functions. ATPDraw, a program used for simulating electromagnetic transients in power systems, servesas a tool to study the behavior of these protection functions under different fault scenarios. It offers a practical and detailed approach, enabling the modeling of transient behavior in electrical systems and the evaluation of how different protection strategies can mitigate interruption risks, contributing to a safer and more reliable electrical grid. The results revealed that the performance of protection functions is strongly influenced by operating conditions, such as the type of short circuit (monophase or three-phase), fault resistance, and fault location. While functions like 87 proved to be highly effective and consistent across all tested scenarios, current- or voltage-based functions such as 51, 27, and 59 exhibited variations under specific conditions, such as high fault resistance or fault locations. These findings highlight the importance of precise adjustments and the integration of diverse protection strategies to maximize system reliability and safety. The research concludes that simulation tools like ATPDraw are essential for validating protection settings and enhancing the performance of complex electrical systems.
Análise modal estática aplicada ao problema de fluxo de carga em coordenada retangular para avaliação de estabilidade de tensão
(Universidade Federal de Goiás, 2024-02-02) Sena, Katarina Leite de; Santos, Vinícius Morais Calixto dos; Brigatto, Gelson; Kopcak, Igor; Kopcak, Igor; Alves, Antônio César Baleeiro; Garces Negrete, Lina Paola
A voltage instability is a phenomenon directly linked to the inability of an electrical power system to meet its demand for active power, caused by inadequate reactive power support. Traditionally, voltage stability margins are calculated through static methods based on the load flow problem. Additionally, Static Modal Analysis has been a tool foridentifying critical buses or areas regarding voltage stability. This methodology uses the representation of voltage phasors in cylindrical or polar coordinates. In contrast, this study proposes applying Static Modal Analysis to the Jacobian of Load Flow in Rectangular Coordinates, where voltage phasors are represented by their Cartesian coordinates on the real and imaginary axes. The aim was to assess whether this new approach could provide additional qualitative information for identifying critical buses regarding voltage stability, complementing traditional Static Modal Analysis. For the analysis of results, well-known transmission networks from the literature were employed, and for comparing the performance of the methods developed in this study, Static Modal Analysis in polar coordinates was adopted as a reference. The results obtained demonstrate that the method developed in this study is satisfactory, as the provided voltage stability margins are close to those of the traditional method, and the Static Modal Analysis applied to Rectangular Load Flow classified the critical buses of the tested systems correctly.
Aplicação de realidade aumentada para monitoramento e manutenção da infraestrutura computacional na indústria 4.0
(2024-12-12) Amaro, Gleycykelly Syssy Indymayer Carnot; Castro, Marcelo Stehling de; Castro, Marcelo Stehling de; Oliveira, Gustavo Dias de; Lemos, Rodrigo Pinto
Industry 4.0 aims to enhance production chains through the integration of various technologies, promoting process optimization and greater operational efficiency. Among these technologies, Augmented Reality (AR) stands out, allowing the overlay of 3D digital objects onto real-world environments, enabling interaction and real-time data visualization. This work aims to develop an Android application for the management and maintenance of computational infrastructures in smart industries, utilizing AR resources and tools such as Unity and Vuforia. The application will provide the visualization of computer metrics and information, in addition to integrating an interactive 3D model, facilitating maintenance and contributing to the reduction of unnecessary downtime.
Aplicação dos metamateriais acústicos na transmissão sonora
(2024-12-18) Silva, Bruno Cordeiro e; Fagundes Neto, Marlipe Garcia; Fagundes Neto, Marlipe Garcia; Kitatani Junior, Sigeo; Rodrigues, Marcos Vinicius Silva
Metamaterials are structures designed to exhibit physical properties not found in natural materials, such as controlled manipulation of sound waves, enhancing acoustic mitigation. When combined with additive manufacturing, this technology offers new possibilities for acoustic insulation. This study aims to explore the concept and application of the acoustic metamaterial model UOM (Ultra-open Metamaterial) proposed by Ghaffarivardavagh (2018), validating its efficiency in sound transmission loss based on its geometry. Additionally, the research investigates the advantages of simulations applied to acoustic silencers, evaluating their accuracy through impedance tube tests in conjunction with additive manufacturing of the silencer. The results indicate that the study's objectives were achieved, showing alignment between analytical, simulated, and tested results, with error margins ranging from 0.4% to 8.0%. The tests also showed proximity to the simulations, with a 3.5% error in the frequencies of maximum transmission loss, despite differences in peak absolute values. However, the fabrication of physical models using 3D resin printing revealed structural weaknesses due to reduced dimensions, highlighting challenges to be addressed for improved practical feasibility.
Atualização de planta de instalação elétrica industrial
(Universidade Federal de Goiás, 2024-02-02) Araujo, Vitor Montero; Araujo, Bruno Alves de; Alvarenga, Bernardo Pinheiro de; Alvarenga, Bernardo Pinheiro de; Santos, Euler Bueno dos; Paula, Geyverson Teixeira de
All elements of a building undergo changes over the years, to a greater or lesser extent, the original characteristics are altered and not recorded, making it difficult to understand for those who did not participate in the processes. This is especially complex for electrical installations, as the addition of new equipment, jobs, and other changes bring problems that can lead to accidents, such as an increase in load beyond what the circuit was designed to handle, generating excessive heating. Renovating a building is a great time for all these issues to be resolved. This is the case of Mercabox Araujo, a supermarket company that is undergoing an expansion that will double its area, a situation that requires the renewal of its entire electrical installation to support the new installed power and a new layout to meet future activities. This final project aims to propose the design of the new Mercabox Araujo installations, taking into account the customer's demands and compliance with the standards that govern low voltage installations, ensuring reliability and safety. Furthermore, a step-by-step guide was also proposed so that the transition between the old and new installation does not cause disruption to the company's activities, which will continue to operate during the works.
Avaliação da confiabilidade em sistemas de distribuição de energia elétrica com alta inserção de geração distribuída: uma abordagem através do método de Newton-Raphson e da simulação de Monte Carlo
(Universidade Federal de Goiás, 2024-12-11) Sabreu , Igor Lyncoln da Silva; Garces Negrete, Lina Paola; Garces Negrete, Lina Paola; Brigatto, Gelson Antônio Andréa; Belchior, Fernando Nunes
This study aimed to evaluate the influence of distributed generation (DG) on reliability parameters in electrical distribution systems, analyzing such metrics before and after allocation. In a different way, this work adopts the integration of the Monte Carlo technique with the Newton-Raphson method, applied to the resolution of the power flow problem, incorporating islanding detection based on Graph Theory. In addition, the use of the goodness-of-fit test (adherence) allows validating the adequacy of probabilistic distribution functions to stochastic variables, ensuring a diagnosis that is more faithful to real behavior. The analyzes were performed using the 14-bus RBTS test system. To this end, five indices were chosen to assess grid reliability: LOLE (Loss Of Load Expectation), LOEE (Loss OF Energy Expectation), FEC (Frequency of Exceeding Capacity), IEC (Interrupted Energy Consumption) and IVR (Interruption Value Rate) , which together provide an excellent perspective of the system. With the addition of decentralized generation, an increase in system availability was noted, a decrease in voltage deviations in the grid, a reduction in unserved energy and fewer occurrences of failures. Thus, the integration of decentralized generation not only improves the parameters evaluated, but also highlights the points in the network that require greater care and allocation of resources, whether in protection devices and systems or in predictive maintenance actions, preventing possible technical complications and excessively high corrective expenses, following the objectives and guidelines of SDG 7 (Sustainable Development Goals), proposed by the UN (United Nations), which aims at clean and affordable energy.
Avaliação da sobreamostragem de dados de voz na classificação automática da doença de Parkinson
(Universidade Federal de Goiás, 2024-12-19) Silva, Matheus Isac da; Felix, Juliana Paula; Felix, Juliana Paula; Silva, Karina Rocha Gomes da; Salvini, Rogerio Lopes
This study investigates a possible bias in oversampling via data windowing of vocal signals. Previous studies indicate that there is a bias for gait data when the data is treated independently, in addition there are statistical studies that show that data from the same individual carry similar information. An approach based on three databases containing vocal signals was used, two of which were unbalanced and one balanced. The K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Naive Bayes and Decision Tree (DT) algorithms were applied, with pre-processing using StandardScaler and PCA behavior analysis. Cross validation was done with k-fold Cross Validation, with k=5, in all 3 bases, adapted for scenarios with and without bias in the training data. Models evaluated without considering bias showed inflated performances, while the rigorous approach showed more modest results. It is concluded that samples from the same individual in training and testing can inflate the performance of models, and it is crucial to apply oversampling correctly to develop reliable models for diagnosing PD.
Avaliação de parâmetros e utilização do tempo de parada na qualidade construtiva de uma peça de alumínio fabricada por manufatura aditiva por deposição a arco
(2023-08-14) Silva, Felipe Cardoso Tavares; Ferreira Filho, Demostenes; Ferreira Filho, Demostenes; Rodrigues, Thales Marcelo Pereira; Kihara, Eliane Alves; Souza, Daniel
An aspect of the Additive Manufacturing (AM) manufacturing process that has been widespread in the Biomedical, Automobile and Aerospace industries is the Wire Arc Additive Manufacturing (WAAM). WAAM consists in the manufacture of metallic components from the deposition of multilayers of weld beads on a substrate, that is, from a three-dimensional model, the object is divided into layers, defining the trajectory in which the addition of metal, performing it by means of a robotic manipulator. The AM has advantages when compared to other manufacturing processes, such as conventional machining, such as a high deposition rate, less material waste, and the possibility of manufacturing complex parts of large proportions. However, it has been facing some challenges to become sufficiently mature and to be applied on a large scale in the industry, thus demanding a series of studies and research for its consolidation and maturation. The objective of this work was to contribute to this maturation through the evaluation of the influence of the parameters used in the manufacture by the WAAM process of an aluminum part with a relatively complex profile and the evaluation of the influence of the use of stop between layers in the manufacture of this part. For this, a pulsed GMAW (Gas Metal Arc Welding) welding system was used coupled to a Yaskawa Motoman HP20D-A80 robotic manipulator, with ER4043 aluminum wires of 1.2 mm in diameter. After programming the trajectories, preliminary tests were carried out in the format of an airfoil, having as standardized parameters the Contact Tip to Work Distance (CTWD) in 18 mm and the rise GAP in 0.5 mm and as varied parameters the deposition velocity (cm/min), the welding current in synergic mode (A) and the pure argon gas flow rate (L/min). Next, two airfoils were manufactured with 100 layers each, the first without stopping and the second with a 30-second stop between each layer, to perform a dimensional analysis and compare the time spent in the manufacture, height, thickness, porosity, and surface finish.
Avaliação do efeito da trajetória na fabricação de roldana de FeCrC-Nb pelo processo de manufatura aditiva por deposição a arco
(Universidade Federal de Goiás, 2023-09-15) Carvalho, Lúcio Mateus Santana; Ferreira Filho, Demostenes; Ferreira Filho, Demostenes; Alves, Renata Rodrigues; Marques, Vitória Ferreira de Oliveira
The welding process with flux cored wire (FCAW), using FeCrC+Nb alloy, made it possible to produce high hardness parts. This project evaluated the use of different weld patterms to analyze hardness values in different areas of the part, along the bead and weld layers. Two sheaves were made by different welding patterns, weaving and concentric circles (CC). The presence of the primary carbides M7C3 and NbC was found to affect hardness. The weaving pattern presented higher hardness than the CC pattern, which is justified by the strong presence of hard carbides.
Avaliação no ATPDraw das proteções diferencial e de sobrecorrente aplicadas a um transformador de potência
(Universidade Federal de Goiás, 2024-08-01) França, André Luiz Almeida de; Almeida, Maria Leonor Silva de; Almeida, Maria Leonor Silva de; Brigatto, Gelson Antônio Andrêa; Kopcak, Igor
This work analyzed the differential and overcurrent protections in ATPDraw for a YDn1 power transformer, with the high-voltage side connected in a grounded wye configuration and the low-voltage side in delta. The protections were implemented in ATPDraw with differential protection blocks (87T), instantaneous overcurrent (50), and time-delayed overcurrent (51). Phase overcurrent functions (50F/51F) and neutral overcurrent functions (50N/51N) were incorporated on the high-voltage side. In 87T block of ATPDraw, it was necessary to include the zero-sequence compensation and angular correction. Faults were simulated on the transformer’s busbars and bushings, covering conventional and internal faults caused by short circuits turn-to-turn and turnto-ground. The 50F and 51F functions demonstrated effectiveness and selectivity in responding to faults on the bushings and busbars. The 50N and 51N functions operate to ground faults, except for those on the low-voltage side, as the delta connection prevents the high-voltage side from contributing to zero-sequence components generated by ground faults. Additionally, the 50N and 51N functions responded quickly to turn-to-ground short circuits in the grounded wye configuration. However, none of the overcurrent functions could defect turn-to-turn or turn-to-ground faults on the delta-connected windings. The differential protection 87T, with zero-sequence compensation and angular correction, performed well for faults on the busbars and bushings. However, it required a high percentage of short-circuited turns on the low-voltage side to operate, specifically 90% for turn-to-ground short circuits and 65% for turn-to-turn short circuits. Therefore, it was recommended that the relay sensitivity be increased by adjusting the slope of the second region to enhance fault detection on the delta side.
Consumo energético: análise no Brasil e no mundo
(Universidade Federal de Goiás, 2024-07-16) Trautenmuller, Artur Schneider; Castro, Marcelo Stehling de; Castro, Marcelo Stehling de; Oliveira , Gustavo Dias de; Belchior, Fernando Nunes
In recent years, companies and institutions have been increasingly investing in information analysis systems both for decision-making and for gaining a deeper understanding of the reality they are embedded in. One of the key areas in this investment is Business Intelligence (BI). This final course work focuses on the collection, organization, and analysis of data to provide valuable insights that drive informed decision-making within organizations (BI)systems help companies better understand their internal processes, the market they operate in, and even future trends. This project proposes to build analytical dashboards related to electricity and primary energy consumption worldwide and in Brazil by consolidating information from different sources.
Coordenação de proteção e seletividade em redes de distribuição com penetração de geração distribuída
(Universidade Federal de Goiás, 2024-08-02) Lobo, João Pedro do Carmo; Pacheco, João Victor Miranda; Kopcak, Igor; Kopcak, Igor; Sena, Carla Silva; Almeida, Maria Leonor Silva de
The work presents a detailed study on the coordination of protection and selectivity in electrical distribution networks with distributed generation, focusing on a 1 MW Small Hydroelectric Power Plant (CGH) located in the interior of Goiás. The study includes the analysis of critical operating scenarios, equipment specification, protection functions, and the parameterization of protection relays. The necessary adaptations to ensure safe and coordinated operation with the utility's protection devices, such as substation relays, reclosers, and fuses, are addressed. The main objective is to highlight the importance of selectivity and coordination in the protection system, especially considering the increasing penetration of distributed generation, which introduces additional challenges such as bidirectional current flow and variations in voltage and frequency levels. This encompasses the need to adapt the parameters of protection relays to maintain selectivity and avoid improper operations that could compromise the continuity of energy supply. The detailed analysis of operating scenarios takes into account different fault conditions and the influence of distributed generation on these events, providing an in-depth understanding of how distributed generation can impact the operation of distribution networks. The specification of equipment and protection functions is carried out in a way that ensures relays operate in a coordinated manner, both with the CGH equipment and with the devices in the distribution network. Correct parameterization of the relays is essential for them to adequately discriminate fault types and act selectively, ensuring system protection and minimizing interruptions. The study highlights the importance of implementing practices and standards that consider the particularities of distributed generation, ensuring reliable and efficient operation of the electrical system.
Deep learning para reconhecimento de sinais da LIBRAS como tecnologia assistiva
(Universidade Federal de Goiás, 2024-12-12) Pedrosa, Samuel França da Costa; Santana, Adriano César; Santana, Adriano César; Castro, Marcelo Stehling de; Nerys, José Wilson Lima
Communication is essential for social inclusion, yet the lack of linguistic accessibility often marginalizes specific groups, such as the Brazilian deaf community. This project proposes the use of deep learning to recognize signals from Brazilian Sign Language (LIBRAS) with the aim of translating them, through model predictions, into written Portuguese. Initially, videos of people signing were collected to form a dataset, which was subjected to processes of extraction and mapping of key body points using open-source tools provided by MediaPipe. The extracted data was processed and used as input for two designed models: one based on Long Short-Term Memory (LSTM) and another on Transformers. The study revealed that model performance is influenced by the alignment methods applied during data processing. The Transformer demonstrated superior results in terms of accuracy and generalization, albeit with higher computational demands. Conversely, the LSTM model showed satisfactory performance in terms of computational efficiency but exhibited limitations as classification complexity increased. One of the primary challenges was the difficulty in building a rich and robust dataset, due to the scarcity of available content for collection and extraction, especially when compared to other natural languages, whether textual or spoken. This limitation partially restricted the models' generalization capabilities. Despite these challenges, the project achieved promising results, suggesting that with enhanced and expanded datasets, its application as assistive technology can be extended to more complex scenarios with broader applicability. This study represents an advancement in the use of deep learning to promote inclusion and accessibility for the Brazilian deaf community.

Navegar

Navegando EMC - Escola de Engenharia Elétrica, Mecânica e de Computação por tipo de documento "Trabalho de conclusão de curso de graduação (TCCG)"

Resultados por página

Opções de Ordenação