PPGEE - Mestrado em Engenharia Elétrica e de Computação

Permanent URI for this collectionhttps://repositorio.ufrn.br/handle/123456789/12008

Browse

Now showing 1 - 20 of 567

Traçamento de raios sísmicos com múltiplas condições iniciais usando Redes de Kolmogorov-Arnold com Informações Físicas
(Universidade Federal do Rio Grande do Norte, 2025-12-05) Marques, Lucas Torres; Barros, Tiago Tavares Leite; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; https://orcid.org/0000-0001-9665-2238; http://lattes.cnpq.br/1321568048490353; http://lattes.cnpq.br/0281845604882859; Silva, Ivanovitch Medeiros Dantas da; https://orcid.org/0000-0002-0116-6489; http://lattes.cnpq.br/3608440944832201; Nose Filho, Kenji; http://lattes.cnpq.br/7933130617235231
Seismic wave propagation modeling is fundamental for seismic imaging and reservoir characterization in the oil and gas industry. While ray tracing is an efficient technique, traditional methods face challenges in complex geological models due to instability and high computational cost, as they require solving for a single initial condition at a time. This work proposes the use of Physics-Informed Kolmogorov-Arnold Networks (PIKANs) to overcome these limitations by allowing the simultaneous tracing of multiple rays with different initial conditions. The proposed methodology includes velocity model smoothing with B-spline filters, the use of a compound loss function, and the application of adaptive weights in the physics-informed loss function to enhance model training. Two experiments were conducted using synthetic data: the first with a parameterized velocity model, and the second with a non-parameterized model widely used as a benchmark in seismic processing, known as the Marmousi model. The results demonstrate the consistency of the approach, with R2 values close to 1 for the first experiment and exceeding 0.90 for the second. Qualitatively, the method accurately traces seismic rays, even in regions not used for training and in areas with rapid spatial velocity variation, proving to be more precise at predicting the ray’s spatial coordinates than the slowness vector.
Substituições de aminoácidos como representação de mutações em transformers para análise e classificação de variantes do SARS-CoV-2
(Universidade Federal do Rio Grande do Norte, 2025-12-19) Silva, Alessandro Soares da; Fernandes, Marcelo Augusto Costa; https://orcid.org/0000-0001-7536-2506; http://lattes.cnpq.br/3475337353676349; http://lattes.cnpq.br/6391394214446756; Câmara, Gabriel Bezerra Motta; https://orcid.org/0000-0002-5736-0782; http://lattes.cnpq.br/3849103334728892; Terrematte, Patrick César Alves; https://orcid.org/0000-0002-0385-0030; http://lattes.cnpq.br/4283045850342312; Barbosa, Raquel de Melo; https://orcid.org/0000-0003-3798-5512; http://lattes.cnpq.br/8174411728088087
The COVID-19 pandemic, caused by SARS-CoV-2, highlighted the need for scalable and automatable computational methods for genomic surveillance in the face of the continuous emergence of variants of concern. In this context, this work proposes an innovative approach based on vector representations of amino acid substitutions, leveraging Transformer models for the analysis and classification of SARS-CoV-2 viral variants, with a focus on the Alpha, Beta, Gamma, Delta, and Omicron variants. Sequences of amino acid substitutions were processed into high-dimensional embeddings, which supported two complementary experiments. In the first, unsupervised techniques, such as Fuzzy C-Means clustering and t-SNE projection, were applied to reveal groupings consistent with known variants, as well as transitional regions and ambiguous samples. In the second, supervised learning models, including SVM, Random Forest, k-NN, and XGBoost, were evaluated for variant classification, with XGBoost achieving accuracy above 99% on an external dataset. The results demonstrate that representations derived exclusively from amino acid substitutions are capable of discriminating viral variants with high performance and capturing mutational signatures of biological relevance, without requiring genomic alignment. Thus, the proposed approach represents a scalable, flexible, adaptable, and purely computational alternative for automated pathogen genomic surveillance, with potential applications in public health scenarios.
Musicalization with educational robotics
(Universidade Federal do Rio Grande do Norte, 2025-12-19) Andrade, José Lindenberg de; Gonçalves, Luiz Marcos Garcia; Alsina, Pablo Javier; http://lattes.cnpq.br/3653597363789712; https://orcid.org/0000-0002-7735-5630; http://lattes.cnpq.br/1562357566810393; http://lattes.cnpq.br/8823342043920921; Curvelo, Carla da Costa Fernandes; http://lattes.cnpq.br/2329425562585943; Souza, Anderson Abner de Santana; https://orcid.org/0000-0001-6353-8674; http://lattes.cnpq.br/2563070123322776
The integration of music and educational robotics remains underexplored, even though both areas offer strong potential to promote active, interdisciplinary, and meaningful learning. This work investigates how musicalization can be incorporated into robotics to enhance student engagement and support the practical understanding of musical, elec tronic, and computational concepts. To this end, the study proposes EcoMusic, an ecolog ical musical device built from electronic components and recycled materials, programmed to play melodies through interactive buttons. The research is grounded in the competen cies of Brazil’s National Common Curriculum (BNCC) and in theoretical foundations from music education, robotics, sustainability, and creative learning. The study was con ducted with prior authorization from the Research Ethics Committee of the Federal Uni versity of Rio Grande do Norte (UFRN), ensuring compliance with ethical standards for research involving human participants. EcoMusic was implemented in a public school with 25 students aged 11 to 17, over two instructional sessions that introduced concepts of sound and frequency, programming, device assembly, and musical experimentation. Data collection involved questionnaires administered to both students and the classroom teacher. Results indicate that integrating musicalization into robotics increases motiva tion, stimulates curiosity, facilitates understanding of musical concepts, and strengthens skills such as collaboration, creativity, and logical thinking. The use of recycled ma terials also encouraged reflections on sustainability and conscious consumption. The findings demonstrate that EcoMusic is an accessible, interdisciplinary, and curriculum aligned pedagogical tool that expands opportunities for active experiences in music and technology within basic education.
Uma metodologia baseada em agentes com avaliação autônoma para extração de conhecimento em textos técnicos
(Universidade Federal do Rio Grande do Norte, 2025-12-19) Andrade, Matheus Gomes Diniz; Silva, Ivanovitch Medeiros Dantas da; https://orcid.org/0000-0002-0116-6489; http://lattes.cnpq.br/3608440944832201; https://orcid.org/0009-0002-0268-2247; http://lattes.cnpq.br/7308297435660284; Brandão, Dennis; https://orcid.org/0000-0003-1558-0581; http://lattes.cnpq.br/6838931677289559; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; Silva, Marianne Batista Diniz da; https://orcid.org/0000-0002-8277-7571; http://lattes.cnpq.br/6470261020797104
In the current scenario of Operational Technology (OT) and Information Technology (IT) convergence and Industry 5.0, the maintenance and resilience of OT systems depend on knowledge continuity. The difficulty and inefficiency in accessing the dispersed technical knowledge within unstructured documentation of legacy systems, such as the PROFIBUS (Process Field Bus) protocol, critically impact maintenance and decision-making. Therefore, this work investigates the use of Generative Artificial Intelligence architectures to retrieve and synthesize this legacy industrial knowledge. The study compares the performance of three distinct configurations: (i) a standalone LLM (Baseline), (ii) a Single-Agent RAG (Retrieval-Augmented Generation) model, and (iii) a Multi-Agent RAG architecture. Through an experiment based on PROFIBUS technical texts, performance was evaluated using quantitative metrics (ROUGE, BERTScore) and qualitative assessment (LLM-as-a-Judge). The results demonstrate that Multi-Agent orchestration establishes a superior performance hierarchy. The collaboration and specialization among agents significantly increase the contextual precision and factual accuracy of the responses, while simultaneously reducing hallucination to 80% faithfulness to the context. The study validates the Multi-Agent RAG as a necessary architectural pattern to guarantee reliability in the transfer of engineering knowledge, providing a pathway for human-centric automation in Industry 5.0.
Trilhas para auditoria aplicada ao sistema único de saúde no contexto das órteses, próteses e materiais especiais (OPMES): estruturação de uma base de dados integrada a partir do projeto fiscalizasus
(Universidade Federal do Rio Grande do Norte, 2025-06-26) Alves, Luca Pareja Credidio Freire; Valentim, Ricardo Alexsandro de Medeiros; https://orcid.org/0000-0002-9216-8593; http://lattes.cnpq.br/3181772060208133; http://lattes.cnpq.br/8782837208577002; Valentim, Janaína Luana Rodrigues da Silva; https://orcid.org/0000-0001-8525-8155; http://lattes.cnpq.br/8236259645905686; Santos, João Paulo Queiroz dos; https://orcid.org/0000-0002-9130-7723; http://lattes.cnpq.br/2413250851590746; Coutinho, Karilany Dantas; https://orcid.org/0000-0002-2051-8611; http://lattes.cnpq.br/8409211766785367; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; Rodrigues Júnior, Methanias Colaço; Costa, Theo Duarte da; https://orcid.org/0000-0002-9355-8382; http://lattes.cnpq.br/8305343735444335
Background: The Unified Health System (SUS) was a historic milestone established by the 1988 federal constitution. However, over the years the SUS has undergone several transformations, especially concerning the amount of data generated about its internal processes. The National Audit Department of SUS (DenaSUS) was created to manage the SUS's internal control and audit processes. As a result, new demands regarding auditing in the SUS have emerged. One of these activities was organized in partnership between DenaSUS, the Federal University of Rio Grande do Norte (UFRN), the Federal Institute of Rio Grande do Norte (IFRN), the Federal Prosecution Office of Rio Grande do Norte (MPF/RN), the Federal Court of Auditors (TCU) and other competent authorities to build the FiscalizaSUS project, which aims to provide intelligent methods for analyzing large volumes of data in the field of health, with the ultimate goal of delivering audit trails focused on Regulation and Special Orthotics, Prosthetics and Special Materials (OPSM) in the SUS domain, as these are inputs that have already been identified as objects of interest in corruption scandals involving health professionals, as was the case with the “máfia das próteses”.Goal : Thus, the objective of this work is to demonstrate how the application of statistical methods and Big Data can enhance the feasibility for developing OPSM audit trails through the integration of public and private databases. Methods: On the first hand, a systematic mapping of the literature was conducted in order to evaluate which methods are being applied for Big Data-based health audit research. From this point of view, it was possible to analyse and adopt approaches to develop a relational model for OPSM audit trials, and also develop metrics based on statistical analysis of the collected data samples. Results : The Systematic Literature Mapping (SLM) initially included a total of 344 articles, obtained from the search string used. This left 40 articles for evaluation and data collection, after the inclusion and exclusion steps. Heterogeneous solutions were identified to address the issue of health auditing, presenting techniques also related to computational methods and Big Data approaches.From an operational perspective, the tables were organized oriented towards a star schema, containing fact and dimension tables, which enable the integration of different data from various sources, also concerning the goals of each audit trail and assessing the presence of common values between the databases. Nevertheless, the integration between the databases also provided a visualization layer by using Apache Superset’s virtual interface, allowing a comprehensive view of the anomalies through geographical location, links between the data sources, and also through the proposed metrics. Conclusion: The results allow us to conclude that the integration of different data sources related to OPSM can contribute to the identification of irregularities in transactional processes involving this type of medical supply. In addition, the Big Data architecture provided by the FiscalizaSUS project also made it possible to aggregate these different data sets and brought new perspectives to the health audit trails within the scope of the project.
Detecção de processos erosivos com aprendizado de máquina e a aplicação da Equação Universal de Perda de Solo (USLE)
(Universidade Federal do Rio Grande do Norte, 2025-08-18) Marinho, Wesley José dos Santos; Alsina, Pablo Javier; http://lattes.cnpq.br/3653597363789712; http://lattes.cnpq.br/9223986328697943; Martins, Allan de Medeiros; https://orcid.org/0000-0002-9486-4509; http://lattes.cnpq.br/4402694969508077; Matos, Maria de Fátima Alves de; https://orcid.org/0000-0002-2864-2027; http://lattes.cnpq.br/7230788237148341; Laura, Tânia Luna; http://lattes.cnpq.br/8142774545918274
The research in question investigates the detection of erosion processes through machine learning and the application of the Universal Soil Loss Equation (USLE). The primary issue addressed by the study is the identification and prevention of soil erosion, a significant environmental problem. Key objectives include implementing techniques to early detect erosion processes near power transmission lines, using the USLE as an additional validation factor. The motivation behind the study is to mitigate damage to power distribution networks caused by erosion and to seek effective methods for monitoring and preventing erosion. The methodology involves using machine learning algorithms, specifically Convolutional Neural Networks (CNNs), to analyze satellite images and identify areas prone to erosion. Additionally, the USLE is applied as a tool for calculating soil loss, aiding in the assessment of the likelihood of erosion in specific locations. The main contributions of the study include integrating machine learning technologies in the detection of erosion processes and validating these techniques with the USLE, demonstrating its effectiveness in this context. The results indicate the feasibility and accuracy of using machine learning for detecting erosion processes, as well as the importance of the USLE as an evaluation method. The study’s conclusions highlight the relevance of the practical application of these techniques for the preservation of power transmission lines and more efficient monitoring of their surroundings. Future research directions include expanding the study to other regions and improving the detection methodology.
Projeto de sensores de micro-ondas para predição e classificação de contaminantes críticos em água usando o método baseado em Decision-Tree Support Vector Machine
(Universidade Federal do Rio Grande do Norte, 2025-10-09) Sousa, Bruno Matias de; Silva Neto, Valdemir Praxedes da; https://orcid.org/0000-0003-3867-3297; http://lattes.cnpq.br/4160231554601828; https://orcid.org/0009-0003-9607-1099; http://lattes.cnpq.br/2574712778132756; D'Assunção, Adaildo Gomes; http://lattes.cnpq.br/4159638862269940; Vasconcelos, Cristhianne de Fátima Linhares de; http://lattes.cnpq.br/3140725798737102; Duarte Júnior, José Garibaldi; https://orcid.org/0000-0002-9708-3118; http://lattes.cnpq.br/4332534673634063
Estratégia aprimorada para avaliações da exposição a campos eletromagnéticos de radiofrequência em prédios próximos à infraestrutura de telecomunicações
(Universidade Federal do Rio Grande do Norte, 2025-07-25) Silva, Ricardo Queiroz de Farias Henrique; Sousa Júnior, Vicente Angelo de; https://orcid.org/0000-0003-2859-6136; http://lattes.cnpq.br/6358312955522220; https://orcid.org/0000-0003-0861-4341; http://lattes.cnpq.br/8490357509264057; Souza Filho, Agostinho Linhares de; http://lattes.cnpq.br/8161570082771243; Campos, Antonio Luiz Pereira de Siqueira; http://lattes.cnpq.br/1982228057731254; Rodrigues, Márcio Eduardo da Costa; http://lattes.cnpq.br/2116993138825543
This study proposes to complement the current regulatory agencies’ methodology for evaluating exposure to Radiofrequency Electromagnetic Fields (RF-EMF) in buildings under the direct incidence of emissions from Base Station (BS) antennas. The key contribution is the refinement of measurement point selection within buildings, ensuring that assessments more accurately capture exposure levels. The approach employs technical criteria for selecting target buildings, considering the location of the BSs and the configuration of the surrounding antenna structures. The proposed approach was applied to measurements in five buildings in Natal, the capital of Rio Grande do Norte, located in the Northeastern region of Brazil. The results show electric field intensity peaks up to 30.53 times higher and averages up to 16.51 times higher than values obtained from measurements conducted at ground level, the position where the evaluations are typically conducted. The highest exposure rates reached 39.19% and 20.44% of the limits established by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) for the frequency bands used in broadcasting services and mobile telephony, respectively. The proposed approach can improve and complement the normative guidelines for RF-EMF exposure assessment, providing more representative evaluations of exposure in indoor building environments near modern telecommunication infrastructure.
Solução para monitoramento adaptativo de métricas de Redes Open RAN
(Universidade Federal do Rio Grande do Norte, 2025-08-07) Dória, Matheus Fagundes de Souza; Sousa Júnior, Vicente Angelo de; Martins, Allan de Medeiros; https://orcid.org/0000-0002-9486-4509; http://lattes.cnpq.br/4402694969508077; https://orcid.org/0000-0003-2859-6136; http://lattes.cnpq.br/6358312955522220; http://lattes.cnpq.br/3322534105634063; Luna, Daniel Rodrigues de; http://lattes.cnpq.br/5084446763958238; Melo, Yuri Victor Lima de; https://orcid.org/0000-0003-3026-7059; http://lattes.cnpq.br/3613517930531317
Monitoring in Open RAN networks involves a continuous exchange of metrics between xApps and the radio access network through the E2 interface. Due to the intense signaling exchange, this monitoring consumes a significant amount of resources (both hardware and network) and can negatively impact system efficiency. This work proposes the development of an xApp for adaptive metric monitoring, adjusting the sampling frequency of Key Performance Indicators (KPIs) based on real-time analysis. The proposal is based on the definition of a Risk Analysis and aims to reduce excessive resource consumption without compromising monitoring quality, contributing to more efficient and cost-effective Open RAN network management. The xApp is implemented using an open-source software-based infrastructure, with FlexRIC as the development platform. For the implementation of the 5G core network and RAN, OpenAirInterface was used, integrated with a Software Defined Radio (SDR) platform, enabling the use of the RAN with Commercial-off-the-shelf (COTS) user equipment. The results are promising, indicating that adaptive metric monitoring offers potential energy savings compared to fixed-frequency monitoring. Furthermore, the energy gain provided by the proposed monitoring becomes even more relevant when applied to a large-scale context, in which the overhead generated by metric collection via the E2 interface poses a significant challenge to system efficiency.
Desenvolvimento de sensor de umidade de solo baseado em antena patch de microfita com capacitor interdigital
(Universidade Federal do Rio Grande do Norte, 2025-05-13) Silva Júnior, Isaú de Sousa; Campos, Antonio Luiz Pereira de Siqueira; http://lattes.cnpq.br/1982228057731254; https://orcid.org/0000-0002-1161-2251; http://lattes.cnpq.br/0089946777137372; Gomes Neto, Alfredo; https://orcid.org/0000-0001-5437-9093; http://lattes.cnpq.br/1403715441701958; Silva, Isaac Barros Tavares da; https://orcid.org/0000-0002-2351-291X; http://lattes.cnpq.br/7304355962395872
Soil water content and water availability are critically important for terrestrial activities, especially those involving agriculture, forestry, hydrology, and engineering. For this reason, many researchers have dedicated efforts to proposing practical and simple methods to obtain this information using diverse measurement techniques. Among existing techniques, those based on dielectric properties, such as time-domain reflectometry (TDR) and frequency-domain reflectometry (FDR), stand out as the most widely used for in situ soil moisture measurements. However, the high cost of such sensors limits their large-scale use. Thus, planar microwave sensors have gained significant attention as an alternative for soil moisture sensing in the field, due to their inherent characteristics, such as low production cost, ease of handling, lightweight design, compact size, and fast measurement speed, making them highly suitable for applications requiring continuous and distributed sensing in situ. In this context, this work presents the design of a low-cost planar microwave sensor for soil moisture monitoring, integrating an interdigital capacitor (IDC) with a rectangular microstrip patch antenna. The device was developed using an FR-4 dielectric substrate (εr = 4.4 and tan(δ) = 0.02) and operates at 1.5 GHz when unloaded. For experimental validation, the sensor was tested on Oxisol soil samples with moisture levels ranging from 0% to 14%, demonstrating a mean error of 3.56%, sensitivity of 1.10%, and frequency resolution of 15.17 MHz, parameters comparable to those reported in the literature for sensors of the same class. This shows that the proposed sensor is viable for applications requiring continuous and distributed monitoring, such as precision agriculture and water resource management.
Sistema ultrassônico aplicado à detecção de interfaces geológicas em caixa de areia
(Universidade Federal do Rio Grande do Norte, 2025-08-11) Figueiredo, Gabriel Araújo Almeida; Salazar, Andrés Ortiz; https://orcid.org/0000-0001-5650-3668; http://lattes.cnpq.br/7865065553087432; http://lattes.cnpq.br/3121557302219300; Silva, Carlos César Nascimento da; http://lattes.cnpq.br/5136096872133075; Fonseca, Diego Antonio de Moura; http://lattes.cnpq.br/5215142664564230; Pinheiro Filho, Ricardo Ferreira; http://lattes.cnpq.br/4758146782593557
Geological modeling in sandboxes is an effective tool for replicating and investigating the propagation of seismic reflections in a controlled environment, providing results that are more representative compared to purely computational simulations. This technique enables a detailed study of phenomena such as reflection, refraction, and wave dispersion at interfaces composed of different materials. However, the implementation of these experiments faces challenges, especially regarding the high cost of suitable ultrasonic sensors, which must have a wide beam angle and broad frequency bandwidth to detect a variety of signals reflected by different materials. Furthermore, most systems described in the literature are composed of decentralized, dedicated devices for each function — requiring computers for control and processing, digital-to-analog converters for signal emission, and high- resolution analog-to-digital converters for data acquisition, which significantly increases the complexity and cost of the system. In this context, the present work proposes the use of more affordable commercial ultrasonic sensors, even with more limited characteristics, combined with a digital signal processor (DSP) responsible for integrating pulse generation, data acquisition, and transmission. Additionally, techniques for selecting and arranging materials within the sandbox will be employed to maximize interface detection. This approach seeks to balance cost and performance, making it possible to conduct experiments with consistent results even in resource-constrained scenarios.
Modelagem computacional da capacidade produtiva e do financiamento da rede assistencial do sus: uma abordagem com foco em inteligência computacional e orçamento secreto
(Universidade Federal do Rio Grande do Norte, 2025-06-12) Silva, Gleyson José Pinheiro Caldeira; Valentim, Ricardo Alexsandro de Medeiros; https://orcid.org/0000-0002-9216-8593; http://lattes.cnpq.br/3181772060208133; http://lattes.cnpq.br/4398013372084295; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; Costa, Theo Duarte da; https://orcid.org/0000-0002-9355-8382; http://lattes.cnpq.br/8305343735444335; Valentim, Janaína Luana Rodrigues da Silva; https://orcid.org/0000-0001-8525-8155; http://lattes.cnpq.br/8236259645905686; Santos, João Paulo Queiroz dos; https://orcid.org/0000-0002-9130-7723; http://lattes.cnpq.br/2413250851590746
This dissertation presents an analysis of the healthcare production of the Unified Health System (SUS) in Brazil, using technical approaches from Computer Engineering to model the relationship between service production and funding mechanisms. Outpatient and hospital production are analyzed, as well as intergovernmental financial transfers, available infrastructure, and human resources. The methodology applied is based on information systems and computational intelligence techniques to integrate and analyze the different databases of the Ministry of Health, such as SIGTAP, CNES, and FNS, ensuring the traceability and auditability of the data. In addition, algorithms were used to detect anomalies in the production presented and in the approval of resources, highlighting possible inconsistencies and fraud in the use of public resources. The results reveal significant discrepancies between the declared healthcare production and the resources perceived, suggesting the need to improve control mechanisms and the interoperability of SUS information systems. The recommendations proposed include improvements in the technical aspects of health systems, aiming to increase management effectiveness and transparency of processes.
Metodologia orientada a agentes de linguagem para assistência automotiva: integrando engenharia de Prompts em Chatbots avançados
(Universidade Federal do Rio Grande do Norte, 2025-07-11) Medeiros, Thaís de Araújo de; Silva, Ivanovitch Medeiros Dantas da; https://orcid.org/0000-0002-0116-6489; http://lattes.cnpq.br/3608440944832201; https://orcid.org/0000-0002-6447-3806; http://lattes.cnpq.br/4746141228253120; Viegas, Carlos Manuel Dias; https://orcid.org/0000-0001-5061-7242; http://lattes.cnpq.br/3134700668982522; Silva, Marianne Batista Diniz da; https://orcid.org/0000-0002-8277-7571; http://lattes.cnpq.br/6470261020797104; Barros, Thiago Medeiros; https://orcid.org/0000-0001-5356-3550; http://lattes.cnpq.br/3844440611390386
The increasing presence of digital systems in vehicles has expanded the number of functionalities available to users. However, recurring doubts related to the use of these features, the interpretation of alerts, and the execution of basic operational procedures still require consulting the owner’s manuals to ensure proper vehicle handling. Although the digitization of these documents represents an advancement over physical formats, access to information remains limited, especially in situations that demand quick responses and accessible language. In this context, this work proposes a language agent–oriented approach, grounded in the Retrieval-Augmented Generation (RAG) technique, with the goal of facilitating specialized consultation of technical content in automotive manuals. The methodology encompasses stages such as segmenting texts into coherent fragments, indexing them in a vector database, crafting prompts tailored to a Large Language Model, and conducting a comparative evaluation of six RAG variants (Conventional, with Gradient Descent, Multi-Query, Step-Back, Self-RAG, and Self-RAG with Gradient Descent). Accordingly, an experiment was conducted in which each variant was evaluated using the LLM-as-a-judge strategy, in which an LLM assigned scores for contextual faithfulness, question relevance, informational completeness, and safety verification. The dataset used comprised twenty questions, ten based on the Volkswagen Polo 2025 manual and ten on the Fiat Argo 2023 manual. Additionally, semantic similarity between pairs of responses was measured using BERTScore. The results indicated that Step-Back achieved the highest overall average score and led in completeness and safety, whereas Self-RAG delivered the best performance in faithfulness and exhibited high semantic convergence with its gradient-based variant. These findings suggest that mechanisms of reformulation, decomposition, and self-assessment improve both the quality and consistency of responses, highlighting the potential of adaptive architectures to enhance technical assistance in embedded systems.
MMCloud: um algoritmo de clustering não supervisionado online para análise do comportamento do motorista
(Universidade Federal do Rio Grande do Norte, 2025-07-10) Medeiros, Morsinaldo de Azevedo; Silva, Ivanovitch Medeiros Dantas da; Silva, Marianne Batista Diniz da; https://orcid.org/0000-0002-8277-7571; http://lattes.cnpq.br/6470261020797104; https://orcid.org/0000-0002-0116-6489; http://lattes.cnpq.br/3608440944832201; https://orcid.org/0000-0001-7624-5301; http://lattes.cnpq.br/2910614991697297; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; Santos, Max Mauro Dias; http://lattes.cnpq.br/6212006974231025
Evolving and adaptive intelligent systems are essential for handling dynamic data streams from real-world environments, especially in resource-constrained settings such as vehicles enabled with the Internet of Things (IoT). This study proposes an online evolutionary clustering approach for real-time driver behavior diagnosis, leveraging Tiny Machine Learning (TinyML) and soft sensors. It introduces the MMCloud algorithm, an incremental clustering model designed to manage continuous data streams. The algorithm eliminates the need for retraining and is capable of adapting to concept drift and varying driving conditions. The methodology integrates onboard diagnostics (OBD-II) with an edge computing framework to classify driver behavior into three categories: cautious, normal, and aggressive. To validate the proposed approach, two case studies were conducted in urban environments under various traffic conditions. One used a Freematics One+ device, while the other deployed the embedded algorithm in a mobile application. The results demonstrate the system’s ability to accurately identify evolving driving patterns, contributing to safer driving practices, fuel consumption optimization, and intelligent transportation systems (ITS). This research advances the field of evolutionary AI by integrating adaptive machine learning models with embedded IoT solutions, enabling autonomous and self-organizing monitoring of driver behavior.
Gerência de largura de banda do reúso fracionário de frequência em cenários dinâmicos com Hotspots utilizando aprendizado por reforço
(Universidade Federal do Rio Grande do Norte, 2025-07-04) Silva, Eriberto de Souto; Sousa Júnior, Vicente Angelo de; https://orcid.org/0000-0003-2859-6136; http://lattes.cnpq.br/6358312955522220; http://lattes.cnpq.br/9594814897805733; Silva Neto, Valdemir Praxedes da; https://orcid.org/0000-0003-3867-3297; http://lattes.cnpq.br/4160231554601828; Melo, Yuri Victor Lima de; https://orcid.org/0000-0003-3026-7059; http://lattes.cnpq.br/3613517930531317
The present work approaches the use of Inter-Cell Interference Coordination (ICIC) techniques, based on techniques Fractional Frequency Reuse (FFR) aimed at mitigating one of the main challenges faced when designing mobile networks: co-channel interference (CCI). ICIC is one of the primary factors contributing to the degradation of mobile communication system performance, especially in dynamic scenarios with randomly distributed user clusters, known as hotspots. Initially, the problem is characterized by the exploration of the chosen ICIC technique without automated parameterization, as well as the system’s performance in the aforementioned scenario. Then, considering the dynamic nature of the environment, the proposed solution involves the use of machine learning techniques—more specifically, the Multi-Armed Bandit (MAB)—customized for optimizing bandwidth management in ICIC techniques for 3GPP systems. The objective of the proposal is to adapt the spatial allocation of bandwidth, using system load variation and hotspot dynamics as decision-making parameters. Proof-of-concept experiments were conducted using the ns-3 network simulator. The results demonstrate that the MAB-based approach proposed in this dissertation outperforms static bandwidth distribution configurations in terms of throughput, even under varying system load conditions, showing significant gains.
Industry 4.0-Compliant Artificial Intelligence-based power transformer fault classification method during data missing conditions
(Universidade Federal do Rio Grande do Norte, 2025-03-28) Dantas, Ingrid Thaís Azevêdo; Costa, Flávio Bezerra; Medeiros, Rodrigo Prado de; https://orcid.org/0000-0002-7554-1480; http://lattes.cnpq.br/7510091283933216; http://lattes.cnpq.br/6791400579819527; Silveira, Luiz Felipe de Queiroz; Oleskovicz, Mário
Power transformers are fundamental components in electrical systems, responsible for the efficient transfer of energy between different voltage levels. Despite their robustness, these devices are subject to failures over time, such as internal electrical faults that can compromise not only the transformer itself but also the stability of the entire interconnected system. In this context, the use of intelligent solutions that enable continuous monitoring and accurate fault classification becomes increasingly relevant, contributing to faster diagnostics and, thus, reducing equipment downtime. This work proposes an innovative approach for classifying internal faults in power transformers, with a particular focus on those occurring in the bushings. The proposed method is based on the development of a classification system that combines advanced mathematical techniques with supervised machine learning models. A key feature of the approach is its ability to operate effectively even with incomplete data, a common condition in real industrial environments where communication failures, sensor defects, or signal noise can compromise measurement integrity and, consequently, the diagnostic process. The methodology integrates the Real-Time Bounded Stationary Wavelet Transform (RT-BSWT) with machine learning models. After applying this transform to the current signals, energy values are extracted from the coefficients and used as the input features for the classification stage. Three supervised algorithms were implemented: Decision Tree, Random Forest, and Logistic Regression. These models were selected for their strong performance in classification tasks and for their interpretability. The system was evaluated using a simulated dataset containing different types of internal faults in transformer bushings, including single-phase, two-phase, three-phase, and phase-to-ground faults, applied to both the primary and secondary sides. The tests also considered variations in fault resistance, fault angle, load conditions, noise, and, most importantly, the absence of one phase current, aiming to simulate realistic operational scenarios. The results demonstrated that the proposed approach is highly effective, achieving accuracy rates above 95% under ideal conditions. Among the tested models, the Random Forest algorithm performed the best. The effectiveness of the method was also validated through a direct comparison with a traditional wavelet-only approach, without machine learning. While the conventional method showed reasonable performance in ideal conditions, with up to 85% accuracy, its effectiveness was severely affected by data loss and noise, dropping to 36.9% accuracy in adverse scenarios. It also struggled to identify specific fault types, such as secondary-side two-phase-to-ground faults or three-phase faults under missing data. In contrast, the machine learning-based method showed higher accuracy and adaptability, correctly classifying all 20 fault types evaluated, even under signal loss and different noise levels. Beyond its technical contribution, this work aligns with the principles of Industry 4.0 by integrating intelligent data analysis, real-time responsiveness, and robustness against data loss. These characteristics significantly expand the potential for applying the proposed methodology in modern industrial environments. The system can be integrated into existing protection and monitoring architectures, providing direct support for operational decision-making. In summary, this work presents a significant contribution to the field of diagnosis and classification of internal faults in power transformers. Its ability to perform under adverse conditions and maintain high accuracy even with incomplete data makes it a promising solution for both academic research and practical applications in the electric power sector. This approach may serve as a foundation for the development of smarter protection systems, contributing to increased safety and operational efficiency.
Predição de aplicação de doses de vacinas com N-BEATS: uma solução de saúde digital para a gestão de imunobiológicos no SUS
(Universidade Federal do Rio Grande do Norte, 2025-02-24) Lemos, Lemyson Oliveira; Oliveira, Luiz Affonso Henderson Guedes de; https://orcid.org/0000-0003-2690-1563; http://lattes.cnpq.br/7987212907837941; http://lattes.cnpq.br/0698665477101903; Santos, João Paulo Queiroz dos; Valentim, Ricardo Alexsandro de Medeiros
Vaccine demand forecasting is essential for the efficient management of stock and the proper distribution of immunobiologicals, ensuring vaccination coverage and preventing waste. Time series models are widely used in this context, allowing the anticipation of vaccine demand and the optimization of available resources. With advancements in deep learning, the N-BEATS (Neural Basis Expansion Analysis Time Series) algorithm has stood out for its effectiveness in modeling complex data without requiring extensive preprocessing. Introduced by Oreshkin et al. (2020), N-BEATS outperforms traditional models, such as autoregressive models and Multi-Layer Perceptrons (MLPs), by handling nonlinear patterns through a block-stacking architecture that iteratively refines predictions. This study applies NBEATS to forecast the weekly number of vaccines to be administered in the states of Rio Grande do Norte and Espírito Santo, using data from the RN+VACINA and Vacina e Confia systems. In Rio Grande do Norte, the forecast focuses on a macro analysis covering state and municipal levels, while in Espírito Santo, a more granular approach is taken, with vaccinespecific predictions. The results highlighted the superior performance of N-BEATS compared to traditional models such as XGBoost. In Rio Grande do Norte, the model achieved an R² of 0.81 and a MAPE of 16.59%, while XGBoost obtained an R² of 0.73. In Natal, a municipality in the state, the values were R² of 0.77 and MAPE of 21.59%. In Espírito Santo, the analysis was conducted at the state level and for the municipality of Cariacica, focusing on the BCG vaccine. The statelevel results showed a MAPE of 1.00% and R² of 0.83, while at the municipal level, the values were MAPE of 2.46% and R² of 0.85. These metrics demonstrate that N-BEATS is a robust and efficient solution for time series forecasting in vaccination systems. Its ability to handle nonlinear and complex patterns, combined with minimal preprocessing requirements, makes it suitable for scenarios where speed and accuracy are essential. Additionally, its superior performance compared to traditional models reinforces the potential of N-BEATS as a reliable tool for managing vaccination campaigns, contributing to a more efficient and strategic distribution of immunizers.
A Real-Time meta-heuristic-based safe navigation approach for mobile robots in unknown environments
(Universidade Federal do Rio Grande do Norte, 2025-02-19) Balza, Micael; Fernandes, Marcelo Augusto Costa; https://orcid.org/0000-0001-7536-2506; http://lattes.cnpq.br/3475337353676349; Silva, Sérgio Natan; Pedrosa, Diogo Pinheiro Fernandes; Oliveira, Fábio Fonseca de
Metodologia e plataforma baseadas em aprendizado de máquina para inspeção de defeitos na indústria têxtil
(Universidade Federal do Rio Grande do Norte, 2025-01-27) Góes, Angelo Leite Medeiros de; Dória Neto, Adrião Duarte; https://orcid.org/0000-0002-5445-7327; http://lattes.cnpq.br/1987295209521433; https://orcid.org/0000-0003-1138-1403; http://lattes.cnpq.br/3037310998903708; Fontes, Aluisio Igor Rego; Florêncio, Heitor Medeiros; Barroca Filho, Itamir de Morais
Quality inspection (QI) in the textile industry is an essential yet challenging process, particularly due to the diversity of defects and reliance on subjective criteria from human inspectors. This work proposes a methodology based on computer vision and deep learning to optimize the QI process in textiles, alongside the development of an associated platform to demonstrate its feasibility and applicability in industrial environments. The platform integrates the proposed algorithms: grid-based detection, which segments images into patches for local defect classification, and ripple refinement, which improves detection quality by analyzing neighboring cells. The experimental results validate the effectiveness of the methodology. A benchmark of machine learning models was conducted, comparing eight approaches, including YOLOv5, YOLOv8 variants, and popular networks in the literature, using the TILDA 400 dataset, composed of five distinct defect classes. The YOLOv8 models stood out, especially YOLOv8 medium, which achieved 90.35% accuracy with an inference time of 0.5ms on a Tesla P100 GPU, surpassing the average precision of 70% from human inspectors. The YOLOv8 small also demonstrated remarkable performance, with a theoretical inspection capacity of up to 46.875 m/min and 86.96% accuracy, exceeding the manual maximum speed of 15 to 20 m/min. The application of this methodology, combined with the developed platform, demonstrates potential to enhance accuracy, speed, and organization in QI, while also reducing operational costs and promoting automation and efficiency in the textile industry.
Blockchain no ecossistema tecnológico da Plataforma revELA: no contexto da esclerose lateral amiotrófica (ELA)
(Universidade Federal do Rio Grande do Norte, 2025-01-13) Fonseca, Aleika Lwiza Alves; Valentim, Ricardo Alexsandro de Medeiros; https://orcid.org/0000-0002-9216-8593; http://lattes.cnpq.br/3181772060208133; http://lattes.cnpq.br/2621599224457086; Morais, Antonio Higor Freire de; Nagem, Danilo Alves Pinto; Barbalho, Ingridy Marina Pierre; Valentim, Janaína Luana Rodrigues da Silva; Oliveira, Luiz Affonso Henderson Guedes de
The National ALS Registry is a digital health solution that aims to collect epidemiological data from various patients with Amyotrophic Lateral Sclerosis (ALS) in Brazil. When using technological solutions such as this one, for electronic health data recording, concerns arise regarding security, data integrity, and the importance of an interoperable platform for sharing with other systems. To address these concerns, blockchain technology has been attracting a lot of attention in several areas that already see the possibilities for progress with its incorporation. Considering the needs for security, auditability, and interoperability, and to make the National ALS Registry increasingly secure and reliable, this paper presents the configuration of a blockchain network with the Hyperledger Fabric framework and the implementation of an API, developed with Node.js, used to integrate the blockchain with the National ALS Registry. To evaluate the performance of the proposed solution, the Hyperledger Caliper tool was used, which is a reference for benchmarking blockchain solutions. The results show the viability of using this technology and the relevance of its use is highlighted throughout the text. Although the performance analysis was not satisfactory, with average latency above 19s and throughput values below 100 transactions per second, the data obtained allows us to identify points for improvement in the implementation and optimization of the network.

SIGAA

Browse

collection.page.browse.recent.head