Navegando por Data de Publicação, começando com "2026-04-24"

Filtrar resultados por ano ou mês

Agora exibindo 1 - 7 de 7

Estruturas híbridas de poliamida 6-madeira (eucalyptus grandis) sobremoldadas por injeção
(Universidade Federal de São Carlos, 2026-04-24) Zanchetta, Giovana Zanetini; Canto, Leonardo Bresciani; https://lattes.cnpq.br/7287108960864123; https://orcid.org/0000-0002-4138-005X; https://lattes.cnpq.br/4463643206121785; https://orcid.org/0009-0004-9368-7274; Bettini, Silvia Helena Prado; Carvalho, Antonio José Felix de; https://lattes.cnpq.br/3687551763124327; https://lattes.cnpq.br/5050955206618507; https://orcid.org/0000-0001-6549-2980; https://orcid.org/0000-0001-8403-1135
Polymer–wood hybrid (PWH) structures represent a promising alternative for sustainable structural applications. This study introduces the development of hybrid joints composed of polyamide 6 (PA6) and Eucalyptus grandis wood produced by injection overmolding, without the use of adhesives or chemical surface treatments. Two scarf joint geometries with single and double chamfers at 45° were designed and evaluated in terms of mechanical performance and interfacial adhesion mechanisms. Shear strength was determined through quasi-static mechanical tests, while the interface was characterized by scanning electron microscopy (SEM) and X‑ray microtomography (µCT). Microstructural analyses revealed a well‑defined interface, with PA6 penetrating wood vessels and forming polymer protrusions (~50 µm) that promote mechanical interlocking. Mechanical tests showed good repeatability, with hybrid joints reaching shear strength values of up to 8.1 ± 0.6 MPa, comparable to those reported in the literature for polymer–wood hybrid lap joints. Complementary digital image correlation (DIC) analyses indicated efficient mechanical coupling between the components, with a more homogeneous strain distribution for the single‑chamfer configuration, suggesting more effective stress transfer under mechanical loading.
Estratégias de acesso e mediação para repositório de informação tecnológica integrada
(Universidade Federal de São Carlos, 2026-04-24) Erba, Ana Guelfi; Zaniro, Dênis Leonardo; https://lattes.cnpq.br/0800017967297543; https://orcid.org/0000-0003-2638-9264; Quoniam, Luc Marie; https://lattes.cnpq.br/4754764003480925; https://orcid.org/0000-0002-6333-6594; https://lattes.cnpq.br/5569779260528830; https://orcid.org/0009-0004-0539-7876; Quoniam, Luc Marie; Hoffmann, Wanda Aparecida Machado; Santos, Fábio José Justo dos; https://lattes.cnpq.br/4754764003480925; https://lattes.cnpq.br/7609135667093837; https://lattes.cnpq.br/2020262358450007
This work presents the development of an integrated repository of scientific and technological effects, with the objective of mitigating the information transfer barrier identified in an originally static knowledge base. The research, of an applied and empirical nature, was conducted to enable the conversion of a vast stock of inventive problem-solving solutions into an accessible and usable knowledge flow. Initially, a diagnostic evaluation of the spreadsheet database was carried out, identifying critical gaps in interoperability and accessibility that limited the strategic use of the data, as advocated by international guidelines for open data. The methodological path involved the logical structuring of the data, adopting a specific data type to favor the coherent representation of the content in multiple languages. Subsequently, a mediation artifact was developed—a Web system structured based on Usability principles, aiming to optimize the efficiency of human-computer interaction and information retrieval. To complement the exact relational search methodologies, the system's architecture incorporated Artificial Intelligence features through the Retrieval-Augmented Generation (RAG) technique. The implemented technological solution met the proposed objectives by establishing multiple specialized access strategies. The system enables searches from different perspectives, connecting the abstract knowledge of inventive problem-solving solutions to industrial and patent classifications, which adds strategic value to technological information. The final adherence evaluation of the artifact evidenced the alignment of the architecture with the pillars of findability, accessibility, interoperability, and reusability. It is concluded that the work provides an empirical mediation model that reiterates the relevance of the Scientific Information area in the organization and architecture of knowledge access for socioeconomic development. The solution not only contributes to broadening access to the resource but also enhances its utilization as a strategic and reusable asset.
Mapeamento de aplicações das ferramentas digitais da indústria 4.0 na cadeia de suprimentos em empresas de máquinas agrícolas
(Universidade Federal de São Carlos, 2026-04-24) Rosa, Jorge Luiz Amaral; Voltarelli, Murilo Aparecido ; https://lattes.cnpq.br/4889939917847330; https://lattes.cnpq.br/5153087261053903; https://orcid.org/0000-0003-1662-5200; Gazzola, Jonathan; Cavichioli, Fabio; https://lattes.cnpq.br/0159433721761670; https://lattes.cnpq.br/0333759662034919
The agricultural machinery sector is characterized by its importance to the Brazilian economy, generating approximately BRL 72 billion in net revenue in 2023. The supply chain is defined as the connected network from suppliers to the delivery of products and services to the customers. With the advent of technological innovations, there has been an evolution toward the concept of Supply Chain 4.0, an offshoot of Industry 4.0. There was a gap in academic research on the challenges of digital tools implementation in the agricultural machinery sector. Through exploratory research with both qualitative and quantitative approaches, the objective was to map which Industry 4.0 tools are applied in the supply chain in farm machinery companies, in the areas of Logistics, Planning and Procurement. A total of 18 employees of five companies of this sector, located in four Brazilian states - São Paulo, Minas Gerais, Parana and Rio Grande do Sul - were interviewed. Through combination of a systematic literature review and the formulation of open and closed questions, the answers were identified and analyzed to define the challenges and maturity of using digital technologies in Supply Chain area.
Flexural behavior of corroded steel beams with and without web openings: experimental, analytical, and numerical investigation
(Universidade Federal de São Carlos, 2026-04-24) Benedito, André Vitor; Krahl, Pablo Augusto; https://lattes.cnpq.br/6910799550680489; https://orcid.org/0000-0002-6172-5481; Martins, Carlos Humberto; https://lattes.cnpq.br/1665294496138141; https://orcid.org/0000-0001-7342-5665; https://lattes.cnpq.br/0497198237377988; https://orcid.org/0009-0005-0890-9969; Bolandim, Emerson Alexandro; Gidrão, Gustavo de Miranda Saleme; Rodrigues, Monique Cordeiro; Pituba, José Julio de Cerqueira; https://lattes.cnpq.br/9169305081866640; https://lattes.cnpq.br/5358252672517976; https://lattes.cnpq.br/9392939752668118; https://lattes.cnpq.br/6503462610408968; https://orcid.org/0000-0001-6220-7646; https://orcid.org/0000-0003-1483-0526; https://orcid.org/0000-0003-3876-4893; https://orcid.org/0000-0002-1480-6238
Corrosion is one of the most critical deterioration mechanisms affecting steel structures, since it reduces plate thickness, modifies geometric and mechanical properties, and may lead to instability, thereby compromising structural safety and load-carrying capacity. In steel beams, especially castellated beams, comprehensive studies that address the combined effects of corrosion on flexural response and local instability mechanisms remain limited, particularly regarding web-post buckling. In this context, this thesis investigates the flexural behavior of corroded steel beams with and without web openings through experimental, analytical, and numerical approaches, considering ASTM A572 Grade 50 and Grade 60 steels. The experimental program comprised tensile tests on coupons extracted from the steel profiles, three-point bending tests on solid-web and castellated beams under both corroded and uncorroded conditions, and Digital Image Correlation analyses to characterize surface imperfections, lateral displacements, and strain fields. The results showed that, in corroded solid-web beams, corrosion promoted local buckling before full cross-sectional yielding, thus changing the governing resistance mechanism. In castellated beams, corrosion significantly increased susceptibility to web-post buckling, leading to substantial reductions in load-carrying capacity of 39.05% for Grade 50 steel and 35.85% for Grade 60 steel, while the steel grade itself had no significant influence on the overall web-post buckling mode. DIC analyses further revealed localized strain concentrations around corrosion pits and earlier yielding in the tensile tie region of the web-post. Pit-depth measurements were statistically described by a log-normal distribution, with logarithmic mean and standard deviation values of 0.7951 mm and 0.3865 mm for Grade 50 steel, and 0.8918 mm and 0.4636 mm for Grade 60 steel. Based on experimental evidence, a non-empirical analytical formulation was proposed to estimate the critical load associated with web-post buckling in both corroded and uncorroded castellated beams. Compared with Steel Design Guide 31 and EN 1993-1-13:2024, the proposed model provided closer agreement with the experimental results, with lower average deviations than the code-based approaches. In parallel, a finite element model was developed in ABAQUS using a stochastic representation of corrosion based on a spatially correlated Gaussian random field. The calibration process indicated that a 15 mm mesh and a spatial correlation factor of ρf = 0.75 yielded the best agreement with the experimental load-displacement curves and successfully reproduced the corrosion-induced shift in structural behavior. Overall, the thesis advances the understanding of corrosion effects on the flexural and instability behavior of steel beams and provides analytical and numerical tools for the assessment of corroded castellated members.
Propriedade magnética de compósitos de nitreto de carbono grafítico com magnetita
(Universidade Federal de São Carlos, 2026-04-24) Rodrigues, Adélia Victória Lázaro; Oliveira, Adilson Jesus Aparecido de; https://lattes.cnpq.br/4640148190073166; https://orcid.org/0000-0003-0145-9107; https://lattes.cnpq.br/9556361772060539; https://orcid.org/0009-0001-9792-2268; Garcia, Ducinei; Varalda, José; https://lattes.cnpq.br/5236758856687873; https://lattes.cnpq.br/8525672131040509
The advent of the discovery of graphene, materials based on carbon structures have been extensively investigated due to their wide range of potential applications, low production cost, and unique electronic structure. Among these materials, graphitic carbon nitride (g-C₃N₄) stands out for its diverse morphologies, such as porous nanosheets, nanotubes, nanoflakes, and nanorods, and is frequently combined with non-toxic oxides. In particular, the incorporation of magnetic materials into g-C₃N₄ opens new perspectives for even more significant applications. While graphitic carbon nitride has been widely explored for hydrogen generation through processes such as water electrolysis and the photodegradation of environmental pollutants, magnetite has been extensively studied for its magnetic properties and its potential applications in areas such as energy storage and water treatment. The aim of this dissertation is to investigate the structural and magnetic properties of the (1–x) g-C₃N₄/(x) Fe₃O₄ composite, with x = 0.01, 0.05, 0.10, 0.20, 0.30, and 0.40. The focus is the comparison between powder and pelletized samples, analyzing the magnetic behavior of the material under different pelletizing pressures. Our results show that the application of pressure to the samples leads to significant changes in remanent magnetization, saturation magnetization, and coercive field, indicating that pressure-induced structural modifications directly affect the magnetic behavior of the composites.
Pseudociência no ensino: análise das percepções e posturas de professores da área de ciências da natureza
(Universidade Federal de São Carlos, 2026-04-24) Jesus, Atair José Bernardino de; Gebara, Maria José Fontana; https://lattes.cnpq.br/6585493065968961; https://orcid.org/0000-0002-9399-3683; https://lattes.cnpq.br/3604597187568726; https://orcid.org/0000-0002-3348-4860; Silva, Fernanda Keila Marinho da; Furlan, Elaine Gomes Matheus; Schappo, Marcelo Girardi; Cunha, Marcia Borin da; https://lattes.cnpq.br/6390725197649567; https://lattes.cnpq.br/3179865576965852; https://lattes.cnpq.br/8685622760352932; https://lattes.cnpq.br/3220989633467080; https://orcid.org/0000-0002-5697-8417; https://orcid.org/0000-0002-5413-7964; https://orcid.org/0009-0005-6362-4413; https://orcid.org/0000-0002-3953-5198
Pseudoscience consists of a body of ideas and practices characterized by difficult demarcation; it presents itself through explanations with a scientific appearance, yet fails to meet the fundamental criteria that characterize the production of scientific knowledge, such as testability, falsifiability, critical analysis, and peer validation. Beyond the difficulties regarding its definition, pseudoscience exhibits recurrent characteristics identified in the literature, such as a tendency to present itself as an alternative or superior form of knowledge, exploiting the limitations, uncertainties, and errors of science itself to challenge its credibility. These theories frequently gain traction by offering simplified and emotionally appealing explanations for complex phenomena, in contrast to science, which is characterized by the constant revision of hypotheses, critical analysis, and falsifiability. Adherence to pseudoscientific beliefs is also linked to cognitive predispositions that favor pattern-seeking, simple explanations, and a sense of control in the face of uncertainty. In this context, pseudoscience produces impacts that transcend the epistemological realm, affecting decisions related to health, social life, and education, particularly within a landscape marked by the widespread circulation of information and the dissemination of misinformation in digital environments. Although not a recent phenomenon, its presence has intensified across different social contexts, especially through conspiracy theories and anti-scientific discourses in digital media. This scenario became particularly evident during the COVID-19 pandemic, when pseudoscientific explanations and false information vied for space with established scientific knowledge on topics related to public health, vaccination, and disease prevention. Given this context, schools assume a significant role in promoting scientific literacy and critical thinking, leaving teachers to mediate between scientific knowledge and the discourses circulating in society. This research investigates the perception of K-12 Natural Sciences teachers regarding pseudoscience and its presence in education. It stems from the understanding that how educators comprehend, interpret, and position themselves toward pseudoscience can directly influence both classroom pedagogical practices and students' scientific development. The investigation was guided by the following research questions: How do K-12 Natural Sciences teachers perceive pseudoscience in education? What teacher positioning profiles regarding pseudoscience can be identified based on these teachers' perceptions? As a general objective, this study sought to analyze how K-12 Natural Sciences teachers perceive pseudoscience in education. Specifically, it aimed to identify and characterize teacher positioning profiles regarding pseudoscience based on the stances assumed by teachers within the context of Natural Sciences education; analyze the conceptions of pseudoscience held by these educators; and investigate how they self-identify in relation to pseudoscience. For data collection, the Investigative Questionnaire on the Perception of Pseudoscience in Natural Sciences Education (QIPPECN) was developed by the author in collaboration with the Research Group on Science Teaching and Communication (GPEDiC). The instrument was administered in a virtual environment via the Survio platform, with the participation of 178 K-12 teachers working across different educational stages, from Early Childhood Education to High School, spanning various regions of the country. Although it engaged participants from different cities and regions, the study does not claim statistical representativeness of the national teacher population. The research adopted a mixed-methods approach using elements of Content Analysis (CA), descriptive statistics, and inferential statistics, with emphasis on Exploratory Factor Analysis (EFA) and Spearman's correlation coefficient (ρ). The results indicate that most participants recognize pseudoscience as a domain distinct from science and reaffirm the centrality of scientific knowledge in Natural Sciences education. However, the data also reveal the coexistence of diverse teacher positions regarding the phenomenon, including perspectives aimed at critically confronting pseudoscience, neutral stances, and attitudes of omission, which may be associated with professional insecurity and the absence of explicit curricular guidelines for addressing the topic. Furthermore, conceptual gaps were identified among a portion of the respondents, who understand pseudoscience as a preliminary stage or a developing science, highlighting weaknesses in epistemological training.
Personalizing mental health support: a retrieval-based llm approach to conversational agent development
(Universidade Federal de São Carlos, 2026-04-24) Passador, Rafael Vinicius Polato; Caseli, Helena de Medeiros; https://lattes.cnpq.br/6608582057810385; https://lattes.cnpq.br/8275474576606344; Papa, João Paulo; Silva, Altigran Soares da; https://lattes.cnpq.br/9039182932747194; https://lattes.cnpq.br/3405503472010994
Mental health disorders constitute a major global challenge, requiring support solutions that are both accessible and scalable. Conversational agents have emerged as promising tools in this context, but purely generative large language model (LLM) systems remain susceptible to unsupported outputs, weak grounding, and limited adaptation to user-specific context. This work investigates whether combining supervised domain adaptation, guideline-grounded retrieval, and persona-based conditioning improves the quality of emotional-support responses. Three system variants built on the same fine-tuned Llama-3 backbone are compared: a Baseline model controlled through layered prompting, a Hybrid Retrieval-Augmented Generation (RAG) variant grounded in authoritative WHO and NICE mental health guidelines through hybrid retrieval, and a Persona-Based RAG variant that further incorporates persona signs into both retrieval and prompt construction. All variants are protected by a pre-LLM crisis guardrail that short-circuits high-risk inputs and returns a predefined safe response. Evaluation was conducted in a single-turn setting using inputs from mental health datasets associated with multiple persona profiles, and outputs were assessed through both human evaluation and an LLM-as-a-judge protocol under a shared rubric covering empathy, topic adequacy, and personalization. Results from both evaluation sources converge in identifying the Persona-Based RAG variant as the best overall configuration, with the clearest gains in empathy and personalization, while the RAG-only variant remained especially competitive in topic adequacy. These findings indicate that combining grounded retrieval with explicit persona conditioning is a promising strategy for the development of more contextually appropriate and supportive conversational agents for mental health.