https://repositorio.unifei.edu.br/jspui/handle/123456789/87 2026-04-23T10:31:17Z https://repositorio.unifei.edu.br/jspui/handle/123456789/3173 Aquisição e validação de sinal de ruído eletroquímico Corrosion is a global problem, which implies costs in industrialized countries of up to 4.5% of GDP, with either economic, but also social and environmental impacts. In the case of Brazil, the waste of water supply networks due to leaks loss is quite significant and much of it is caused by network degradation, indicating that corrosion control should be promoted whenever possible. This study proposes a corrosion monitoring system, in system subject to the use of inhibitor, with the approach of passive technique for monitoring corrosion by electrochemical noise (EN), in which the classification of events in a corrosion sensor by EN is part of methodological study for structural integrity (or “health”) monitoring system (SHM). Due to very dynamic and stochastic nature of the signal, this study and analysis of EN measurements (ENM) considers numerical and graphic characteristics of two corrosion systems both in saline aqueous solution: carbon steel and stainless steel. These experiments are repeated for accumulating data, which allow the generation of several graphs in time and frequency domains, from which at least one characteristic is extracted, which has a good correlation with data from corrosion processes. Then, based on a supervised machine learning system, the training data allows the model to be calibrated. From the test data, the correctness rate of the model above 50% is verified. Dissertação 2022-02-17T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4387 Riscos climáticos e vulnerabilidades de plantios florestais comerciais no brasil sob condições climáticas atuais e futuras The Brazilian forest sector is primarily supported by planted forests, which cover ~10.52 million hectares, account for more than USD 15.7 billion in exports, and generate ~ 2.8 million direct and indirect jobs. Traditionally, this commercial production has been dominated by exotic species such as Pinus spp., although there is growing interest in more profitable and high-value-added species, including teak (Tectona grandis L.f.), Brazilian mahogany (Swietenia macrophylla King), and African mahogany (Khaya spp.). Despite advances in forest management and silvicultural practices, the wood volume and quality of these species are likely to be impacted by climate conditions. To address these challenges, using climatic and bioclimatic risk indicators is an effective strategy for quantifying the vulnerabilities of Brazilian plantations under current and future climates. Although these indicators are essential tools for assessing magnitude, spatial distribution, typology, and severity levels, their application to Brazilian forest species of commercial interest remains limited. To fill this gap, this dissertation assesses the risks and vulnerabilities of commercial plantations of Pinus spp., T. grandis, S. macrophylla, and Khaya spp. in Brazil by developing and applying 36 specific indicators under current and future climate conditions. To achieve this aim, daily data of near-surface minimum (Tasmin, °C), mean (Tas, °C), and maximum (Tasmax, °C) air temperature, precipitation (P; mm day⁻¹), relative humidity (Hurs; %), and global solar radiation (Rsds; MJ m⁻² day⁻¹) were obtained from sixteen Global Climate Models (GCMs) from the NASA Earth Exchange Global Daily Downscaled Projection (NEX-GDDP-CMIP6). All risk indicators developed in this study were simulated for the baseline period (BP; 1995–2014) and projected for the near future (NF; 2041–2060) and far future (FF; 2081–2100) under two socioeconomic scenarios (SSP3-7.0 and SSP5-8.5). By the end of the 21st century, the greatest risks to Pinus plantations are associated with productivity losses (approximately 84% of the area under moderate to very high risk in the far future), followed by the wood wasp (Sirex noctilio) (17–87%), tip blight caused by Sphaeropsis sapinea (58–66%), stress related to basal temperatures (5–64%), needle blight caused by Calonectria pteridis (12–53%), and water deficit stress (32–45%). For T. grandis, the main risks are basal temperature stress (88–98%), productivity losses (84–88%), and canker incidence (53–59%). Lastly, S. macrophylla and Khaya spp. shown greater vulnerability to thermal stress, particularly at basal (48–89%) and maximum temperatures (41–80%). To mitigate productivity losses and minimize phytosanitary risks in these plantations, the adoption of effective and economically viable adaptation measures will be essential. These measures include selecting species and hybrids more tolerant to water and thermal stresses, implementing appropriate spacing and planting density management, performing thinning and pruning at appropriate intervals for each species, and strengthening control strategies for Sphaeropsis sapinea, Calonectria pteridis, Sirex noctilio, Hyblaea puera, and Hypsipyla robusta. Dissertação 2026-02-23T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4386 Detecção de defeitos em máquinas de fabricação de copos de papel utilizando sinais acústicos e de vibração com algoritmos de aprendizado de máquina e redes neurais Early detection of defects in industrial machinery is essential to ensure high operational availability, reduce maintenance costs, and prevent productivity losses, particularly within the context of Industry 4.0. This study proposes and evaluates an approach for fault diagnosis in paper cup manufacturing machines by integrating acoustic and vibration signals to detect slack in Chain 1 of the transmission system, a critical component for equipment synchronization. Data were collected during a real production process in a packaging company, ensuring practical representativeness of operational conditions. Nineteen experimental acquisitions were conducted based on a structured Design of Experiments (DOE), ensuring systematic variation of machine operating conditions. In total, approximately 50 minutes of recordings were obtained. The acquired signals were segmented into 5-second windows with 50% overlap, resulting in a dataset comprising 1,453 instances. The modeling stage was structured into two complementary approaches. The first was based on manual feature engineering, extracting statistical and spectral descriptors from audio and vibration signals and using them as input to traditional statistical models (Logistic Regression and Linear Discriminant Analysis) and classical machine learning algorithms, including Random Forest, Support Vector Machine, Multilayer Perceptron, and a soft-voting Ensemble model. The second approach employed deep learning through Convolutional Neural Networks (CNNs) applied to Mel spectrograms extracted exclusively from the audio signal, enabling automatic learning of relevant time–frequency representations for fault diagnosis. Model evaluation was performed using stratified internal validation and an external test set composed of completely unseen experimental runs, ensuring a rigorous estimation of generalization capability. To mitigate stochastic effects, experiments were repeated across multiple independent executions, and average performance metrics along with their confidence intervals were analyzed. Results indicate that multisensory integration of audio and vibration signals significantly improved performance and robustness compared to single-sensor approaches, substantially reducing the gap between validation and external testing. The feature-based approach achieved an average external test accuracy of 94% in its best multisensory configuration, while the CNN-based approach reached an average accuracy of 92% using audio alone, demonstrating competitive performance even under less intrusive instrumentation. Overall, the findings confirm that multisource sensory integration enhances diagnostic robustness, while deep learning approaches provide a promising alternative in scenarios with instrumentation constraints. The proposed method shows strong potential for predictive maintenance systems and quantitatively advances the application of Artificial Intelligence techniques in industrial fault diagnosis. Dissertação 2026-02-24T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4385 Teoria de perturbações lineares num modelo intrinsecamente simétrico Contemporary cosmology is largely grounded in the standard ΛCDM model, based on the Friedmann–Lemaître–Robertson–Walker (FLRW) metric and on the Cosmological Principle, which assumes large-scale homogeneity and isotropy. This model has been highly successful in describing cosmic expansion, the anisotropies of the cosmic microwave background, and the formation of large-scale structures, although its formulation depends on the introduction of components that are not yet fully understood, such as dark matter and dark energy, in addition to assuming global symmetries that contrast with the observed inhomogeneities, such as clusters, filaments, and cosmic voids. In light of these limitations, alternative models have been developed that relax the conditions of the Cosmological Principle, such as the Lemaître–Tolman–Bondi (LTB), Szekeres, and Locally Rotationally Symmetric (LRS) models, allowing for more general matter distributions. More recently, intrinsically symmetric models have emerged, which preserve internal symmetries without requiring global homogeneity, providing a more flexible framework to investigate the effects of inhomogeneities on cosmological parameters. In this work, we develop the linear scalar perturbation framework for intrinsically symmetric models, analyzing how the presence of spatial gradients in the geometric background alters the evolution equations of the fluctuations. Our goal is to establish the theoretical basis to confront these models with the ΛCDM paradigm and investigate potential observational signatures of inhomogeneity. Dissertação 2026-02-20T00:00:00Z