https://repositorio.unifei.edu.br/jspui/handle/123456789/55 2026-09-03T06:09:45Z https://repositorio.unifei.edu.br/jspui/handle/123456789/3173 Título: Aquisição e validação de sinal de ruído eletroquímico Abstract: 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. Tipo: Dissertação 2022-02-17T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4469 Título: Avaliação de parâmetros associados ao envelhecimento de para-raios submetidos a descargas de corrente Abstract: Surge arresters are protective devices installed throughout power systems to limit overvoltages caused by lightning impulses, switching operations, and network faults. However, these devices are subject to degradation processes that may compromise their performance, lead to failure, and affect system reliability. In this context, it is essential to investigate the mechanisms associated with surge arrester aging and to identify the most sensitive parameters for monitoring their operational condition. For this purpose, an aging procedure based on current impulse stresses was applied to three groups of surge arresters from different manufacturers, each group consisting of three 15 kV samples. Throughout the aging process, the devices were evaluated through leakage current and partial discharge measurements. At the end of the aging procedure, microstructural analyses were performed on both new and aged samples to investigate the effects of degradation. The results indicated that leakage current-derived parameters exhibited the highest sensitivity for monitoring surge arrester aging, with dissipated power and third-order harmonics standing out as the most representative indicators. Partial discharge activity was detected only in a subset of the samples and during the final stages of the degradation process. Furthermore, samples exhibiting partial discharge activity also showed changes in the final leakage current measurements. Microstructural analyses confirmed the degradation evidence observed in the electrical tests. Therefore, it was concluded that the combined application of the employed techniques was effective for assessing the aging of distribution surge arresters, with leakage current-derived parameters emerging as the most promising indicators for monitoring the degradation of these devices. Tipo: Dissertação 2026-06-29T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4468 Título: Electric motor fault detection under nonstationary operation using koopman operator spectral analysis. Abstract: Accurate motor fault detection and flow rate estimation in industrial centrifugal pumps is essential for diagnostics, performance monitoring, and process optimization, yet traditional frequency-domain techniques such as Motor Current Signature Analysis (MCSA) often rely on fixed spectral bands and assume stationarity, making them less effective under variable-speed operation, changing loading conditions, and multiphase flow environments. This work introduces a Koopman-theoretic framework for motor fault and flow-rate classification based on Hankel-embedded Higher-Order Dynamic Mode Decomposition (HODMD), enabling extraction of physically interpretable spectral-dynamical features from multichannel current, voltage, and vibration measurements. The proposed pipeline computes Koopman eigenvalues, modal energies, growth rates, temporal coherence, and spectral entropy to construct a compact descriptor of pump electromechanical behaviour, which is subsequently evaluated through supervised and unsupervised learning pipelines incorporating SMOTE balancing, nonlinear UMAP projection, and extensive hyperparameter optimization via TPESearchCV. Using 2 laboratory datasets, the first consisting of 106 signals describing 4 motor faulty operation states with different input frequencies, its results were as high as 96.8% cross validation accuracy and 80.7% balanced test accuracy. The second dataset consists of 1304 signals spanning multiple rotational speeds, heterogeneous fluid compositions, and 11 unbalanced flow-rate range classes, the system achieves up to 97.8% cross-validation balanced accuracy and up to 92.6% balanced test accuracy while revealing coherent manifold structure aligned with physical flow-rate regimes. Although overlapping operating states and speed-dependent spectral shifts introduce classification challenges, the results demonstrate that Koopman spectral feature captures the essential dynamic structure of the pump more effectively than traditional Fourier-based approaches, highlighting their potential for robust, reliable, and generalizable monitoring and diagnostics of industrial pumping systems. Tipo: Dissertação 2026-07-08T00:00:00Z https://repositorio.unifei.edu.br/jspui/handle/123456789/4467 Título: In-context learning versus trained models: generative AI and machine learning for surface roughness prediction in dry Ti-6Al-4V turning Abstract: Surface roughness Ra is a critical quality indicator in dry turning of Ti-6Al-4V titanium alloy, a material widely used in aerospace and biomedical applications. Empirical powerlaw models remain the industrial standard but cannot represent interaction effects or non-linear responses. Machine learning offers greater flexibility, yet its deployment is constrained by small experimental datasets and the need for rigorous validation protocols. Large language models represent an alternative predictive paradigm, potentially encoding implicit knowledge of machining physics that may support quantitative prediction through in-context learning alone. This work compares five predictive models, comprising two conventional mathematical approaches (Power Law and RSM) and three machine learning models (SVR, Gaussian Process Regression, and XGBoost), against three LLMs (Claude Sonnet 4.5, GPT-4o-mini, and Gemini 3 Flash Preview) for Ra prediction in Ti-6Al- 4V turning. A Central Composite Design with k = 3 factors and two carbide inserts of distinct nose radii yielded N = 38 observations. The conventional and ML models were evaluated under nested LOOCV with 5-fold grid search where applicable. LLMs were accessed via API at temperature zero under two prompting strategies, few-shot and chain-of-thought, across three independent runs, with results reported as mean ± standard deviation. Claude Sonnet 4.5 in few-shot mode achieved the best overall performance (RMSE = 0.288 ± 0.016 μm, R2 = 0.985), outperforming SVR, the best trained model (RMSE = 0.341 μm), by 16% without task-specific training or fine-tuning. The effect of prompting strategy was model dependent: few-shot prompting outperformed chain-ofthought for Claude and GPT, whereas Gemini CoT achieved a substantially lower mean RMSE than Gemini FewShot but exhibited markedly greater run-to-run variability. This variability was especially pronounced for Gemini CoT, whose RMSE ranged from 0.31 to 0.99 μm across identical executions. The results show that a capable LLM supplied with well-structured experimental demonstrations can surpass optimised trained models for Ra prediction without task-specific training, while also demonstrating that predictive performance and stability depend strongly on both model architecture and prompting strategy. Tipo: Tese 2026-08-06T00:00:00Z