https://repositorio.ufu.br/handle/123456789/5465 2026-09-28T16:31:51Z https://repositorio.ufu.br/handle/123456789/50464 Title: Aprendizado profundo para estimação de instantes de chegada com aplicação à detecção e localização de descargas parciais em cabos isolados de média tensão Abstract: Partial discharge is one of the main precursors of failure in insulated medium-voltage cables. Diagnosis based on its measurement can answer two questions: whether there is partial discharge activity in the cable and where the source is located. This thesis proposes treating detection and localization as parts of a single problem: estimating, in each signal, the set of arrival times of the partial discharge pulses, whose number is not known in advance and may be zero. To this end, the proposed deep learning model produces, for each sample of the signal, a value indicating the confidence that the corresponding instant is the arrival of a pulse. Under this formulation, detection occurs when at least one arrival time is found, and localization follows from single-ended time-domain reflectometry, from the difference between the arrival times of the direct pulse and its reflection. Training and internal evaluation use the author’s own datasets, published under an open license, comprising laboratory and field measurements. Robustness is further assessed on a dataset acquired and published by an independent institution, to which the formulation is applied without any readjustment. The estimation error is assessed under two aspects: identification, which measures whether the correct pulses were found, and timing accuracy, which measures the temporal error of each pulse found. The comparison is made against classical arrival time estimation procedures, calibrated on the same data and under the same protocol. In identification, the formulation achieves an F1 of 0.987, against 0.699 to 0.799 for those procedures, and keeps this advantage throughout the operating range of signal-to-noise ratio evaluated. In timing accuracy, the bias is sub-sample (0.16 sample) and the dispersion is 1.4 samples (14 ns), 32% lower than that of the best procedure compared. Under a change of measurement scenario, with different instrumentation and discharge sources, the formulation retains its performance, with an F1 of 0.962. Two case studies demonstrate the diagnostic outputs. In a field campaign of 8135 records, analyzed with no prior screening, all 1251 records containing partial discharge activity are detected, with false positives in 0.68% of the remaining records. In a laboratory test with a documented defect position, the source is located 1.70 m from the actual position, in a 144 m cable. 2026-09-04T00:00:00Z https://repositorio.ufu.br/handle/123456789/50363 Title: Sistema de avaliação da confiabilidade dielétrica de cabos isolados de alta tensão baseado em descarga parcial como módulo de digital shadow Abstract: The evaluation of the dielectric reliability of medium and high voltage power cables is a very important procedure for the prevention of failures and operational continuity in industrial environments. Insulation degradation, driven by mechanisms such as thermal aging, moisture penetration, and localized defects, manifests itself early through partial discharges, whose systematic monitoring and interpretation make it possible to anticipate critical conditions before asset failure. In this context, the present work proposes a methodology for evaluating the dielectric reliability of medium and high voltage cables from the monitoring of partial discharges, designed to operate as the dielectric evaluation module within a Digital Shadow architecture aimed at the management of electrical assets. The methodology is organized in a sequential pipeline of five steps: (i) signal acquisition by means of sensors (either in offline or online regimes), (ii) separation of signals from partial noise discharges, by hierarchical clustering by Ward's method, (iii) evaluation of dielectric severity by fuzzy logic, (iv) estimation of the cable state in the Weibull reliability curve, through a degradation model and (v) weighting adjustments by the signal quality index and final evaluation by condition ranges based on IEEE Std 400.3 and CIGRE TB 841 guidelines. These methodological procedures, when added together, aim to qualify the reliability of the cable based on its discharge signals. This set must work within the concept of "Digital Shadow", receiving data from the sensors installed in the physical world, processing it and producing, at its end (usually in a cloud system) a coherent physical indication of the state of the (insulation) of the cable under monitoring. 2026-07-31T00:00:00Z https://repositorio.ufu.br/handle/123456789/50351 Title: Técnicas de Inteligência Artificial aplicadas à otimização multiobjetivo de equipamentos eletromagnéticos saturáveis: um estudo de caso em reatores a núcleo saturado Abstract: This thesis proposes and validates a multi-objective optimization methodology for the design of large-scale Saturated Core Reactors (SCRs), based on replacing three-dimensional Finite Element simulations with Artificial Neural Networks (ANNs) trained as surrogate models. The work begins by contextualizing the design problem of electromagnetic equipment and identifying the research gap concerning the application of ANN-based surrogate models to three-dimensional saturated magnetic structures. A reference reactor, rated at 500 kVAr in a Twin-Trippler topology, was designed, modeled in ANSYS Maxwell software, and experimentally validated through the fabrication of a physical prototype and factory acceptance electrical tests, confirming the accuracy of the adopted computational model. From a simplified model, shown to be equivalent to the detailed manufacturing model, a parametric sweep using Finite Element simulations was conducted to generate a training dataset. A comparative study among machine learning techniques demonstrated the superiority of ANNs in capturing the strongly nonlinear behavior of the equipment under deep saturation, achieving coefficients of determination (R²) above 0.96 for the reactive power and core loss objective functions. The trained ANNs were then employed to exhaustively explore a design space of more than 247 million combinations, enabling the construction of the Pareto frontier between core losses and volume within minutes of processing. The selected optimal design, validated through detailed Finite Element simulation, achieved reductions of 12.86% in magnetic core losses and 14.37% in core volume relative to the reference equipment, while preserving the rated reactive power absorption capacity within acceptable design margins. The results confirm the effectiveness of the proposed methodology as a computationally efficient and experimentally validated design tool for electromagnetic equipment operating under deep magnetic saturation. 2026-08-17T00:00:00Z https://repositorio.ufu.br/handle/123456789/49315 Title: Novo conversor isolado de múltiplas entradas e múltiplas saídas multidirecional para arquiteturas modernas de distribuição de energia CC de baixa tensão Abstract: This work presents the modelling, analysis, and simulation-based evaluation of a multidirectional, multi-winding, multiport flyback DC-DC converter proposed for the integration, within a single conversion stage, of sources and loads operating at different voltage levels in low-voltage DC architectures. By doing so, the need for intermediate DC-DC converters and the losses inherent to multistage power processing are reduced, thereby favoring higher power density. As a proof of concept, a three-port configuration composed of a photovoltaic source, a DC load, and a battery energy storage system was adopted to develop the analytical formulations, the control strategy, and the validation studies. For converter operation, an operating scheme is adopted in which the magnetization and demagnetization processes of the coupled inductor take place individually for each port, thereby reducing the dynamic coupling among ports, increasing the flexibility of the turns-ratio design, and simplifying control implementation. Energy coordination is performed through a Power Flow Control Strategy (PFCS), in which the voltage of port 2 acts as the global coordination variable and is maintained within a band delimited by four voltage levels, from which the operating modes are selected according to the balance among photovoltaic generation, load demand, and the power-exchange capability of the storage system. Within this framework, port 1, connected to the photovoltaic system, operates either in Maximum Power Point Tracking (MPPT) mode or in voltage-droop-based power-limiting mode, referred to as Voltage Droop Control Mode (VDCM). For this port, a Finite Control Set Model Predictive Control (FCS-MPC) strategy is employed to track the photovoltaic array voltage reference, which is generated either by the MPPT algorithm, featuring an adaptive step size and based on observer-estimated photovoltaic current without the need for a current sensor, or by the VDCM, depending on the system operating mode. Port 3, associated with battery energy storage, is bidirectionally controlled through two voltage-droop strategies, one dedicated to the charging process and the other to the discharging process. Under surplus-energy conditions, the system selects the Battery Voltage Droop Control by Charging Mode (BVDMcha), in which the battery absorbs power; under power-deficit conditions, it selects the Battery Voltage Droop Control by Discharging Mode (BVDMdis), in which the battery supplies power to the system. When no energy exchange is required, port 3 remains in a standby state characterized by the IDLE mode. In a complementary manner, port 2, associated with the DC load, incorporates protection functions that allow the load either to remain connected or to be disconnected. Simulation results under islanded operation and different irradiance and load-demand conditions showed that the voltage of port 4 2 remained within the specified range under both surplus- and deficit-power scenarios, with port 3 transitioning between the BVDMcha and BVDMdis modes to absorb or supply energy as required. In addition, a comparative evaluation of alternative control strategies for port 3 was carried out, highlighting differences in dynamic compatibility with the photovoltaic control loop. In summary, the obtained results indicate that the proposed multiport topology, together with the new operating scheme and the developed control strategy, constitutes a technically viable solution for integrating, within a single conversion stage, elements operating at different voltage levels in DC architectures. 2026-06-12T00:00:00Z