Estimação de parâmetros de sinais com saltos de magnitude, fase ou frequência com aplicações na calibração de unidades de medição fasorial

• Autor(a) principal: Martins, Marcelo Britto
• Data de Publicação: 2022
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações do LNCC
• Texto Completo: https://tede.lncc.br/handle/tede/365
• topic: Redes elétricas inteligentes; Unidades de medição fasorial; Calibração; Metrologia; Sincrofasores; Afundamentos de tensão; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::SISTEMAS ELETRICOS DE POTENCIA::TRANSMISSAO DA ENERGIA ELETRICA, DISTRIBUICAO DA ENERGIA ELETRICA

The measurement of synchrophasors with Phasor Measurement Units (PMUs) is one of the key technologies for the development of smart electrical grids and has been increasingly used for the prevention of blackouts, protection, monitoring, and evaluation of network disturbances, among other possible applications [1]. With the increasing inclusion of renewable energy sources and non-linear loads in the electrical grid, the measurement of electrical quantities becomes more challenging, due to the greater presence of disturbances in the grid [2]. Anomalies such as harmonics, interharmonics, voltage dips and frequency variations are adverse conditions that can put the reliability of synchrophasor measurementes below acceptable limits for typical applications. For this reason, in recent years, electrical metrology, traditionally developed using steady state signal analysis, has turned to dynamic state signal analysis. For this, signal analysis tools have been used, in order to obtain greater accuracy of measurements and shorter response time in the measurement of quantities of interest [3 –9], or evaluate in real time the measurements reliability [10–12]. For metrological evaluation of PMUs, calibration systems have been developed in recent years in several National Metrology Laboratories around the world [13–16], including INMETRO, in Brazil [17–20]. In the context of dynamic regime synchrophasor measurement, the development of calibration systems capable of reproducing adverse conditions typically present in electrical networks is necessary, in order to metrologically characterize PMUs under more realistic conditions. In turn, the evaluation of the calibration systems themselves is a task that entails the use of techniques for analyzing digitized signals adapted to dynamic regime. Specifically, there is a growing concern with the local estimation of the electrical system frequency performed by PMUs, especially in networks with high insertion of renewable energies generation [21, 22], whose protection systems are particularly sensitive to errors in these measurements. In this thesis, motivated by the pressing needs of metrology laboratories, we propose and evaluate appropriate tools for analyzing AC signals that contain magnitude, phase and frequency steps. At first, we propose a parametric phasor estimator with a magnitude or phase step, using the Levenberg–Marquardt algorithm [23, 24], together with definitions of intermediate phasors for AC signals with a magnitude or step phase [19]. Then, we carried out pioneering work in the proposition of detectors and estimators for the location of step jumps in the magnitude, phase and frequency of an AC signal, based on analyses of instantaneous functions of magnitude and phase of the analytical Hilbert signal associated with the AC signal [20]. We then evolved such analyses for the development of electrical system frequency estimators from AC signals with one step in magnitude, phase or frequency. The signal analysis techniques used in the proposed electrical system frequency estimators allow, additionally, to estimate the height of the frequency step, which also allowed us to propose discriminators of AC signals containing a step in frequency from those containing a step in magnitude or phase. Such discrimination is particularly desirable to improve the decision making performance of network protection systems with high insertion of energy generation from renewable sources, such as wind and solar. In all cases, we performed extensive Monte Carlo simulations to estimate mean errors and uncertainties associated with the measurements performed by each provided estimator, as well as its performance sensitivity with respect to several parameters. This allows estimating the contribution of each estimator to the total uncertainty of a synchrophasor estimation system that makes use of it. Finally, we implemented the proposed estimators in prototypes of PMU calibration systems at INMETRO’s Laboratory of Metrology in Electric Energy, which allowed us to estimate the total uncertainty of measurements, assess the suitability of the equipment used to carry out standardized tests, and provide subsidies for discussions on the development of synchrophasor technology and possible revisions to related standards.