Overvoltages due to maneuvers are among the most severe transients that damage the electrical insulation of components in power systems, especially extra-high voltage systems (EHV). These overvoltages are generated due to the non-synchronized closing of the mechanical poles of the circuit breakers, which are characterized by high peaks of short duration. As the exact prediction of the closing times is uncertain, the problem requires a statistical treatment for the estimation of the switching (maneuver) surge voltages that are essential for the design of the electrical insulation of the components. Usually, the probability distribution of overvoltages due to maneuvers in power systems is computed considering the soil modeled by its constant electrical parameters. However, such studies have not yet been conducted for power systems located on real soils, whose electrical parameters (resistivity and permittivity) are variable with frequency due to the displacement currents in the ground, which is a significant factor especially in regions of high resistivity soil. In this context, it is expected that surge probability distributions will be modified when these factors are taken into account in the transient analysis. This project proposes to investigate the probability distributions of overvoltages due to switching operations in EHV lines located on high resistivity and frequency-dependent soils. This study is justified due to the fact that at these voltage levels, overvoltages due to switching are more important than overvoltages generated by lightning strikes. In addition, it is known that when the soil is modeled by its frequency-dependent electrical parameters, the overvoltages obtained have reduced peak values compared to those obtained for the soil with constant parameters. In this context, the cost of the insulator string can be reduced when a more realistic model is used. The study includes several sensitivity analyses to investigate the impact of different parameters of the power system on the overvoltages such as the line length, the nominal system voltage, the electrical line parameters, the inclusion of frequency-dependent soil electrical parameters, and the line topology. Furthermore, to carry out more realistic studies, the metallic structures and tower-footing grounding systems are computed by full-wave electromagnetic models and subsequently represented in the form of black-box type models. From the studies performed, it is expected to provide more accurate transmission line models for the simulations of statistical electromagnetic transient events adding new features to EMTP-type programs.
News published in Agência FAPESP Newsletter about the scholarship: