Contribuições para o Processo de Análise e Emissão de Pareceres sobre Pedidos de Indenização por Danos Elétricos Através de Estimadores de Estado

Abstract

Given the increasing number of claims and reimbursement related to electrical damage in equipment and processes, electric utility companies, together with governmental and regulatory agencies, have sought ways to improve analysis procedures. Current practices, which rely primarily on a simple correlation between reported damages and events occurring in the power grid, have proven to be technically fragile and insufficient. In response, recent years have seen a growing number of studies aiming to develop more robust, science-based analytical methods. Among these, a prominent approach is based on correlating the dielectric and thermal impacts of electrical disturbances on consumer units with the physical and operational withstand capabilities of the affected equipment and processes. Generally, the procedures in use involve computational simulations to assess causal relationships. Anomalous phenomena observed in the field are simulated and propagated through the modeled electrical network up to the point of connection of the claimant, where the affected equipment is represented. A specialized software then compares the voltages and currents manifested in the electronic devices with the dielectric and thermal withstand limits of the claimed components, producing technical report regarding the validity of the compensation request. Although this method represents a significant advancement, its reliability can be compromised using inaccurate models or parameters for the electrical network and its components. As a result, the effects observed at consumer connection points may not provide the required confidence for definitive analysis. To address these limitations, this thesis proposes an improved evaluation methodology based on measurements of voltage and current obtained from event recorders installed at monitored buses and lines. Using these field data, the method estimates the electrical quantities at unmonitored points. In doing so, the causal relationship is assessed directly through the impactful electrical stresses, eliminating the need for computational propagation studies — a crucial step toward enhancing the reliability of PID analysis.