Implementação computacional do estimador de estado de mínimos quadrados considerando elementos shunt e a presença de transformadores com controle de tensão e defasadores

Abstract

Presented in 1970 by Schweppe, State Estimation is a tool that uses the parameters of the electrical system and a set of strategically located measurements to determine the operating point of an electrical power system and to monitor the behavior of the electrical system in real time. Traditionally, state estimation uses the weighted least squares iterative method to estimate the values of the state variables of an electrical power system. Although the classical formulation of the least squares method is the most used by experts and researchers, this formulation cannot accurately represent test systems with phase-shifting transformers, transformers with taps, and shunt devices in the buses, requiring some approximations to estimate the operating point. The general objective of this work is to develop computationally in MATLAB software the classical method of Weighted Least Squares State Estimation that allows obtaining the operating point and performing error analysis of electrical power systems considering the taps of transformers, phase-shifting transformers, and shunt elements in the buses in the power flow formulation. For this, classical state estimation techniques were consulted, and adaptations were made to the formulation necessary to include the parameters of these elements, with the premise of the state estimation algorithm. The validation of the results considered the electrical systems published in the IEEE of 14, 30, and 57 buses, considering three scenarios for each case: the first scenario being the set of measurements without errors, the second scenario with a measurement with Gaussian error, and the third scenario with a measurement with gross error in the set of measurements. For the three cases, considering the set of measurements without error, the algorithm was satisfactory for WLSSE with the new parameters included in the formulation. While for the second scenario containing a Gaussian error measurement in the three cases, the method did not identify the presence of this error. For the third scenario, in the first case of 14 buses, the algorithm estimated the state of the EPS, detected, and correctly identified the measurement containing gross error; in the second case, the method did not detect the gross error; and in the third case, the method detected the presence of gross error but did not correctly identify it.