Estimating the Frequency Variation of the Mexican Grid by Kalman Filtering

Abstract

This work describes the use of Kalman filtering to estimate the changes of frequency of the Mexican Grid. An off-line analysis of the frequency variation is presented. This analysis was done using real data extracted from the Mexican Grid during August 2004. It is shown that, although the Kalman indexes do not converge to an exact value, they maintain a low variation following the initial iterative steps. The precision of the estimation is limited by unknown load disturbances but this technique could be useful to predict the frequency variation and also the load variation of the Mexican grid.