Noise Effect on Real-Time Estimation of Harmonics in Electrical Power Substations

This paper studies the effect of noise on real-time estimation of harmonics in a typical power substation. The study is carried out based on studying signal-to-noise ratio as a criterion for optimization and Monte Carlo simulation is used to compute the critical values of signal-to-noise ratio for minimizing the noise effect on estimation. To alleviate problem of skewness found in legacy state estimators that are employed in traditional power substations, the high rate measurements casted by the Merging Units installed at substation bay level is transmitted to a Kalman Filter function which is used as a dynamic estimator for estimating the harmonics. A collection of Matlab MEX functions are built to facilities Merging Unit function, A/D and D/A conversions function and the network communication function in conjunction with built-in Simulink block library are incorporated in one Matlab/Simulink-based simulator model and runs in XPC target to co-simulate execution of task in real-time. Representative results in evaluating performance of estimated harmonies as they affected by noise ratio are presented in this paper.