Wyniki 1-1 spośród 1 dla zapytania: authorDesc:"Jumana ALSHAWAWREH"

Adaptive Technique based on Fast Fourier Transform for Selecting the Modelled Harmonics' orders in Kalman filter DOI:10.15199/48.2018.08.23

Czytaj za darmo! »

Harmonics are existed in power system due to the power electronics devices, nonlinear loads and Photo Voltaic PV plants.. etc[1] , where the Total Harmonics Distortion THD is not fixed, and it may increase rapidly under certain operating conditions, such as low radiation in PV system [2]. The increment of THD may be due to the increment of the harmonics amplitude or the increment of the numbers of the harmonics’ components in the signals[2]. Many of passive and active filters have been used to estimate the fundamental component in distorted input signal, where exact value of the fundamental signal is necessary for the protection devices to work probably. Kalman filter is one of the most of filters that have been used in power system and power system apparatus for different purposes [3]-[6] especially for harmonic signal estimation in power system[7]-[10] and it showed a better performance compared to other filters [11], but it is required to model the system exactly which means that all the harmonics in the signal must be modelled. It has been discussed in details the effect of harmonic modelling in the Kalman filter in [12], and it was concluded that the best performance of Kalman filter achieved, when all the harmonics in the input signal are modelled, and it was also concluded that, estimating extra harmonics, which are not existed in the measured signal, doesn’t affect the Kalman filter output. The error signal (difference between the input signal and the estimated output by the Kalman filter), becomes zero only when all the harmonics of the input signal are included in the Kalman model. The problem here, when the order of harmonics are changed, while the modelled harmonics are unchanged this will cause the error signal to exceed an acceptable limit, and the order of the estimated harmonics must be modified to reduce that error, for that reason an adaptive Kalman filter is proposed in this paper to model t[...]

 Strona 1