T.K. Abdel-Galil,∗ H.H. Zeineldin,∗∗ E.F. El-Saadany,∗∗∗ and M.M.A. Salama∗∗∗
Artificial neural networks, capacitor switching, total least square- estimation of signal parameters via rotational invariance
This paper presents a new method in identifying the switched ca- pacitor location from the measured transient voltage waveform. The measured transient voltage waveform due to capacitor switching is composed of a summation of damped sinusoidal components. Total Least Square-Estimation of Signal Parameters Via Rotational In- variance Technique (TLS-ESPRIT) is used to estimate the transient signal parameters of different modes contained in the signals. These parameters are functions of the switched capacitor location. Feed Forward Artificial Neural Network (FF-ANN) is employed to iden- tify the location of the switched capacitor based on the frequencies and amplitudes of various modes, contained in the transient voltage signal, extracted using TLS-ESPRIT. The simulation results prove the competence of the proposed methodology.
Important Links:
Go Back
