Studying the propagation of harmonics in power grids represents a very important task nowadays. However, this demands for nonlinear models of power system devices representing a critical tradeoff between the accuracy and the computational cost. Simplified Volterra models with quasi-sinusoidal excitation have been recently proposed and they appear to be particularly suitable for the purpose. In this paper, these models have been applied to predict, for a given input voltage, the current harmonics of a full-bridge rectifier feeding an ohmic-capacitive load. Accuracy has been evaluated by means of both numerical simulations and experiments. Robustness has also been assessed by considering a set of realistic voltage waveforms different from those employed for the identification. Results highlight the remarkable accuracy of the proposed models in spite of the relatively small set of coefficients to be identified. This feature is extremely important when models have to be employed in frequency-domain simulations of large networks.
Behavioral Representation of a Bridge Rectifier Using Simplified Volterra Models
Faifer, Marco;Laurano, Christian;Ottoboni, Roberto;Toscani, Sergio;Zanoni, Michele
2019-01-01
Abstract
Studying the propagation of harmonics in power grids represents a very important task nowadays. However, this demands for nonlinear models of power system devices representing a critical tradeoff between the accuracy and the computational cost. Simplified Volterra models with quasi-sinusoidal excitation have been recently proposed and they appear to be particularly suitable for the purpose. In this paper, these models have been applied to predict, for a given input voltage, the current harmonics of a full-bridge rectifier feeding an ohmic-capacitive load. Accuracy has been evaluated by means of both numerical simulations and experiments. Robustness has also been assessed by considering a set of realistic voltage waveforms different from those employed for the identification. Results highlight the remarkable accuracy of the proposed models in spite of the relatively small set of coefficients to be identified. This feature is extremely important when models have to be employed in frequency-domain simulations of large networks.File | Dimensione | Formato | |
---|---|---|---|
08488577.pdf
Accesso riservato
Descrizione: Articolo principale
:
Publisher’s version
Dimensione
1.89 MB
Formato
Adobe PDF
|
1.89 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.