Adaptive discrete-time grid-voltage sensorless interfacing scheme for grid-connected DG-inverters based on neural-network identification and deadbeat current regulation

Yasser Abdel Rady Ibrahim Mohamed, Ehab F. El-Saadany

Research output: Contribution to journalArticlepeer-review

77 Scopus citations

Abstract

This paper presents an adaptive discrete-time grid-voltage sensorless interfacing scheme for grid-connected distributed generation inverters, based on neural network identification and deadbeat current regulation. First, a novel neural network-based estimation unit is designed with low computational demand to estimate, in real-time, the interfacing parameters and the grid voltage vector simultaneously. A reliable solution to the present nonlinear estimation problem is presented by combining a neural network interfacing-parameters identifier with a neural network grid-voltage estimator. Second, an adaptive deadbeat current controller is designed with high bandwidth characteristics by adopting a delay compensation method. The delay compensation method utilizes the predictive nature of the estimated quantities to compensate for total system delays and to enable real-time design of the deadbeat controller. Third, the estimated grid voltage is utilized to realize a grid-voltage sensorless average-power control loop, which guarantees high power quality injection. Theoretical analysis and comparative evaluation results are presented to demonstrate the effectiveness of the proposed control scheme.

Original languageBritish English
Pages (from-to)308-321
Number of pages14
JournalIEEE Transactions on Power Electronics
Volume23
Issue number1
DOIs
StatePublished - Jan 2008

Keywords

  • Deadbeat current control
  • Digital control
  • Distributed generation (DG)
  • Grid-voltage sensorless control
  • Neural network identification
  • Pulsewidth modulated (PWM) inverters

Fingerprint

Dive into the research topics of 'Adaptive discrete-time grid-voltage sensorless interfacing scheme for grid-connected DG-inverters based on neural-network identification and deadbeat current regulation'. Together they form a unique fingerprint.

Cite this