Artificial neural network based scheme for voltage and harmonic compensation

Xie, Boluo (2006) Artificial neural network based scheme for voltage and harmonic compensation. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (5MB)


This thesis presents the analysis and design of the application of a novel hybrid active power filter (APF). The proposed APF scheme employs the recurrent artificial neural network (RANN) structure to determine each harmonic component in the distorted line voltage and the nonlinear line current. The compensation unit of the APF employs the multi-loop feedback control strategy to operate the PWM voltage source inverter to generate the replica of the harmonics and inject them into the system for compensation. The implementation of the proposed APF can maintain the utility supply current and the voltage at the point of common coupling (PCC) in a distribution network almost sinusoidal and minimize the total harmonic distortion (THD) levels in the network. Applying the harmonics extraction method and inverter control scheme, three compensation topologies were investigated to examine the functionality of the system under different source-end disturbances and load-end conditions. The obtained simulation results showed that the compensation system is capable of mitigating the harmonic and distortion caused by the operation of two kinds of nonlinear loads. In order to identify the contribution of design parameters and their interactions on the performance of the overall compensation system, the response surface method (RSM) was used to investigate the influence of the controlling parameters such as inverter controller gains and low-pass filter specification.

Item Type: Thesis (Masters)
Item ID: 10173
Additional Information: Includes bibliographical references (leaves 167-172).
Department(s): Engineering and Applied Science, Faculty of
Date: 2006
Date Type: Submission
Library of Congress Subject Heading: Neural networks (Computer science)--Industrial applications; Power electronics.

Actions (login required)

View Item View Item


Downloads per month over the past year

View more statistics