![](/static/media/logo.jpg)
An International Publisher for Academic and Scientific Journals
Author Login
Scholars Journal of Engineering and Technology | Volume-3 | Issue-09
A Neural Network Controller with Anti Windup Based Shunt Active Power Filter for Harmonic Mitigation
Angayarkanni S, Vanitha U
Published: Sept. 30, 2015 |
117
143
DOI: 10.36347/sjet
Pages: 724-731
Downloads
Abstract
Power quality problem is the most sensitive problem in a power system. The objective of this paper is to
reduce “harmonics” using a compensation technique. Shunt Active Power Filter (SAPF) is used to eliminate harmonic
current and to compensate reactive power for a three phase four wire unbalanced system. Here, a PI controller, Fuzzy
Logic and Artificial Neural Network based Controller are employed to maintain DC bus voltage. The advantage of fuzzy
control is that it provides linguistic values such as low, medium, high that are useful in case where the probability of the
event to occur is needed. The implementation of Artificial Intelligence as controller is that it does not require an accurate
mathematical model of the system. A MATLAB/SIMULINK has been used to perform the simulation. Simulink model
is developed for three phase four wire system under balanced supply and unbalanced supply condition. The
performance of PI, Fuzzy Logic ANN based controller is measured and their results are compared. Simulation results
obtained shows that the performance of neural network based controller is found to be better than PI controller and Fuzzy
Logic Controller. Anti-windup circuit is added with the PI controller in addition to reduce peak overshoot of DC link
overshoot so that steady state is obtained soon.