Eyad K. Almaita

Doctoral Dissertation Announcement

---

Candidate: Eyad K. Almaita

Degree of: Doctor of Philosophy

Department: Electrical and Computer Engineering

Title: Adaptive Radial Basis Function Neural Networks-Based Real Time Harmonic Estimation and PWM Control for Active Power Filter

Committee:
Dr. Johnson Asumadu, Chair
Dr. Ala Al-Fuqaha
Dr. Liang Dong

Date: Friday, February 17, 2012 3:30 p.m. to 5:30 p.m.
D-212 Parkview Campus

Abstract:
With the proliferation of nonlinear loads in the power system, harmonic pollution becomes a serious problem that affects the power quality in both transmission and distribution systems. Active power filters (APF) have been proven to be one of the most successful methods for mitigating harmonics problems. So far, different techniques have been used in harmonics extraction and control of APF to satisfy the fast response and the accuracy required by the APF. Neural networks techniques have been used successfully in different real-time and complex situations. This dissertation demonstrates four main tasks; (i) a novel adaptive Radial basis function neural networks (RBFNN) algorithm which can be used in different signal processing or control applications, (ii) Dynamic identification for the total harmonics content in converter waveforms based on RBFNN and p-q (real power-imaginary power) theory, (iii) RBFNN is used to dynamically identify and estimate selective harmonic components in converter waveforms, and (iv) a novel adaptive hysteresis current control algorithm with nearly constant switching frequency.
The proposed RBFNN filtering algorithms are based on a computationally efficient training method called hybrid learning method, which requires negligible training time. Both of the proposed algorithms in this dissertation, adaptive RBFNN algorithm and adaptive hysteresis current controller, are simple, effective, and easy to implement.