Harmonic minimization using SHE-PWN and artificial neural network for transistor clamped cascaded h-bridge multilevel inverter

Moataz M.A, Alakkad (2022) Harmonic minimization using SHE-PWN and artificial neural network for transistor clamped cascaded h-bridge multilevel inverter. Masters thesis, Universiti Teknikal Malaysia Melaka.

	Text (24 Pages) Harmonic minimization using SHE-PWN and artificial neural network for transistor clamped cascaded h-bridge multilevel inverter.pdf - Submitted Version Download (329kB)
	Text (Full Text) Harmonic minimization using SHE-PWN and artificial neural network for transistor clamped cascaded h-bridge multilevel inverter.pdf - Submitted Version Restricted to Registered users only Download (6MB)

Abstract

For more than two decades, inverter have been progressing in different topologies and control strategies with many applications. The inverter can be classified according to several standards such as the topologies structure which can be mainly classified into conventional two-level inverter and multi-level inverter. In general the inverter can be classified into two types which are the current source inverter (CSI) and voltage source inverter (VSI). Nowadays, the research gap of the inverter focuses on the improvement of producing an effective power converter taking into consideration of cost, efficiency and output quality. Recently, many newly developed topologies have been derived from the fundamental topologies of inverter. The transistor clamped cascaded H-bridge multilevel inverter (TC- CHB-MLI) is one of those topologies, which is derived from the CHB-MLI types. The TC- CHB-MLI topology has proved its effectiveness in reducing the required components compared to existing inverters especially in the higher levels. Therefore, the overall cost and complexity are greatly reduced, particularly for higher output voltage levels. The major problem of the multilevel inverter output however is the harmonic contents. Various modulation control strategies have been developed in order to produce high quality output with a minimum distortion. The selective harmonic elimination pules width modulation (SHE-PWM) strategy is considered as fundamental switching frequency modulation (FSF) strategies which is widely accepted in high and medium power applications. The benefit of using the SHE-PWM strategy centered in its ability to provide high control accuracy of MLI. The aim of this thesis is to investigate the performance and effectiveness of the open loop five-levels, nine-levels and thirteen-levels TC-CHB-MLI for single phase power electrical system controlled based on the SHE-PWM strategy combined with artificial neural network (ANN) technique. The switching angles have been optimized by solving the transcendental nonlinear equation of the SHE-PWM strategy by using the ANN technique in order to minimize the harmonic order at the output voltage waveform of the MLI and produces a low total harmonic distortion (THD). This thesis includes investigation using both the MATLAB/SIMULINK simulation and hardware experiment. The results of using ANN technique have been compared with the Newton Raphson (N-R) technique. The proposed ANN technique shows better performance over N-R technique during both of MATLAB simulation and hardware experiment. Compared with the N-R technique, reduction of 7.78%, 19.83% and 20.61% was achieved in the output voltage harmonics for the 5 levels, 9 levels and 13 levels respectively. As for the output current harmonics, reduction of 1.04%, 14.75% and 13.97% was achieved for the 5 levels, 9 levels and 13 levels respectively compared with the N-R technique. The benefit of the proposed technique centered on the reduction of the additional filter before feeding the sensitive load like as the communications systems and medical devices.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Electric inverters, Harmonic (Electric waves), Wavelets (Mathematics)
Subjects:	T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Library > Tesis > FKE
Depositing User:	F Haslinda Harun
Date Deposited:	03 Jul 2023 12:12
Last Modified:	03 Jul 2023 12:12
URI:	http://eprints.utem.edu.my/id/eprint/26880
Statistic Details:	View Download Statistic

Actions (login required)

View Item