DoctorateOpen Access

Design of cascaded three-level inverter with space vector pulse width modulation using artifical neural network in region and sector determination

2012
0 views
0 downloads
Advisor: Prof. Dr. Sedat Sünter

Abstract (EN)

Direct current is converted to alternating current to meet the requirements of some AC systems. For this reason, inverter circuits were developed. Because the switching losses increase with use of inverters at high power levels, multilevel inverter topologies have been developed. The output voltage waveforms in multilevel inverters can be generated at low switching frequencies with high efficiency and low harmonic distortion.The aim of using multilevel inverters is to control the output frequency and voltage magnitude with high quality of sinusoidal output waveforms. Space vector pulse width modulation technique is one of the most popular technique gained interest recently because; it offers better harmonic performance and can more conveniently handle overall switching patterns and constrains and it is simple to implement using DSPs.In this thesis, modeling and simulation of the cascaded multilevel (three level) inverter with separated and isolated DC sources have been performed by using MATLAB/Simulink package program. First, sector and region of three level SVPWM technique have been identified with using artificial neutral network. Then the conventional three level space vector pulse width modulation technique has been simulated both an RL load and motor load. The inverter system has been designed and experimental results were taken for various operating and load conditions.Keywords: Multilevel Inverters, Three Level Inverters, Space Vector PWM

Author

Ayşe Kocalmış Bilhan

How to Cite

Ayşe Kocalmış Bilhan (Doctorate thesis). Design of cascaded three-level inverter with space vector pulse width modulation using artifical neural network in region and sector determination, 2012, Fırat University.

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Fırat University