Three-Level Inverter Performance Using Adaptive Neuro-Fuzzy Based Space Vector Modulation

Abstract

Space vector modulation is an optimal pulse width modulation technique in variable speed drive application. This paper presents Adaptive Neuro-fuzzy based space vector modulation technique for a three level inverter. It uses uses hybrid learning algorithm (combination of back propagation and least square methods) for training due to this the required training error is obtained with less number of epoches compared to other techniques like Neural, fuzzy etc. The proposed scheme uses the d-axis and q-axis voltages information at the input side and the corrected two-level duty ratios for switching pulses, two-level index are generated as output. The performance measure in-terms of the total harmonic distortion (THD) of inverter line-line voltage has been evaluated with Adaptive Neuro-fuzzy based system is compared with the conventional based SVM method.