Title :
Voltage Losses Compensation Using Artificial Neural Network for Estimation Nonlinear Characteristic of Switches
Author :
Pothi, N. ; Premrudeepreechacharn, S. ; Rakpenthai, C.
Author_Institution :
Naresuan Univ. Phayao, Phayao
Abstract :
This paper presents the voltage losses compensation technique for three-phase inverter by considering nonlinear characteristic of switches, dead time, and turn on/off delay. An artificial neural network (ANN) is employed to estimate the nonlinear characteristic of switches. The compensation method improves the modulation index of the open-loop Space Vector Pulse Width Modulation Voltage-Source Inverter (SVPWM-VSI) performing on the constant V/f control technique. DSP TMS320LF2407 was used to control the effect of the voltage losses on the efficiency. The principle is explained qualitatively and extensive experiments have been carried out to verify the propose method.
Keywords :
PWM invertors; neural nets; power engineering computing; switching convertors; ANN; DSP TMS320LF2407; artificial neural network; estimation nonlinear characteristic; modulation index; open-loop space vector pulse width modulation inverter; switches; three-phase inverter; voltage losses compensation; voltage-source inverter; Artificial neural networks; Delay effects; Delay estimation; Diodes; Open loop systems; Pulse width modulation inverters; Space vector pulse width modulation; Switches; Temperature; Voltage control; SVPWM; modulation index; nonlinear characteristic of switches; voltage compensation;
Conference_Titel :
Power Electronics and Drive Systems, 2007. PEDS '07. 7th International Conference on
Conference_Location :
Bangkok
Print_ISBN :
978-1-4244-0645-6
Electronic_ISBN :
978-1-4244-0645-6
DOI :
10.1109/PEDS.2007.4487828