An artificial-neural-network-based space vector PWM of a three-phase high power factor converter for power quality improvement

Abstract

In this paper, an artificial neural network (ANN) based space vector pulse width modulation (SVPWM) for a three-phase, two-level high power factor converter is presented. A multilayer feedforward neural network with backpropagation is used. The ANN receives amplitude and angle of the reference vector to calculate the duty cycles of various space vectors in different sectors which can be used for the generation of PWM pulses for the control of converter. The ANN significantly reduces the computational efforts of the modulation technique and makes the implementation of space vector modulation algorithm very fast without losing precision compared to the conventional SVM algorithm implementation using look up table. The performance of the converter in terms of its power quality on the input and output side is investigated through computer simulations and the results have been found excellent. A mathematical model of the rectifier is also developed. A comparative analysis of the converter using space vector modulation without and with ANN implementation is also presented to validate the usefulness of the ANN implementation of space vector modulation. The principle can be extended to multilevel converters with certain modifications.