Paper Title
An Alternative Technique To Minimise Thd in Sinusoidal Pwm Controllers with Ann and Hybrid LC-RLC filter
Abstract
The enormous growth in Artificial Neural Network (ANN) has increased the rate at which researchers investigate
how the field can be used to enhance the performance of power converters. In recent years, several research works have been
conducted to modify various power electronic devices with ANN. This paper also aims to contribute to the enhancement of
power electronic converters with ANN. As a result, the work proposes a Sawtooth-Sinusoidal PWM controller to control a
Voltage Source Inverter (VSI). The proposed technique was made feasible by incorporating an ANN, based on Feed-
Forward Network, into the traditional Sinusoidal PWM. The ANN functioned to train the reference signals together with the
carrier signal of the controller. To further reduce the THD, a hybrid LC-RLC passive low pass filter was connected between
the VSI and the Load. MATLAB/Simulink was applied in designing both the proposed and the conventional techniques to
observe the viability of the suggested idea. The results were observed and presented in this paper. Based on the result, it was
deduced that the proposed Sawtooth-Sinusoidal PWM can be replaced with the conventional Sinusoidal PWM technique.
Keywords - Artificial Neural Network, Voltage Source Inverter, Passive Filter, Sinusoidal PWM