Journal of Electrical & Electronic Systems

GNN-Based Voltage Source Converter Controlled Synchronous Reluctance Motor with Reduced Torque Ripple

Abstract

Tausif Ahmad*, Juhi Chaudhary and Chandra Bhushan Mahto

This paper introduces a Generalized Neural Network (GNN) based voltage control strategy for a Voltage Source Converter (VSC) fed Synchronous Reluctance Motor (SyRM) drive system with reduced torque ripples. The aim is to provide an economical, concise, and reliable control method for VSC-controlled SyRM applications. The algorithm utilizes a Generalized Neuron (GN) model created with fuzzy compensatory operators to handle the dynamic capabilities of the drive, thus improving training time. Furthermore, the neuron is decomposed in the synchronous reference frame currents to generate a more accurate set of reference inputs for the active currents component for converter switching. Additionally, the approach avoids conventional pulse-width modulation, resulting in less computational burden on the control. This enables the regulation of converter voltage within a specified time frame to minimize torque ripples. A simulation model of the drive system was developed and evaluated to assess the effectiveness of the proposed method. Both experimental and simulation results demonstrate that the drive system offers rapid speed response and effective disturbance rejection while minimizing torque ripple. The results indicate that the suggested GNN algorithm is efficient and offers technical advantages.