Five-level DTC-ANN with Balancing Strategy of DSIM

Abstract

This paper presents an artificial neural networks controller devoted to improve the performance of direct torque control strategy of double star induction machine fed by two five-level diode-clamped inverters. The resulting control scheme presents enough degrees of freedom to control both torque and flux with very low ripple and high dynamics. Unfortunately, the diode-clamped inverter has an inherent problem of DC-link capacitors voltages variations. To overcome this problem, an artificial neural network based direct torque control with balancing strategy is proposed to suppress the unbalance of DC-link capacitor voltages. Simulations results are given to show the effectiveness and the robustness of the suggested control method.