A computationally efficient torque and flux ripple reduction by active vector optimization approach using direct torque control–based model predictive torque control for matrix converter–fed permanent magnet synchronous motor

Abstract

The Matrix Converter–fed Finite Control Set–Model Predictive Control is an efficient drive control approach that exhibits numerous advantageous features. However, it is computationally expensive as it employs all the available matrix converter voltage vectors for the prediction and estimation. The computational complexity increases further with respect to the inclusion of additional control objectives in the cost function which degrades the potentiality of this technique. This paper proposes two computationally effective switching tables for simplifying the calculation process and optimizing the matrix converter active prediction vectors. Here, three prediction active vectors are selected out of 18 vectors by considering the torque and flux errors of the permanent magnet synchronous motor. In addition, the voltage vector location segments are modified into 12 sectors to boost the torque dynamic control. The performance superiority of the proposed concept is analyzed using the MATLAB/Simulink software and the real-time validation is conducted by implementing in the real-time OPAL-RT lab setup.