Torque ripple minimization for PMSM using voltage matching circuit and neural network based adaptive state feedback control

In this paper a new neural network based adaptive state feedback controller used to torque ripple minimization for PMSM is presented. A simple adaptation algorithm is proposed to determine an appropriate value of inverter DC voltage and, in consequence, non-linear coefficients of the controller. The novelty of the proposed approach lays in precisely control the level of inverter DC voltage required for proper operation of the motor as well as in neural network based approximation of state feedback controller coefficients. An additional voltage matching circuit with buck converter and state feedback controller is depicted. Simulation studies illustrate significant torque ripple reduction for the proposed approach in comparison to motor fed by 2-level and 3-level inverters.