Tuning of pi speed controller in direct torque control ofdual star induction motor based on genetic algorithms and neuro-fuzzy schemes

Thanks to the positive characteristics of the double stator machine (DSIM), its high reliability, and reduced rotor torque ripples, it has become among the most important multiphase machines included in industrial applications. This article aims to apply the two techniques of artificial intelligence represented by the adaptive neuro-fuzzy inference systems (ANFIS) and the genetic algorithm (GA) for direct torque control (DTC) of the DSIM to improve the performance of the machine. The ability to learn and the parallelism of operation characteristics made exploiting the GA to control the machine possible instead of using the proportional integral controller (PI). Fixed switching frequency obtained, given with the vector selection table and hysteresis, allowed the inclusion of the ANFIS technique in the DTC strategy. Two -level inverters are included to feed the DSIM. Several results prove that the two techniques applied, ANFIS and GA, improve the quality of the electromagnetic torque and flux and the dynamic responses of the DSIM.