Growing Neural Gas-based Maximum Torque per Ampere (MTPA) Technique for SynRMs

This paper proposes a maximum torque per ampere (MTPA) technique specifically developed for Synchronous Reluctance Motors (SynRMs). The proposed MTPA is based on a self-organizing artificial neural network, called Growing Neural Gas (GNG). The GNG gas been trained in order to learn the real maximum torque per ampere points of the SynRM under test. The proposed MTPA has been tested experimentally on a suitably developed test set-up. The obtained experimental results clearly highlight a significant increase of maximum producible torque, with respect to the previously developed MTPA techniques.