ANFIS Based Speed Sensorless Induction Motor Fed by 3-Level NPC Inverter

The use of an adaptive Neuro-fuzzy inference system (ANFIS) for speed estimation of an induction motor is presented in this paper. The ANFIS (neuro-fuzzy inference adaptive system) speed observer was created in response to the limitations of mechanical sensors. It is based on an artificial intelligence technique that combines the ideas of fuzzy inference systems and neuron networks. Since there is no need for a speed sensor, the ANFIS rotor speed estimator is simple to use in practice and relies only on readily available, measurable stator quantities (voltages and currents). This lowers costs. Furthermore, the vector controlled induction motor with stator field orientation (SFO) is also covered in this work. It is commonly known that the vector control approach relies on the concurrent calculation of the flux vector's magnitude and argument. By providing decoupling between an induction motor's torque and flux, this control method effectively addresses the intricate control challenge of these machines. On the other hand the multilevel inverter has been proposed as an alternative solution for applications involving high power and medium voltage. This paper presents also specifically the NPC topology and outlines the most known control technique SVM technology space vector modulation.