Neural Network Observer Design for Sensorless Control of Induction Motor Drive

In this paper a new neural network observer design is presented for a three phase Induction Motor drive. Neural Network observers have the advantage that these do not require the exact mathematical knowledge of the system. As such the observer design is easy, parameter independent and accurate. The speed control is achieved by Fuzzy Logic controller. Fuzzy logic controller is useful for non-linear systems in particular leading to robust control. Simulation is carried out in MATLAB/Simulink on a three phase Induction Motor drive in order to verify the physical behaviour of the system when subject to disturbances, change in load and speed.