ANN Based Speed Estimator for Vector Controlled Induction Motor Drive

The speed estimation in Field Oriented Control (IFOC) also known as vector control of induction motor drives (IM) is based on the reduced order of the IM model, the rotor fluxes are estimated using only the stator currents and voltages. The speed estimator, in the form of a single stage neural network using back propagation techniques is presented. The neural network is able to accurately estimate these speed of an induction motor. The observations indicate that the neural network speed estimation may be a feasible alternative to other speed estimation methods like programmable DSP kit. The comparative performance of both has been presented in this work with the help of a practical three-phase induction motor drive.

The present approach avoids the use of flux and speed sensors which increase the installation cost and degrade the mechanical robustness. The ANN based estimator is found to be a very useful technique to obtain high performance speed response.