Design of Feed-Forward Neural Network Based On-line Flux Estimator for Sensor-less Vector Controlled Induction Motor Drives

The sensor-less vector controlled induction motor drive requires accurate estimation of speed and flux. The speed estimation depends on the motor flux, which has to be measured or estimated. The flux measurement is difficult and expensive and hence generally estimated. Conventional voltage model equations for flux estimation encounter major drawbacks at low frequencies/speed. Neural network based estimators for flux provides an alternate solution with added robustness to motor parameter variation. Hence such model assumes importance for real time/on-line estimation. The required neural model uses feed-forward architecture whose design is more an art than a science. This paper proposes a novel design methodology to optimize both performance and cost. Using the proposed methodology an on-line flux estimator is designed and validated through simulation. The results obtained are presented.