Simulation of neural networks to sensorless control of switched reluctance motor

Neural networks have been applied to two aspects of sensorless switched reluctance motor operation. First a neural network is trained to predict position from inductance and phase current data and thereby eliminate the position sensor. Second, a neural network is trained to provide a current reference that minimises torque ripple. Torque ripple minimisation is achieved without a torque sensor. A model built in Matlab is used to simulate the system and show successful operation provided the training data is well chosen.