High resolution rotor angle estimation for permanent magnet synchronous machines with Hall effect position sensors using neural networks

A permanent magnet synchronous motor (PMSM) with sinusoidal flux distribution is often commutated using discrete rotor position feedback from Hall sensor devices. The use of such discrete rotor position feedback usually means a six-step current waveform must be used to excite the stator. However, the application of non-sinusoidal currents to the stator results in less than optimal operating conditions, such as torque ripple and reduced efficiency. In this paper, a pseudo-sensorless position estimator is used together with Hall sensors to provide sinusoidal current excitation in place of the six-step strategy. A rotor position estimator based on the machine model is designed and simulated. In addition, a recurrent neural network is constructed and trained to perform the same task. Performance evaluation of the rotor position estimator in a PMSM drive is provided through simulation.