The Proposal of discriminating stable control bandwidth using ANN in sensorless speed control system for PMSM

In the sensorless control of a permanent magnet synchronous motor (PMSM), there is a method of determining stability by analyzing the pole placement of the closed-loop transfer function regarding the influence of the three control bandwidths of current control, speed control, and sensorless control on stability. However, due to various effects, there is a problem that the theoretical analysis and the range of the control bandwidth that can be stably driven in the actual machine do not match. In this paper, we propose a method using ANN (Artificial Neural Network) in order to determine the condition of the stable control band more accurately.