Trajectory Planning of Permanent Magnet Synchronous Motor Based on Improved Bidirectional Long Short Term Memory Network

Because the operation cycle of the CNC system is much longer than the control cycle of the permanent magnet synchronous motor (PMSM), it is necessary to do high-precision interpolation of the rotation Angle of the PMSM rotor on the position instruction sequence given by the CNC system. At the same time, to ensure the smooth transition of PMSM acceleration, this paper proposes a Bi-LSTM based on attention mechanism (Bi-LSTM-Att) interpolation model. The model introduces the attention mechanism optimization interpolation process, and uses the bidirectional long short-term memory network to analyze the time series data from the front and back directions with the global trend. The evaluation indexes of model verification results and simulation results show that Bi-LSTM-Att has smaller interpolation errors, higher interpolation accuracy and more stable control effect than other interpolation models.