Optimal control of permanent magnet motors using dynamic programming

Firstly, dynamic programming is employed in the development of an optimal torque controller for sinusoidal permanent magnet synchronous motors (PMSM). In particular, a solution that can handle non-zero state tracking is of interest. Secondly, it is shown that the resulting optimal control solution converges to that of linear quadratic regulators (LQR). An LQR is then presented as an alternative control method for PMSM. A comparison of the performance of the LQR with that of traditional proportional and proportional-integral controllers is made using time and frequency domain techniques.