Brushless dc motor (BLDCM) is a multivariable
nonlinear system, whose current ripple is very obvious
due to the non-linearity of armature reaction, phase
conversion and many other causes. The application in
servo-drive system of high precision is restricted, just
because it is difficult to get ideal square wave current
under actual control. The conventional PID control
selects error as control signal directly, resulting in the
conflict of fast response and overshoot. What’s worse,
its intrinsic phase-shifting effects are harmful to
BLDCM control which requires control back EMF
(ElectroMotive Force) and current in-phase. In order to
solve this problem and enhance the motor performance,
neural networks, Fourier analysis, prediction control
and hysteresis control are applied to BLDCM and some
effects have been achieved[1-4].
The active disturbance rejection controller (ADRC)
is composed of three parts: tracking differentiator (TD),
extended state observer (ESO) and nonlinear state error