Control method of brushless direct current motor based on neural network feedforward compensation

The invention discloses a control method for a brushless direct current motor based on neural network feedforward compensation. The control method comprises the following steps of taking the brushless direct current motor as a second-order nonlinear system; calculating a position error according to the difference between the expectation of the control object and the current rotating speed or current of the motor; an equation of a closed-loop control system being obtained by designing an input u of the system as a PD control rate based on feed-forward and compensation; establishing a three-layer radial basis function neural network, and performing approximation on an expression of disturbance of environmental factors to the motor by using the three-layer radial basis function neural network; the output of the RBF neural network being used as the disturbance compensation of the brushless direct current motor; the input of the system being designed to be a PD adaptive control law of a brushless direct current mo