Introduction  In order to reduce the influence of the positioning accuracy of the electric actuator by the change of its internal parameters， and to reduce the maintenance frequency of the electric actuator.

  Method  After the model of the electric actuator was improved， the influence factors of the steady-state error were obtained through simulation experiments. The compensation loop was further added to reduce the steady-state error. The parameters of the compensation loop were obtained by utilizing the BP neural network algorithm.

  Result  The simulation result shows that the steady-state error of the electric actuator is related to the width of the dead zone of the controller， the backlash width of the reducer and the sign of the input signal. After adding the compensation circuit， the absolute value of the steady-state error can be guaranteed to fluctuate within 0.1% when the controller dead zone width and reducer backlash width change.

  Conclusion  It solves the contradiction between the positioning accuracy and stability of electric actuator， and the problem of positioning accuracy being reduced due to the wear of reducer.