Hydraulic servo system neural adaptive control method based on state observer

Abstract

The invention belongs to the technical field of fluid transmission and control, and relates to a state observer-based neural adaptive control method for a hydraulic servo system, which comprises the following steps of: S1, modeling by using a typical hydraulic servo position control system, and constructing a dynamic model of the hydraulic servo position control system; s2, establishing a nonlinear expansion state observer according to the dynamic model of the hydraulic servo position control system, and determining a to-be-tested state and first disturbance; s3, estimating second disturbance of the hydraulic servo position control system by using the adaptive neural network; s4, according to the to-be-tested state, the first disturbance and the second disturbance of the hydraulic servo position control system, a neural adaptive controller based on a retreating control framework is designed; and S5, verifying the stability of the closed-loop hydraulic servo position control system, and completing neural adaptive control of the hydraulic servo position control system. According to the method, the problems of strong nonlinearity, parameter uncertainty and internal and external disturbance of the hydraulic servo position control system are solved, and the control precision and robustness of the hydraulic servo position control system are improved.

Images