Integral Separation PID Control of Certain Electro-hydraulic Servo System based on RBF Neural Network Supervision

摘要

Electro-hydraulic servo control system is a nonlinear and uncertain system with time-variation parameters and external disturbance. The traditional Proportion Integration Differentiation (PID) controller can hardly control the nonlinear and time-variant systems, whereas the Radial Basis Function (RBF) controller may solve the problem under the condition of the control parameters' reasonable selection. In order to ensure the static and dynamic performance of certain electro-hydraulic servo system, an Integral Separation PID Controller (ISPID) based on RBF Neural Network Supervision (RBFNNS) was proposed. Taking advantage of the ISPID Controller and RBFNNS, optimum control of certain electro-hydraulic servo system was achieved. The results of MATLAB simulation and the prototype tests show that the overshoot of the proposed system is 2.5%, the steady state error of step response is about 0.10～0.91mm, and the error of prototype depth control is no more than ± 10mm. Compared with the traditional PID controller and the RBF controller, the proposed system has advantages of speedy response, smaller overshoot, high steady-state accuracy and strong robustness. It proves that the proposed control scheme is effective and suitable.