Abstract:

In view of the requirements of complex working environment and high precision welding in pool welding process of nuclear power plant, gradient descent method is used to update parameters of parameter estimator step by step, so as to realize the adaptive adjustment of parameters, and improve the performance and robustness of adaptive control. A MRAC-VSIC(Model Reference Adaptive Control-Variable Stiffness Impedance Control) based on gradient descent method is proposed by combining model reference adaptive control with variable stiffness impedance control. It is used to adjust the stiffness parameters in the adaptive compliance controller. The closed-loop stability of MRAC-VSIC method is analyzed by using Lyapunov stability theory, and the stability of the closed-loop system is proved. The simulation results show that the proposed method can improve the dynamic response ability of the robot and the robustness of the control system, reduce the steady-state force error, and effectively improve the control precision and stability in the welding process.

Key words: nuclear power pools, underwater welding robot, impedance control, model reference adaptation, gradient descent method, parameter estimator, end effector, force tracking