RBF neural network disturbance observer-based backstepping boundary vibration control for Euler-Bernoulli beam model...

Area of Science:

• Mechanical Engineering
• Control Systems
• Applied Mathematics

Background:

• Euler-Bernoulli beam systems are susceptible to external disturbances and input saturation, compromising their stability and performance.
• Existing vibration control methods often struggle to effectively address both boundary disturbances and actuator constraints simultaneously.

Purpose of the Study:

• To develop an advanced vibration control strategy for Euler-Bernoulli beam systems facing external disturbances and input saturation.
• To enhance the robustness and stability of flexible beam systems through adaptive control techniques.

Main Methods:

• Derivation of a nonlinear partial differential equation (PDE) model using Hamilton's principle.
• Design of an adaptive radial basis function (RBF) neural network-based control law for estimating boundary disturbances.
• Application of the backstepping approach combined with a hyperbolic tangent function to handle input saturation.

Main Results:

• The proposed adaptive RBF neural network controller effectively estimates and compensates for boundary disturbances.
• The hyperbolic tangent function successfully enforces input constraints, preventing actuator saturation.
• The backstepping framework ensures the stability and convergence of the closed-loop system under saturation.

Conclusions:

• The developed vibration control methodology offers a superior approach for managing disturbances and input saturation in Euler-Bernoulli beam systems.
• The integration of RBF neural networks and backstepping control provides a robust solution for complex dynamic systems.