Visual Feedback Vibration Control of Flexible Hinged Plate System Based on Reinforcement Learning Algorithm

Zhi-cheng Qiu,Yi-hong Liu
DOI: https://doi.org/10.1016/j.ymssp.2024.112005
IF: 8.4
2025-01-01
Mechanical Systems and Signal Processing
Abstract:In order to simulate the bending and torsional coupled vibration of solar panels under external disturbances, an experimental system of a fixed flexible hinged plate is established. The dynamic model is established by finite element method (FEM), and the model parameters are accurately identified by wavelet transform and simulated annealing (SA) optimization method. Two laser displacement sensors and binocular vision are combined to detect the bending and torsional vibrations and used for feedback control. A combined piezoelectric (PZT) control scheme is designed to suppress the bending and torsional vibrations of the flexible hinged plate. A vibration control method based on divergence augmented policy optimization (DAPO) reinforcement learning (RL) for the fixed flexible hinged plate is developed to train the modal controllers. Simulation and experiments are carried out to verify the effectiveness of the applied vibration control scheme and algorithm. The simulation and experimental results show that DAPO RL modal controllers have better vibration suppression effect than large gain proportional derivative (PD) controllers, especially for small amplitude residual vibration.
What problem does this paper attempt to address?