SEMI-ACTIVE NEUROCONTROL ON STAY CABLE WITH MR DAMPER

The inverse dynamic behaviour of MR damper is emulated based on neural network theory. In order to improve the calculation and generalization of the neural network model of MR damper, the Levenberg-Marquardt algorithm and Bayes Regularization method are adopted here. With the neural network model of the MR damper, a new control strategy for the stay cables is proposed. A comparison with LQR active control method is also made to check the validity of the proposed control method. Numerical results show that the proposed semi-active neurocontrol method is effective, and its results are comparable to the LQR.