A new training algorithm that does not require desired response is proposed. The algorithm uses cost function as training criterion. A sensitivity evaluation algorithm is also proposed. Therefore, emulator neural network is not needed any more for the training of controller neural network. An active mass driver (AMD) system on the top roof is used as an exciter. The control signals are made by a neural network that is trained by minimizing a sub-optimal performance index. The performance index is a function of both the output responses and the control signals. Structure having nonlinear hysteretic behavior is also trained and controlled by using proposed control algorithm. Both the time delay effect and the dynamics of hydraulic actuator are included in the simulation. Example shows that neuro-controller can be trained by proposed training algorithm and with the sensitivity data found by proposed evaluation algorithm.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £52 +P&P)