digilib@itb.ac.id +62 812 2508 8800

2008 GL PP HERLIEN D. SETIO 1.pdf
PUBLIC Vika Anastasya Kovariansi

A study on actively controlled responses of non-linear system using Artificial Neural Network (ANN) is presented in this paper. The capability to learn, the ability to generalize to new inputs and the ability to perform fast computations are the major advantages in application of neural network. The responses of the non-linear system, which is governed by Duffing’s equation and driven by a harmonic force, are determined through the utilization of equivalent linearization method. The active force at any step of amplitude is determined by using an optimal control algorithm which is then used as input data for neural network training. The advantage of utilizing the neural network as a neuro-controller is that it can considerably reduce computing steps and time.