Abstract: A hybrid modeling method based on neural network (NN) is developed and used to model the hysteretic restoring force of wire cable isolation system for a certain electronic equipment. Firstly, a framework model for the nonlinear hysteretic restoring force is identified by using the experimental data from period loading. Secondly, the remained characteristic of hysteretic restoring, which cannot be parametrically modeled in an easy way, is identified by using NN method through network training. Thus, a hybrid model for the nonlinear hysteretic restoring force can be obtained. The model is then used to predict the dynamic responses of isolation system under harmonic and broad-band random excitations, respectively. The comparative study is also carried out to validate the effectiveness and prediction accuracy of the model. The results show good prediction accuracy with the developed hybrid model.

Key words: Hybrid modeling, Hysteretic restoring force, Neural network, Wire cable isolator