Abstract
This paper presents two different response evaluators which evaluate the response of the system and decide the control input to the system. The characteristics of the response evaluators were designed based on the source region and ground motion of earthquakes. They were applied to a base isolated structure with a semi-active oil damper whose damping coefficient has large or small values. The response evaluators were designed by a multilayer neural network which uses the building response and the peak ground velocity as input data and outputs a switching signal for the damping coefficient as output data. A genetic algorithm was used to adjust the neural network parameters which determine the performance of the response evaluators. A computer simulation was carried out using several kinds of earthquake motions, and absolute acceleration and displacement of each story, which are in a trade-off relationship with each other, are reduced to intermediate values of the response in the case where constant damping coefficients are large or small. From the result of the simulations, the control effect of the response evaluators was verified.