1995 年 61 巻 586 号 p. 2253-2258
This paper describes a convenient method of identification of non-linear vibration system by a neural network using time series data. The neural network using suitable system of inputs and outputs are applicable for system identification of non-linear vibration by learning of impact response. By learning of only time series datas of impact response, it is possible to predict both forced vibration response and impact response in time domain by the neural network. The effectivenss of the method presented is demonstrated by numerical simulations for Duffing system and piece wise linear system.