CAJ | 학술논문

讨论了在输入未知时结构模态参数基于频率响应函数的神经网络识别方法，并研究了不同的噪声水平和网络输入层结点数目的变化对网络输出误差的影响.讨论了网络对不同阶数模态参数的识别精度.数值结果表明该方法是可行的.
토론료재수입미지시결구모태삼수기우빈솔향응함수적신경망락식별방법，병연구료불동적조성수평화망락수입층결점수목적변화대망락수출오차적영향.토론료망락대불동계수모태삼수적식별정도.수치결과표명해방법시가행적.
This paper proposes a method for identifying structural modal parameters based on neural network. The influence of variety of noise and node number of the input layer of neural network to the error of the network are also taken into consideration.Furthermore, the identification precision of the neural network to different order of the modal parameters is also studied. Numerical results show that this method is very practical.