论文部分内容阅读
为了更准确地判断简支梁中裂纹的损伤位置和损伤程度,使用模态固有频率的变化率判断出简支梁中裂纹的存在,通过模态振型的变化判别出其损伤位置,并利用基于模态振型的BP神经网络理论,对有限元数值分析得到的模态振型差作为输入向量进行网络训练。结果表明:通过固有频率的下降和振型的变化可以快速地判断裂纹的位置,BP神经网络预测值与理论值误差非常小,可以很准确地判断裂纹的损伤程度。
In order to more accurately determine the damage location and damage degree of the crack in simply supported beam, the existence of cracks in the simply supported beam is judged by the rate of change of modal natural frequency, and the damage location is determined by the change of mode shape. Based on the modal modal BP neural network theory, the modal shape difference obtained from finite element numerical analysis is used as input vector to train the network. The results show that the position of crack can be quickly judged by the decrease of natural frequency and the change of mode shape. The prediction error between BP neural network and theoretical value is very small, and the damage degree of crack can be judged accurately.