摘要
应用有限元仿真法 ,对带有裂纹损伤钢板结构的裂纹长度和位置对结构不同阶频率和模态的影响进行了分析 ,并与理论结果进行了比较 ,得到了带伤钢板结构动态特性的一般规律 ,针对裂纹损伤与结构动力特性之间存在着的很复杂的尚未明确的非线性关系 ,利用人工神经网络模型对结构裂纹深度和位置进行了诊断和预测的定量和定性研究 ,结果表明 ,可以应用此方法对带伤钢板进行动态特性分析 。
This paper analyzed the influence of the length and location of crack of steel plate on inherent frequencies and modal using FMP and compared with the results of theory. The general regularity was obtained about the dynamical properties of steel plate with crack. The relationship between the crack damage and dynamic properties of the structure was non-linear. Taking the inherent frequency as characteristic parameter of inputs, the depth and position of crack fault in the steel plate was diagnosed and forecasted, applying artificial neural network (ANN). It showed that it is feasible to identify the crack fault in structure using ANN.
出处
《沈阳工业大学学报》
EI
CAS
2002年第3期248-251,共4页
Journal of Shenyang University of Technology
关键词
有限元
动力特性
神经网络
固有频率
诊断
finite element
dynamic property
neural Network
inherent frequency
diagnosis