小波神经网络在基桩缺陷诊断分析中的应用

Application of Wavelet and Neural Network to Defect Diagnosing of Piles

将小波分析作为神经网络的前置处理手段,从基桩动测信号小波变换的分量中提取特征,然后将这些特征输入人工神经网络进行训练和分类,进而实现基桩缺陷位置和程度的诊断。仿真试验的结果表明,该方法对桩身完整性的评价是快速有效的,特别是对于多个缺陷的判别较其他方法具有优越性,在此基础上进行了模型桩的现场试验研究。

A diagnostic method for defects in piles combing with the advantage of wavelet and neural network is presented in this paper.As a fore processing medium,wavelett transform method is used to extract the characteristics which reflect the information of defects.These characteristics are fed into the neural network as the input patterns for training and classifying.Farther,it can be used to diagnose the location and magnitude of defects.The results of simulated test indicated that this method can be applied to the identification and diagnosis of defects with high efficiency and accuracy,especially for testing of the piles which contains more than one defect.The field tests verify the feasibility of the method.