中文摘要：

声发射信号是表征材料内部损伤状态的重要指标,由于源信号无法直接获取,研究时通常采用其传播一段距离后失真的信号,这对准确判断材料损伤状态带来了影响,为了降低传播路径对声发射源信号引起的衰减和失真带来的不利影响,用压电陶瓷产生的脉冲信号求解传递通道频响,采用神经网络方法设计了具有任意幅频响应的FIR滤波器,经过试验对比发现,发现该方法能够一定程度地恢复声发射信号在传播过程中衰减的高频成分,为材料损伤状态的准确判断提供参考。

英文摘要：

Acoustic emission signals is an important basis for characterizing the state of material internal damage, since the source signals cannot directly obtain, and it is the signal after propagation distance generally distorted signals, which will affect the identification of the material damage state. In order to reduce the adverse effects of the propagation path of the acoustic emission source signal attenuation and distortion caused, it uses piezoelectric to generate a pulse signal, and solve delivery channel freqnency response, using neural network designed FIR filter with arbitrary amplitude-frequency response. It is found, this method can restore some degree of acoustic emission signal attenuation during propagation of high frequency components, and it gives references to accurately determine the material damage state.