论文部分内容阅读
超声TOFD检测信号是一种非线性、非平稳的信号,常受噪声干扰甚至因材料的厚度过小导致波形混叠,故较难分析出各混叠回波的精确到达时间。本文应用一种基于经验模态分解(EMD)的方法,对超声TOFD检测回波到达时间进行准确识别。该方法首先把采集到的原始回波信号进行经验模态分解,得到若干个固有模态函数,通过计算出一个能量临界值与固有模态函数的能量相比较,恰当地选择其中几个分量进行重构信号,将其进行H ilbert变换得到该信号的峰值包络,则该包络的峰值所对应的时间表示为各个回波的精确到达时间。最后通过超声TOFD实验信号验证了该方法的可行性和正确性。
Ultrasound TOFD detection signal is a non-linear, non-stationary signal, often by the noise interference and even because the material thickness is too small lead to waveform aliasing, it is more difficult to analyze the exact arrival time of each aliasing echo. In this paper, a method based on Empirical Mode Decomposition (EMD) is applied to accurately identify echo arrival time in TOFD detection. In this method, the original echo signals collected are empirically mode decomposed to obtain several intrinsic mode functions. By comparing the energies of an energy critical value with the intrinsic mode functions, some of these components can be properly selected Reconstruct the signal and perform Hilbert transform to obtain the peak envelope of the signal, then the time corresponding to the peak of the envelope is expressed as the exact arrival time of each echo. Finally, the feasibility and correctness of this method are verified by the ultrasonic TOFD signal.