基于自适应滤波的薄层厚度超声测量方法

Ultrasonic measurement method for thin layer thickness based on adaptive filtering

针对应用超声脉冲回波法测量三层结构中的硅橡胶薄层厚度时回波混叠的问题,提出一种基于RLS(recursive least square)自适应滤波的解决方法。该方法将硅橡胶层下界面回波发生混叠的信号作为输入信号,将无硅橡层时的回波信号作为期望信号,通过RLS自适应滤波算法的处理提取出相互分离的有效下界面回波信号,实现硅橡胶薄层的测厚。研究不同滤波器阶数和遗忘因子对信号分离及测厚精度的影响,以输出信号的信噪比为指标选择最优的滤波参数。结果表明:该方法能够有效分离出发生混叠的硅橡胶薄层界下界面回波,能够测量0.15,0.17,0.19,0.21 mm 4种厚度的硅橡胶薄层,对回波部分混叠和完全混叠两种情况均有良好的适用性。

To solve the problem that the thickness of the silicone rubber layer in a three-layer structure can’t be measured due to echo aliasing,an ultrasonic measurement method based on RLS adaptive filtering has been proposed.In this method,the aliasing signal from the bottom interface of the silicone rubber layer was input as the original signal,and the echo signal without silicon rubber layer was input as the desired signal.The effective signal of the bottom interface of the silicone rubber layer was extracted by adopting RLS adaptive filtering,and then the thickness of the thin silicone rubber layer was measured.The effects of the filter order and the forgetting factor on signal separation were studied,and then the optimal filter parameters were determined according to the signal-to-noise ratio of the output signal.The results show that this method can effectively separate the aliasing echoes of the thin silicone rubber layer and that thin silicone rubber layers with thicknesses of 0.15,0.17,0.19,0.21 mm can be measured,and that it is applicable for both partially mixed signals and completely mixed signals.