基于S变换和时频谱空间滤波的超声缺陷回波检测方法

发布时间：2018-03-28 20:25

本文选题：超声　切入点：S变换　出处：《振动与冲击》2017年22期

【摘要】：缺陷回波的检测是超声探伤的一项重要内容,为减弱噪声的影响准确检测缺陷回波,提出基于S变换时频分析和时频谱空间滤波的信号处理方法。讨论了高斯回波模型下的到达时间和中心频率与S变换时频谱的关系,说明了利用S变换时频谱幅值矩阵的极值提取回波到达时间和中心频率的合理性;为检测回波,首先对原始信号作S变换,然后对得到的时频谱幅值矩阵应用最大熵法自适应选择去噪阈值,对S变换时频谱作空间滤波完成降噪;从降噪后的区域中提取反映缺陷的到达时间和中心频率;对降噪后的时频谱作S逆变换,获得缺陷回波明显的时域信号。仿真研究表明,基于S变换和时频谱空间滤波的方法能够有效去除噪声,检测回波。棒材试块的实验结果同样表明了该方法在缺陷检测上的有效性。
[Abstract]:The detection of defect echo is an important part of ultrasonic flaw detection. This paper presents a signal processing method based on S-transform time-frequency analysis and time-spectrum spatial filtering, and discusses the relationship between the time of arrival and center frequency under Gao Si's echo model and the S-transform time-frequency spectrum. The rationality of extracting the arrival time and center frequency of echo by using the extreme value of spectrum amplitude matrix in S transform is explained. In order to detect the echo, the original signal is first transformed by S transform. Then the maximum entropy method is used to select the denoising threshold of the time-spectrum amplitude matrix, and the S-transform spectrum is filtered by spatial filtering, and the time of arrival and the center frequency of the defect are extracted from the de-noised region. The time-domain signal with obvious defect echo is obtained by S-inversion of the time spectrum after noise reduction. The simulation results show that the method based on S-transform and time-spectrum spatial filtering can effectively remove the noise. The experimental results of echo and bar test also show that the method is effective in defect detection.
【作者单位】：浙江大学流体动力与机电系统国家重点实验室;北京航天发射技术研究所;
【基金】：浙江省自然科学基金(LY14E050013) 浙江省公益技术研究工业项目(2015C31052) 重庆齿轮箱有限责任公司"海面平台洋流发电装备研制与开发"
【分类号】：TB553

【相似文献】