首页|阵列脉冲远场涡流检测的信号降噪方法

阵列脉冲远场涡流检测的信号降噪方法

扫码查看
针对普通脉冲远场涡流检测(PRFECT)无法定位管道缺陷周向位置的缺点,通过增加接收线圈数量,改变接收线圈和激励线圈的相对位置,设计了一种新型阵列式脉冲远场涡流检测探头.并且针对检测出的信号微弱且混杂噪声干扰的问题,提出了一种连续变分模态分解(SVMD)和奇异值分解(SVD)联合的信号降噪方法.首先将信号通过连续变分模态分解分解为一系列模态函数,然后通过皮尔逊相关系数筛选出用于重构的信号分量,再使用奇异值分解降噪方法对这些分量进行降噪,将降噪后的分量叠加可以得到重构信号,仿真和实验结果证明,新探头可以有效定位管道缺陷位置,新算法可以将关键信号降噪误差比降低至9.30,相较于目前算法有明显性能提高.
Array pulsed remote field eddy current testing signal denoising method
To address the limitations of conventional pulsed remote field eddy current testing(PRFECT)in accurately determining the circumferential position of defects,we propose an innovative array-based PRFECT probe.This method involves increasing the number of receiving coils and modifying the relative positions of the receiving and excitation coils to enhance defect localization capabilities.To address the issue of weak signals detected,a hybrid signal denoising technique combining successive variational mode decomposition(SVMD)and singular value decomposition(SVD)is introduced.Initially,the signal is decomposed into a series of modal functions using SVMD.Subsequently,components for reconstruction are selected based on the Pearson correlation coefficient.The retained components are then denoised using the SVD method,and these denoised components are superimposed to reconstruct the signal.Both simulations and experimental results demonstrate that the proposed novel probe effectively locates defect positions,The proposed algorithm can improve the Denoising Signal-to-Noise Ratio of the measured key signal to 9.30,better than traditional algorithm.

PRFECTsignal denoisingSVMDSVD

朱冠诚、刘大生

展开 >

上海交通大学电子信息与电气工程学院 上海 200240

脉冲远场涡流 信号降噪 连续变分模态分解 奇异值分解

2024

电子测量技术
北京无线电技术研究所

电子测量技术

CSTPCD北大核心
影响因子:1.166
ISSN:1002-7300
年,卷(期):2024.47(19)