水冷壁管磁性检测信号的小波处理方法

针对锅炉管道磁性无损检测信号的特性,研究了运用多尺度小波分析进行信号奇异性识别、特征提取和检测数据的去噪压缩方法.根据锅炉管道的两类不同缺陷（表面缺陷和壁厚减薄）,分别选用偶对称和奇对称小波函数进行奇异性分析,较好地实现了缺陷识别和定位,并通过选择具有较好正交性和规则性的双正交小波基来进行检测数据的压缩处理,为电站锅炉管道检测过程中“海量数据”的实时传输和存储找到一种可行的方法.

Some Methods of Signal Processing in Magnetic Non-destructive Testing for Boiler Tube Based on Wavelet Transform

Based on the characteristics of the boiler tube magnetic NDT signals,some signal processing methods for singularity detection,feature extraction,denoising and compressing by multi-scale wavelet transform(WT) are discussed. According to the two different faults(the surface fault and the wall loss),the even and odd symmetry function are chosen to analyze the singularity point respectively,and the faults are recognized and located. And the bi orthogonal wavelet,with good orthogonality and regularity,is applied to compress and reconstruct the signal. It eliminates measure noise effectively,and provides a feasible approach to implement the data real time transmission and storage.