基于小波包能量谱的管道缺陷磁记忆检测信号特征研究

Study on Magnetic Memory Testing Signal Feature of Pipeline Defect Based on Wavelet Packet Energy Spectrum

现有磁记忆检测技术判定准则，只能指示应力集中位置，无法进一步获取应力集中信息。为获取应力集中信息，提出一种基于小波包能量谱的磁记忆信号分析方法，进行试件拉伸试验。拉伸应力为200MPa时，信号小波包能量谱分布较为均匀，各频带能量占总能量之比均小于15％，不存在集中分布的频带范围。拉伸应力为410MPa时，信号小波包能量谱最大值分布在1，3，4频带，1。4频带能量之和占总能量的73．8％，小波包能量主要集中在低频段。试件屈服后，信号小波包能量谱最大值分布在1，2频带，能量谱分布极为分散，能量主要集中在低频段的1，2频带，1-3频带能量之和占总能量的87．3％。管道试件应力集中程度与磁记忆信号的小波包能量谱分布特征有关，应力集中程度越低，小波包能量谱分布越均匀；应力集中程度越高，小波包能量谱分布越集中，能量主要向低频段集中。

At present, the judging principle of magnetic memory testing technology （MMT） can only indicate the stress concen- tration zone, but can not get information of the stress concentration further. In the research, to get information of the stress concentra- tion zone, an MMT signal analysis method based on wavelet packet energy spectrum is proposed. The specimen tension load experi- ment shows that when the tension load is P=200 MPa, the signal wavelet packet energy spectrum distributes uniformly, the percent- age of separate energy is lower than 15%, and it has no concentration zone. When the tension load is P=410 MPa, the maximum wavelet packet energy locates in spectrum 1,3 and 4, the percentage of spectrum 1 - 4 energy to total energy is 73.8%, the distribu- tion of wavelet packet energy spectrum is separate, and the main wavelet packet energy locates in low frequency area. After the spec- imen yields, the maximum wavelet packet energy locates in spectrum 1 and 2, and the percentage of spectrum 1 - 3 energy to total energy is 87.3 %. The distribution of wavelet packet energy spectrum is more separate, and the main wavelet packet energy locatesin spectrum 1 and 2. The stress concentration degree is related to the MMT signal＇ s wavelet packet energy spectrum distribution characteristic. If the stress concentration degree is low, then the wavelet packet energy spectrum distribution is equaJ,; if the stress concentration degree is high, then the wavelet packet energy spectrum distribution is concentrated, and the main energy is concen- trated in low frequency area.