摘要
多频涡流检测是有效实现多参数检测和干扰抑制的无损检测方法。在谱分析型多频涡流检测中,采用同步方式合成多频激励信号,将涡流检测信号进行放大后,采集进行DFT变换,在频域进行特征提取分析。基于遗传算法,对多频激励信号的初始相位进行优化,降低激励信号的峰值因数,减小驱动电路工作电压范围。对检测信号进行谱分析,通过谱能量的变化对缺陷进行识别,提出谱变化趋势对缺陷进行分类。对于表面下缺陷,采用峰值频率点来近似识别缺陷的深度位置。通过涡流检测试验,验证了所提出方法的正确性。
Multi-frequency eddy current testing（MFECT） is an effective NDT method for multi-parametric detection and interferer elimination.In multi-frequency eddy current testing based on spectrum analysis,its multi-frequency exciting signal was synthesized synchronously.The eddy current testing signal was amplified and sampled,then it was converted to frequency domain by discrete Fourier transform（DFT） and spectrum features could be extracted and analyzed.The initial phase of each frequency component in the exciting signal was optimized based on genetic algorithm（GA）,thus the crest factor was reduced and the work voltage of driving circuit was decreased too.When the testing signal was analyzed in frequency domain,spectrum energy could be adopted to identify defects and spectrum variation trend was proposed to classify defects.Peak frequency point was adopted to identify the depth position of subsurface defects approximately.The eddy current testing experiment results show that the present methods are correct.
出处
《电子测量与仪器学报》
CSCD
2011年第1期16-22,共7页
Journal of Electronic Measurement and Instrumentation
基金
国家"十一五"预研(编号:51317030106)资助项目
关键词
多频涡流检测
谱分析方法
峰值因数优化
缺陷识别
缺陷分类
multi-frequency eddy current testing
spectrum analysis method
crest factor optimization
defect identification
defect classification
