摘要
结构材料的分散性使得裂纹长度评定时误差较大,为此提出了基于高斯过程模型的裂纹长度评定方法。采用主动Lamb波监测技术对疲劳裂纹扩展进行监测,从所测的传感信号中提取能够反映裂纹扩展的4种损伤因子,建立基于组合协方差函数的高斯过程先验模型,通过极大似然法完成模型训练,进而实现裂纹长度的评定。在航空结构中常用的LY12-CZ铝合金试件上进行了孔边裂纹的疲劳试验,结果表明该方法能够有效地减小由于结构材料的分散性造成的裂纹长度评定误差。
The dispersion of structure material leads to large error in crack length evaluation,an assessment method for crack length evaluation utilizing Gaussian process （GP） model is proposed to solve this problem.The active Lamb monitoring technology is adopted to monitor the fatigue crack growth; four damage factors are extracted from the measured sensor signal,which can reflect the fatigue crack growth; and the prior GP model based on composite covariance function is established.Maximum likelihood method is used to complete the model training,and then the crack length evaluation is realized.The fatigue test of hole-edge crack was performed on LY12-CZ aluminum alloy specimen commonly used in aerospace structures; the test results show that the proposed method could efficiently decrease the evaluation error of crack length caused by the dispersion of structure material.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2014年第3期580-585,共6页
Chinese Journal of Scientific Instrument
基金
国家杰出青年科学基金(51225502)
国家自然科学基金(51205189)
江苏省高校优势科学建设工程资助项目
关键词
结构健康监测
疲劳裂纹
LAMB波
高斯过程
长度评定
structural health monitoring （SHM）
fatigue crack
Lamb wave
Gaussian process （GP）
length evaluating
