摘要
针对复合材料异常检测或健康监测的问题,对电阻抗谱法与支持向量机相结合应用于飞机复合材料构件健康监测进行了研究。制作了同面2电极传感器和两块碳纤维复合材料样板,设计实验电路测量了两个碳纤维被测样本的电阻抗谱,提取了不同类型的电阻抗特征参数;在Matlab软件平台上利用支持向量机对复合材料健康状态进行辨识,并比较分析了不同类型特征参数下的支持向量机辨识准确率。结果表明,相比于Cole-Cole曲线分段各段电阻抗实部或虚部幅值的平均值和频率——实部或虚部曲线分段线性拟合斜率,选取复合材料样板的Cole-Cole曲线分段线性拟合斜率作为特征参数时,支持向量机具有更高的健康状态辨识准确率。初步实验表明,支持向量机与电阻抗谱法结合起来进行飞机复合材料构件的健康监测是可行的。
Aiming at the problem of composite anomaly detection and health monitoring,the impedance spectroscopy combined with support vector machine（SVM）is applied to aircraft composite materials component health monitoring.A uniplanar dual electrode sensor and two carbon fiber composite samples are fabricated,and an experiment circuit is designed,and two carbon fiber material electrical impedance spectrum is measured.On Matlab software platform,the support vector machine（SVM）are used to identify health status of the composites after extracting the characteristic parameters of the different types of electrical impedance.The analysis of the support vector machine（SVM）identification accuracy under different types of characteristics parameters is carried out.The results show that comparing with the Cole-Cole curve segment real or imaginary part of electrical impedance average,and the frequency-real or imaginary part of electrical impedance piecewise linear fitting curve slope,the support vector machine has higher identification accuracy while the Cole-Cole piecewise linear fitting curve slope is used to be a characteristic parameter.Preliminary experiment shows that the support vector machine（SVM）combined with electrical impedance spectrum method for aircraft health monitoring of composite components is effective.
出处
《压电与声光》
CAS
CSCD
北大核心
2016年第1期115-120,共6页
Piezoelectrics & Acoustooptics
基金
装备维修科学与改革项目"飞机复合材料构件异常检测技术研究"(2011325)
关键词
电阻抗谱法
电阻抗测量
复合材料
支持向量机
健康监测
electrical impedance spectrum method
electrical impedance measurement
composite materials
support vector machine（SVM）
health monitoring
