摘要
研究基于超声信号的机器学习方法在钛钢爆炸复合棒材拉剪性能分类中应用。提出基于超声信号特征值的概率神经网络(Probabilistic neural network,PNN)评估分类方法,首先获取120组工件样本的水浸超声检测全序列A扫信号,对该信号进行时域分析和改进的协方差功率谱密度估计,得到钛钢上结合层深度、上复合层的反射频率、频谱能量、下复合层的反射频率、频谱能量以及下表面二次反射波衰减等6种特征值作为PNN输入;然后进行拉伸试验得到拉剪强度值作为PNN输出;最后以96个样本特征信号和拉剪强度值建立分类训练模型,其余24个样本超声特征信号作为测试集,对这些样本的拉剪强度值进行分类预测。实验结果表明,连续24次预测准确率为94.35%。本文研究为实现钛钢爆炸复合棒材拉剪性能快速、全覆盖检验找到新思路。
The application of machine learning method based on ultrasonic signals in tensile and shear per-formance evaluation of titanium steel explosive composite bars is studied.This paper proposes a probabilis-tic neural network(PNN)evaluation and classification method based on the eigenvalues of ultrasound sig-nals.Firstly,120 samples of workpiece are taken as the object to obtain the full sequence A-scan signals of water immersion ultrasonic testing.The signals are analyzed in time domain and improved covariance pow-er spectral density estimation.Six characteristic values are used as PNN input:depth of the composite lay-er,reflection frequency of the upper composite layer,spectral energy,reflection frequency of the lower composite layer,spectral energy,and attenuation of the secondary reflected wave on the lower surface.Then,a tensile test is performed on the workpiece sample to obtain the tensile and shear strength values as the PNN output.Finally,a classification training model is established based on 96 sample characteristic sig-nals and tensile and shear strength values.The remaining 24 samples are used as the test set,and the ten-sile and shear strength values of these samples are classified and predicted.Experimental results show that the accuracy rate of 24 consecutive predictions is 94.35%.This article finds new ideas for the fast and full coverage evaluation of the tensile and shear properties of titanium steel explosive composite bars.
作者
刘云轩
吴伟
陈曦
廖翔
LIU Yunxuan;WU Wei;CHEN Xi;LIAO Xiang(Key Laboratory of Non-destructive Testing Technology,Nanchang Hangkong University,Nanchang,330063,China)
出处
《数据采集与处理》
CSCD
北大核心
2020年第2期307-314,共8页
Journal of Data Acquisition and Processing
基金
中国特种设备检测研究院NQI(2016YFF0203001-5)资助项目
南昌航空大学研究生创新专项(YC2018041)资助项目。
关键词
爆炸复合棒材
水浸超声
概率神经网络
拉剪强度
explosive composite bar
water immersion ultrasound
probabilistic neural network
tensile strength
