轮胎橡胶颗粒红外光谱的多元分类研究

Multivariate classification of infrared spectroscopy about tire rubber particles

交通肇事现场轮胎橡胶颗粒的检验鉴别是一项常见的工作。为降低检验成本,提高检验效率,提出一种基于红外光谱及其化学模式研究的轮胎橡胶颗粒鉴别方法。实验采集并获取朝阳等4种品牌共计46个样本的红外光谱数据,预处理采用自动基线校正、峰面积归一化、多元散射校正、Savitzky-Golay 7点平滑,借助主成分分析和判别分析构建分类模型。结果表明,46个样本主要由丁苯橡胶、顺丁橡胶和异戊橡胶三种类型,PCA模型对46个样本的区分能力相对较弱,判别分析模型则实现了100%的准确区分和归类,分类结果理想,该方法快速、无损、准确,具有一定的普适性和借鉴意义。

It is common to verify the rubber particles in the traffic accident scene.In order to reduce the inspection cost and improve the inspection efficiency,a method for identifying tire rubber particles based on infrared spectroscopy and its chemical mode is proposed.The paper obtained infrared spectra data of 46 samples from 4 brands including Chaoyang and used automatic baseline correction,peak area normalization,multiple scatter correction and Savitzky-Golay 7-point smoothing to pretreat the data.Besides,combing principal component analysis with discriminant analysis to constructed the model.The results showed that all samples belongs to three types:styrene butadiene rubber,butadiene rubber and isoprene rubber.The PCA model had a relatively weak ability to distinguish all samples,the discriminant analysis model attained 100%accurate classification,which is fast,non-destructive,accurate,and has certain universality and reference significance.