论文部分内容阅读
提出了一种在小样本条件下建立光谱定量分析的新方法-Bootstrap-SVM模型。以道路沥青为研究对象,共收集29个来自6个不同单位的沥青样本,利用所提方法建立了沥青针入度定量分析模型。Bootstrap-SVM由Bootstrap重抽样、噪声注入及SVM三个步骤组成。为了对比所提方法的优势,对比了目前常用的PLS模型以及SVM模型。研究结果表明Bootstrap-SVM,PLS,SVM预测均方根误差分别为0.773 5,2.889,1.784 4,所提方法预测精度最好,为小样本条件下光谱定量分析提供了一种新的有效方法。
A new method, Bootstrap-SVM, for establishing quantitative spectroscopy analysis under small sample conditions is proposed. A total of 29 bitumen specimens from 6 different units were collected from the road bitumen, and a quantitative analysis model of bitumen penetration was established by using the proposed method. Bootstrap-SVM consists of Bootstrap re-sampling, noise injection and SVM three steps. In order to compare the advantages of the proposed method, we compare the commonly used PLS model and SVM model. The results show that the prediction root mean square errors of Bootstrap-SVM, PLS and SVM are 0.773 5, 2.8989 and 1.784 4, respectively. The proposed method has the best prediction accuracy and provides a new effective method for spectral quantitative analysis under small sample conditions.