基于激光诱导击穿光谱技术的土壤快速分类方法研究

发布时间：2018-06-03 15:04

本文选题：激光诱导击穿光谱 + 土壤分类　；参考：《光谱学与光谱分析》2017年01期

【摘要】：为实现不同种类土壤的快速分类鉴别,实验研究了基于激光诱导击穿光谱技术的土壤快速分类方法。由于不同类型的土壤在元素组成上会存在较大差异,所以利用激光诱导击穿光谱技术进行土壤分类具有可行性。不同土壤在相同实验条件下产生的等离子体温度会存在较大差异,可以作为分类的重要依据,所选择的7类土壤中,赤红壤的等离子体温度最高。选取土壤中6种常量元素Si,Fe,Al,Mg,Ca和Ti的光谱强度作为分类指标,利用主成分分析(principal component analysis,PCA)对7种土类的25个样品进行了分类,其中砖红壤和赤红壤分类出现了交叠,而不同高山草甸土样品之间元素差异较大,并没有实现较好的聚类。利用反向传播神经网络(back-propagation artificial neural network)结合土壤的LIBS光谱对土壤进行了分类,分类结果与PCA结果相近,赤红壤与砖红壤出现了识别错误。当用PCA分析获得三个主成分值作为BP神经网络的输入量时,获得了较好的分类结果,因为简化了输入量,降低了BP神经网络的误差,此时只有一个高山草甸土被识别成褐土,而高山草甸土的等离子体温度显著低于褐土,所以结合不同土壤类型的等离子体温度差异,能够实现不同土壤的分类识别。实验证明激光诱导击穿光谱技术可以应用于土壤分类,为土壤普查和合理利用提高了一种新的技术。
[Abstract]:In order to realize the rapid classification and identification of different kinds of soils, a rapid classification method based on laser induced breakdown spectroscopy was studied. Because there are great differences in element composition among different types of soil, it is feasible to classify soil by laser induced breakdown spectroscopy. The temperature of plasma produced by different soils under the same experimental conditions will be quite different, which can be used as an important basis for classification. The plasma temperature of lateritic soil is the highest among the selected seven types of soils. The spectral intensities of six kinds of constant elements, Si-FeFeAlMg-Ca and Ti, were selected as classification indexes. 25 samples of 7 soil groups were classified by principal component analysis (principal component analysis), in which the classification of latosol and lateritic red soil appeared overlapping. However, the elements of different alpine meadow soil samples are different, and there is no good clustering. Soil classification was carried out by back-propagation artificial neural network) and LIBS spectrum of soil. The classification results were similar to those of PCA, and there were some errors in identification between lateritic soil and latosol. When using PCA analysis to get three principal component values as the input of BP neural network, a better classification result is obtained, because the input quantity is simplified and the error of BP neural network is reduced, only one alpine meadow soil is recognized as cinnamon soil. The plasma temperature of alpine meadow soil is significantly lower than that of cinnamon soil, so combining the plasma temperature difference of different soil types, the classification and recognition of different soils can be realized. The experimental results show that laser induced breakdown spectroscopy can be applied to soil classification, which improves a new technique for soil survey and rational utilization.
【作者单位】：中国科学院安徽光学精密机械研究所环境光学重点实验室;
【基金】：国家(863计划)项目(2014AA06A513,2013AA065502) 安徽省杰出青年科学基金项目(1508085JGD02) 国家自然科学基金项目(61378041) 中国科学院STS项目(KFJ-EW-STS-083);中国科学院合肥研究院院长基金项目(YZJJ201502)资助
【分类号】：S155

【相似文献】