用于火星表面物质探测的拉曼光谱技术研究

发布时间：2019-03-20 06:59

【摘要】：拉曼光谱技术是识别物质成分的重要手段之一,它具有谱峰尖锐、探测范围广和微区域探测的特点。论文在火星表面物质成分探测的大背景下,分析了拉曼光谱技术比其他曾用在火星表面物质探测的光谱技术(可见近红外光谱技术、穆斯堡尔光谱技术和X射线荧光光谱技术)的一些优势。并结合RLS和SHERLOC两台即将用于火星探测的拉曼光谱仪,指出了拉曼光谱技术在火星探测方面的可行性。本论文综合考虑了拉曼光谱技术的难点,并结合火星表面的实际探测情况,解决了用于火星表面物质探测的远程拉曼光谱仪中的两个关键问题。第一个关键问题是卡塞格林望远系统的设计,主镜和次镜均采用抛物面,入瞳直径200mm,物距1.5-2.5m,次镜可移动。第二个关键问题是预处理算法的编写,采用小波变换和多项式拟合的算法,分别去除拉曼光谱原始信号中的随机噪声和荧光背景。小波变换采用‘sym5’小波母函数进行5层小波分解,软阈值去噪后再重构信号。多项式拟合的关键点是先对去噪后的信号进行了求导,用导数在0附近一定范围内(约±10)的点进行多项式拟合,拟合多次让背景更准确,再从去噪后的谱图中扣除荧光背景,得到最终的拉曼光谱图。为了对远程拉曼光谱仪进行实验验证,并初步验证它用于火星表面物质探测的可行性,选用了在火星表面已被探测到的四种物质——方解石、橄榄石、赤铁矿和辉石,以及一种有机物——葡萄糖,作为实验样品。利用远程拉曼光谱仪对几种样品进行了探测,并用编写的小波变换和多项式拟合的算法对原始信号进行了光谱预处理,对最终数据进行光谱分析,得到了比较可靠的结果,初步验证了仪器的性能。
[Abstract]:Raman spectroscopy is one of the important methods to identify the material composition. It has the characteristics of sharp spectrum peak, wide detection range and micro-region detection. In this paper, under the background of Mars surface material composition exploration, we analyze the Raman spectroscopy technology compared with other spectral techniques used in Mars surface material exploration (visible near infrared spectroscopy). Mossbauer spectroscopy and X-ray fluorescence spectroscopy) some advantages. The feasibility of Raman spectroscopy in Mars exploration is pointed out by combining two Raman spectrometers, RLS and SHERLOC, which will be used in Mars exploration. In this paper, the difficulties of Raman spectroscopy are considered comprehensively, and two key problems in the remote Raman spectrometer for Mars surface material exploration are solved in combination with the actual exploration of Mars surface. The first key problem is the design of Cassegrin telescopic system. The primary mirror and the secondary mirror are paraboloid, the pupil diameter is 200mm, the object distance is 1.5m 2.5m, and the secondary mirror can be moved. The second key problem is the preparation of pre-processing algorithm using wavelet transform and polynomial fitting algorithm to remove random noise and fluorescence background from the original Raman spectra respectively. Wavelet transform uses' sym5' wavelet generating function to decompose 5-layer wavelet, soft threshold de-noising and signal reconstruction. The key point of polynomial fitting is that the de-noising signal is derived first, and polynomial fitting is carried out with the point whose derivative is around 0 (about 卤10). The fitting makes the background more accurate many times, and then deducts the fluorescence background from the de-noising spectrum map. The final Raman spectra are obtained. In order to test the remote Raman spectrometer and preliminarily verify its feasibility for the Mars surface material exploration, four substances-calcite, olivine, hematite and pyroxene-which have been detected on the surface of Mars, have been selected. And an organic substance, glucose, as an experimental sample. The remote Raman spectrometer is used to detect several samples, and the original signal is preprocessed by wavelet transform and polynomial fitting algorithm. The final data are analyzed by spectral analysis, and the reliable results are obtained. The performance of the instrument is preliminarily verified.
【学位授予单位】：中国科学院研究生院(西安光学精密机械研究所)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P185.3

【相似文献】