Raman spectra and its hyphenated techniques are the most impprtant tools for complex sample analysis. Considering the complexity of environment and characteristics of Raman signal, some overlapping spe-crum peaks can't be avoided in complex sample analysis. The wavelet transform was used to deduct fluorescent signals so as to obtain useful ones and to have overlapping Raman spectrum' s peak signals separated by making use of immune algorithm characteristics. The results show that this method can separate overlapping Raman specrum' s peak signals accurately and meanwhile obtain the quantitative information of each component with a relative error less than 2%.%拉曼光谱及其联用技术的快速发展使其成为复杂样品分析的重要手段,但由于环境的复杂性以及拉曼光谱信号的特点,在复杂样品分析中难免会出现一些重叠谱峰.为此,采用小波变换扣除原始信号中的荧光背景得到有用信号,并基于免疫算法将该方法对重叠拉曼光谱谱峰信号进行解析.结果表明:该方法可以准确地将重叠拉曼光谱谱峰信号进行分离,并且同时得到了各组分的定量信息,绝对误差在2％之内.
展开▼
