基于径向基神经网络和BP神经网络对日落黄荧光光谱的分析

对食用合成色素日落黄的荧光光谱进行研究,发现在最佳激发波长370nm紫外光的激励下,荧光峰值波位于576nm;经分析认为,日落黄溶液之所以能产生荧光是因为分子中偶氮键将一个苯环和一个萘环连接在一起,形成大共轭结构,并且取代基与—SO3Na与—OH处于萘环的对位,大大增强了日落黄分子的共轭程度,使其具有强的吸光功能,发出强荧光。另外,结合径向基神经网络和BP神经网络对未知样本进行浓度预测,结果精确,平均相对误差分别为3.51%和5.45%,RSD分别为1.83%和2.95%。该方法有望成为对食用合成色素进行高效检测的有效方法。

Identification of Sunset Yellow by Detection of Fluorescence Spectra Based on Radial Basis Neural Network and BP Neural Network

The fluorescence spectra of synthetic edible pigment of sunset yellow were analyzed. The fluorescence peak wavelength was 570nm when induced by the best excitation wavelength 370 nm. Sunset yellow solution can produce fluorescence because nitric group linked a benzene ring and a naphthyl together to form a conjugate structure located at the eontraposition of naphthyl in place of the group and --OH, and greatly increases the conjugate extent of sunset yellow elements to absorb and give off fluorescence. In addition, solutions of unknown sample were predicted by using radial basis neural network and BP neural network, the training time was less and the prediction results were precise with the average relative error of 3.51% and 5.45 %, the relative error of 1.83 % and 2. 95%. This method would become an effective method to detect synthetic food dyes efficiently.