激光与光电子学进展, 2017, 54 (7): 073001, 网络出版: 2017-07-05
最小二乘支持向量机在对羟基苯甲酸甲酯钠荧光检测中的应用
Application of Least Squares Support Vector Machine in Fluorescence Detection of Sodium Methylparaben
光谱学 荧光光谱 对羟基苯甲酸甲酯钠 粒子群优化算法 最小二乘支持向量机 spectroscopy fluorescence spectra sodium methylparaben particle swarm optimization algorithm least squares support vector machine
摘要
对羟基苯甲酸甲酯钠是一种常见的食品添加剂,如果长时间食用或者超量食用会对人体造成一定的危害。采用FS920荧光光谱仪对对羟基苯甲酸甲酯钠橙汁溶液和水溶液进行检测,实验结果表明两者的特征峰发生了明显的变化。经分析得出,对羟基苯甲酸甲酯钠橙汁溶液的荧光光谱受到橙汁荧光特性干扰,一定浓度范围的溶液光谱图存在较大差距,对羟基苯甲酸甲酯钠浓度与荧光强度之间的关系复杂。为了精确地检测橙汁中对羟基苯甲酸甲酯钠的浓度,结合荧光光谱法与最小二乘支持向量机,建立了橙汁溶液中对羟基苯甲酸甲酯钠的检测模型,使用改进的粒子群优化算法得到影响模型性能的正则化参数和核函数。实验得到了较为理想的结果,与普通反向传播(BP)神经网络、基本粒子群寻参的最小二乘支持向量机等方法相比,该方法性能最优,得到的平均回收率为97.05%,平均相对误差为2.71%,均方根误差为3.04%,模型输出与真实值之间的相关系数是0.9999。该方案可以做为橙汁中对羟基苯甲酸甲酯钠浓度的精确检测方法。
Abstract
Sodium methylparaben is a common food additive, and long-period or excessive ingestion will do harm to the human body. An FS920 fluorescence spectrometer is used to detect sodium methylparaben in orange juice and aqueous solutions. The experimental results show that the characteristic peaks of the orange juice and the aqueous solutions have obvious difference. It is inferred that the interference is mainly from the fluorescence characteristics of orange juice. There exists notable difference between the two solutions with certain concentration range of sodium methylparaben, and the relationship between the relative fluorescence intensity and the sodium methylparaben concentration is complex. Therefore, in order to accurately detect the content of sodium methylparaben in orange juice, fluorescence spectroscopy and least squares support vector machine are combined to establish a model to detect sodium methylparaben in orange juice, and the regularization parameter and the kernel function are obtained with the improved particle swarm optimization algorithm. Compared with ordinary back-propagating (BP) neural network and least squares support vector machine based on particle swarm optimization, the model proposed has optimal performance, the average recovery rate is 97.05%, the average relative error is 2.71%, the root mean square error is about 3.04%, and the correlation coefficient between the model output and the real value is about 0.9999. This method can be used for accurate determination of sodium methylparaben in orange juice.
王书涛, 张彩霞, 王志芳, 张强, 马晓晴, 郑亚南. 最小二乘支持向量机在对羟基苯甲酸甲酯钠荧光检测中的应用[J]. 激光与光电子学进展, 2017, 54(7): 073001. Wang Shutao, Zhang Caixia, Wang Zhifang, Zhang Qiang, Ma Xiaoqing, Zheng Ya′nan. Application of Least Squares Support Vector Machine in Fluorescence Detection of Sodium Methylparaben[J]. Laser & Optoelectronics Progress, 2017, 54(7): 073001.