| 基于太赫兹时域光谱和PCA-SVM算法的甜蜜素含量分析 |
| 投稿时间：2024-03-04 修订日期：2024-03-13 点此下载全文 |
| 引用本文： |
| 摘要点击次数: 134 |
| 全文下载次数: 0 |
|
| 基金项目:国家自然科学基金(批准号：61927813, 61991432) |
|
| 中文摘要:光谱分析是研究太赫兹（THz）辐射与物质相互作用的重要手段。采用全光纤式THz时域光谱系统（TDS）测试了不同含量甜蜜素样品的透过率光谱，发现甜蜜素的特征吸收峰位置在1.4 THz和1.7 THz附近；采用主成分分析结合支持向量机（PCA-SVM）的方法建立了甜蜜素含量回归模型，并与遗传算法结合偏最小二乘（GA-PLS）模型预测结果进行分析比较，引入决定系数和预测均方根误差来评价建模效果，对以10%含量梯度制作的样品集进行检测；研究结果表明，采用SVM方法、GA-PLS方法和PCA-SVM方法建立的预测模型，其预测均方根误差分别为1.885%、1.926%和2.432%，因此PCA-SVM方法的预测效果最优，且预测数据与实际数据均表现出良好的相关性，获得了效果良好的含量回归预测模型，为甜蜜素含量的检测与分析提供了一种有效手段。 |
| 中文关键词:太赫兹时域光谱 主成分分析 支持向量机 含量回归预测模型 |
| |
| Analysis of saccharin content based on terahertz time-domain spectroscopy and PCA-SVM algorithm |
|
|
| Abstract:Spectral analysis is an important means of studying the interaction between terahertz (THz) radiation and matter. The transmittance spectra of samples with different levels of saccharin were tested using an all-fiber THz time-domain spectroscopy system (TDS), and it was found that the characteristic absorption peaks of saccharin were located around 1.4 THz and 1.7 THz; A regression model for saccharin content was established using principal component analysis combined with support vector machine (PCA-SVM), and compared with the prediction results of genetic algorithm combined with partial least squares (GA-PLS) model. Correlation coefficients and root mean square error were introduced to evaluate the modeling effect, and the sample set made with a 10% content gradient was tested; The research results show that the prediction models established using SVM, GA-PLS, and PCA-SVM methods have root mean square errors of 1.885%, 1.926%, and 2.432%, respectively. Therefore, the PCA-SVM method has the best prediction performance, and the predicted data shows good correlation with the actual data. A content regression prediction model with good performance has been obtained, providing an effective means for the detection and analysis of saccharin content. |
| keywords:terahertz time-domain spectroscopy principal component analysis support vector machine content regression prediction model |
| HTML 查看/发表评论 下载PDF阅读器 |