基于MCR-ALS-PLS对发酵液中葡萄糖的定量分析  

MCR-ALS-PLS Algorithm for the Quantitative Analysis of Glucose in Fermentation Broth

作　　者：于永爱 李子璇 刘雅娇 陈娟 Yu Yongai;Li Zixuan;Liu Yajiao;Chen Juan(Department of Computer and Information Engineering,Shanghai Polytechnic University,Shanghai 201209,China;Shanghai Oceanhood Opto-electronics Tech Co.,Ltd.,Shanghai 201216,China)

机构地区：[1]上海第二工业大学,计算机与信息工程学院,上海201209 [2]上海如海光电科技有限公司,上海201216

出　　处：《实验与分析》2024年第2期45-52,共8页LABOR PRAXIS

摘　　要：葡萄糖浓度是生物发酵过程中的重要参数,精确控制葡萄糖浓度,可以优化发酵条件,提高生产效率和产品质量。然而大多数情况下,由于发酵液中物质复杂,导致葡萄糖的拉曼特征峰难以明确。针对这种情况下的葡萄糖定量分析,首先使用多元曲线分析-交替最小二乘(MCR-ALS)算法将葡萄糖拉曼光谱与发酵液基质光谱分离,然后采用竞争自适应重加权采样(CARS)算法进行降维和偏最小二乘2(PLS)算法进行浓度建模。结果表明使用这种方法建模的预测效果得到明显提高,测试集R由0.86026提升到了0.98031,均方根误差(RMSE)由0.00183降低到了0.00061。Glucose concentration is an important parameter in biological fermentation processes.Precise control of glucose concentration can optimize fermentation conditions,improve production efficiency,and enhance product quality.However,in most cases,due to the presence of various substances in the fermentation broth,the Raman characteristic peaks of glucose in the fermentation broth are difficult to distinguish.For the quantitative analysis of glucose under such circumstances,the Multivariate Curve Resolution-Alternating Least Squares(MCR-ALS)algorithm is first used to separate the Raman spectra of the target substance glucose from the matrix spectra of the fermentation broth.Then,the Competitive Adaptive Reweighted Sampling(CARS)algorithm is employed for dimensionality reduction,and Partial Least Squares(PLS)algorithm is used for concentration modeling.The results show that using this method 2 for modeling significantly improves the predictive performance,with the R of the test set increasing from 0.86026 to 0.98031,and the root mean square error(RMSE)decreasing from 0.00183 to 0.00061.

关 键 词：MCR-ALS 拉曼光谱 定量分析 葡萄糖 

分 类 号：O657.37[理学—分析化学]