首页 | 本学科首页   官方微博 | 高级检索  
     检索      

近红外光谱法在红曲菌固态发酵过程参数检测中的应用
引用本文:黄常毅,范海滨,刘飞,许赣荣,彭秀辉.近红外光谱法在红曲菌固态发酵过程参数检测中的应用[J].分析测试学报,2014,33(1):13-20.
作者姓名:黄常毅  范海滨  刘飞  许赣荣  彭秀辉
作者单位:1.江南大学轻工过程先进控制教育部重点实验室自动化研究所;2.江南大学工业生物技术教育部重点实验室
基金项目:江苏高等学校优秀科技创新团队、江苏高校优势学科建设工程资助项目;国家高技术研究发展计划项目(863计划,2012AA021302)
摘    要:研究了近红外光谱技术快速检测红曲菌固态发酵过程参数水分含量和pH值的可行性。针对传统基于间隔策略波长选择方法忽略非线性因素的缺点,采用一种基于最小二乘支持向量机(Least squares support vector machines,LS-SVM)非线性模型的波长筛选算法:联合区间最小二乘支持向量机(Synergy interval least squares support vector machines,siLS-SVM),并将新算法与相关系数法、iPLS算法、siPLS算法对比。实验结果显示,联合siLS-SVM算法和LS-SVM模型取得了最好的预测效果,水分含量、pH值的预测集相关系数(Rp)分别为0.962 1、0.976 1,预测均方根误差(RMSEP)分别为0.012 9、0.145 2,表明模型具有较好的拟合度和预测性能。应用近红外光谱法进行红曲菌固态发酵过程的水分含量和pH值的快速检测可行,该方法为进一步实现其过程参数的在线检测及发酵条件优化提供了技术基础。

关 键 词:近红外光谱  联合间隔最小二乘支持向量机  最小二乘支持向量机  红曲菌  固态发酵  水分含量  pH  

Application of NIR Spectroscopy in Detection of Process Parameters for Solid-state Fermentation of Monascus
Abstract:The feasibility of rapid detection of the process parameters such as moisture content and pH value in solid-state fermentation of Monascus was studied by near infrared technique.In view of the traditional wavelength selection methods based on interval strategy ignoring the disadvantages of the nonlinear factors,this paper adopted a new wavelength selection algorithm based on LS-SVM nonlinear model which was named synergy interval least squares support vector machines(siLS-SVM),and the new algorithm was compared with correlation coefficient method,iPLS algorithm and siPLS algorithm.The results showed that the combination of siLS-SVM and LS-SVM model achieved the best prediction result.The prediction correlation coefficients(Rp) of moisture content and pH value were 0.962 1 and 0.976 1,respectively.The root mean square errors for prediction set(RMSEP) were 0.012 9 and 0.145 2,respectively.The obtained results demonstrated that the fitting and the predictive accuracy were satisfactory,and it was feasible to apply NIRS method to fast determination of moisture contents and pH value in solid state fermentation of Monascus.This study provided a technical foundation for the further online detection of process parameters in solid-state fermentation of Monascus and the optimization of fermentation conditions.
Keywords:NIR  siLS-SVM  LS-SVM  Monascus  solid-state fermentation  moisture content  pH value
本文献已被 CNKI 等数据库收录!
点击此处可从《分析测试学报》浏览原始摘要信息
点击此处可从《分析测试学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号