首页 | 本学科首页   官方微博 | 高级检索  
     

基于BP神经网络检测面粉中滑石粉含量的研究
引用本文:刘翠玲,董秀丽,孙晓荣,吴静珠,吴胜男.基于BP神经网络检测面粉中滑石粉含量的研究[J].北京轻工业学院学报,2012(5):77-80.
作者姓名:刘翠玲  董秀丽  孙晓荣  吴静珠  吴胜男
作者单位:北京工商大学计算机与信息工程学院,北京100048
摘    要:利用近红外光谱技术对掺杂滑石粉的小麦面粉进行了检测,采用多元散射校正对谱图进行预处理,利用BP神经网络中的SCG反向传播算法训练函数建立了面粉中滑石粉的定量分析模型,并对校正集和预测集进行了定量分析,分析结果为R2=0.997 3,RMSEC=0.436 7,RMSEP=1.708 8.结果表明,BP神经网络结合近红外光谱技术检测面粉中滑石粉含量具有快速、精度高、泛华能力强的优点,可用于面粉中滑石粉含量的快速准确检测.

关 键 词:BP神经网络  近红外光谱  滑石粉  小麦面粉

Rapid Determination of Talc-containing Flour Based on BP Neural Network
LIU Cui-ling,DONG Xiu-li,SUN Xiao-rong,WU Jing-zhu,WU Sheng-nan.Rapid Determination of Talc-containing Flour Based on BP Neural Network[J].Journal of Beijing Institute of Light Industry,2012(5):77-80.
Authors:LIU Cui-ling  DONG Xiu-li  SUN Xiao-rong  WU Jing-zhu  WU Sheng-nan
Affiliation:(School of Computer Science and Information Engineering, Beijing Technology and Business University, Beijing 100048, China)
Abstract:Near infrared spectral technology (NIR) was used to test talc-containing spectrum was preprocessed with muhiplicative scatter correction. The quantitative anal wheat flour. The ysis model of talc containing flour was built using SCG back propagation algorithm training function of BP neural network, and the calibration set and prediction set were quantitatively analyzed. R2 was 0. 997 3, the root mean square error of calibration (RMSEC) was 0. 436 7, and the root mean square error of prediction ( RM- SEP) was 1. 708 8. The results showed that BP neural network with NIR for the determination of talc-con- taining flour has the advantages of fast, high precision, and the ability of Fanhua, and can be used for talc-containing flour.
Keywords:BP neural network  NIR  talcum powder  wheat flour
本文献已被 维普 等数据库收录!
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