摘要
选取市售米粉作为研究对象,测定样品的近红外光谱,用国标分光光度法测定样品中硼砂的含量作为标准值,结合偏最小二乘法,建立了米粉中硼砂含量的近红外光谱快速检测模型。结果表明,采用二阶导数谱,在9767.73~4594.29 cm^(-1)光谱区间,经多元散射校正、七点平滑处理建立的模型为较优模型。该模型的校正集线性相关系数(Rc)值为0.9907,校正集均方根偏差(RMSEC)值为0.252,验证集线性相关系数(Rp)值为0.9631,验证集均方根偏差(RMSEP)值为0.709,主因子数(factors)为6,相对分析误差(RPD)为为3.1。模型有较好的稳定性,外部验证结果准确,可以用于米粉中硼砂含量的快速检测。
Near infrared spectroscopy(NIR)combined with partial least square(PLS)was adopted to establish a rapid method for determinating borax content in rice flour,and the predictive model parameters were optimized.Different kinds of rice flour saled in supermarkets were selected as research objects.Using the national standard method of ultraviolet spectrophotometry to determinate the borax content in the samples firstly,then the near infrared spectra of each sample were scanned by Fourier transform near infrared spectrometer,and the rapid prediction model of borax content in rice flour were established by using TQ Analyst software.It was found that the optimal spectral preprocessing method was multiplicative scatter correction(MSC)combined with second derivatives(SD)and seven dots smoothing,and the predictive model with optimal results was constructed using the spectral range of 9767.73~4594.29 cm^(-1) and 6 PLS factors,with the root mean square error of calibration(RMSEC)of 0.252,the root mean square error of prediction(RMSEP)of 0.709,the correlation coeffificient for calibration(Rc)up to 0.9907,the correlation coeffificient for prediction(Rp)up to 0.9631,and the relative percent deviation(RPD)up to 3.1 as well.The verifification results showed the model-predicted value had a good linear relationship with the experimental value.In conclusion,the NIR prediction model could accurately predict the content of borax in rice flour.
作者
梁奇峰
曾令聪
郭红卫
李岚
郭远凯
何江
张丰如
Liang Qifeng;Zeng Lingcong;Guo Hongwei;Li Lan;Guo Yuankai;He Jiang;Zhang Fengru(School of Chemistry and Environment,Jiaying University,Meizhou 514015,China)
出处
《广东化工》
CAS
2023年第16期152-153,184,共3页
Guangdong Chemical Industry
基金
嘉应学院自然科学项目(2019KJY04)。
关键词
近红外光谱
偏最小二乘法
硼砂
米粉
快速检测
near infrared spectroscopy
partial least square method
borax
rice flour
rapid detection
