为提高白酒固态发酵的副产物黄水中淀粉含量预测模型精度和建模效率。采用傅里叶变换近红外光谱仪采集黄水光谱信息,利用一阶导数对光谱进行预处理,并结合偏最小二乘回归(partial least squares regression,PLSR)建立黄水淀粉定量预测...为提高白酒固态发酵的副产物黄水中淀粉含量预测模型精度和建模效率。采用傅里叶变换近红外光谱仪采集黄水光谱信息,利用一阶导数对光谱进行预处理,并结合偏最小二乘回归(partial least squares regression,PLSR)建立黄水淀粉定量预测模型。使用决定系数(R^(2))和预测均方误差(root mean square error of prediction,RMSEP)评价模型性能。光谱中含有大量冗余信息,为有效提升黄水淀粉含量检测精度和优化模型效率,将不同特征提取方法的优点结合,发现使用竞争性自适应重加权算法(competitive adaptive reweighted sampling,CARS)结合连续投影算法(successive projections algorithm,SPA)提取的光谱特征所建立的PLSR模型,相较于未使用特征提取或仅使用单一特征提取所建立的模型均有明显提升。在单一使用CARS时,模型的R^(2)为0.9654,RMSEP为0.2012%,而结合SPA后,R2为0.9738,RMSEP为0.1748%。此外,光谱维度从2203个减少到了126个,不仅提高了预测精度,也提升了建模效率。本研究提出的方法可作为黄水近红外定量模型优化的有效途径。展开更多
Objectives:The composition and content of fatty acids are critical indicators of vegetable oil quality.To overcome the drawbacks of traditional detection methods,Raman spectroscopy was investigated for the fast determ...Objectives:The composition and content of fatty acids are critical indicators of vegetable oil quality.To overcome the drawbacks of traditional detection methods,Raman spectroscopy was investigated for the fast determination of the fatty acids composition of oil.Materials and Methods:Rapeseed and soybean oil at different depths of the oil tank at different storage times were collected and an eighth-degree polynomial function was used to fit the Raman spectrum.Then,the multivariate scattering correction,standard normal variable transformation(SNV),and Savitzky–Golay convolution smoothing methods were compared.Results:Polynomial fitting combined with SNV was found to be the optimal pretreatment method.Characteristic wavelengths were selected by competitive adaptive reweighted sampling.For monounsaturated fatty acids(MUFAs),polyunsaturated fatty acids(PUFAs),and saturated fatty acids(SFAs),44,75,and 92 characteristic wavelengths of rapeseed oil,and 60,114,and 60 characteristic wavelengths of soybean oil were extracted.Support vector regression was used to establish the prediction model.The R^(2)values of the prediction results of MUFAs,PUFAs,and SFAs for rapeseed oil were 0.9670,0.9568,and 0.9553,and the root mean square error(RMSE)values were 0.0273,0.0326,and 0.0340,respectively.The R^(2)values of the prediction results of fatty acids for soybean oil were respectively 0.9414,0.9562,and 0.9422,and RMSE values were 0.0460,0.0378,and 0.0548,respectively.A good correlation coefficient and small RMSE value were obtained,indicating the results to be highly accurate and reliable.Conclusions:Raman spectroscopy,based on competitive adaptive reweighted sampling coupled with support vector regression,can rapidly and accurately analyze the fatty acid composition of vegetable oil.展开更多
在光谱建模过程中,采用不同的变量筛选算法进行光谱特征波段的提取已成为提高模型效果的重要方法。以真空包装的冷却羊肉细菌菌落总数作为研究指标,比较了两种变量筛选算法对其高光谱偏最小二乘(partial least squares,PLS)模型效果的...在光谱建模过程中,采用不同的变量筛选算法进行光谱特征波段的提取已成为提高模型效果的重要方法。以真空包装的冷却羊肉细菌菌落总数作为研究指标,比较了两种变量筛选算法对其高光谱偏最小二乘(partial least squares,PLS)模型效果的影响。研究提取了样品肌肉感兴趣区域(ROIs)的羊肉光谱并进行预处理,进而采用遗传算法(genetic algorithm,GA)和竞争性自适应重加权法(competitive adaptive reweighted sampling,CARS)分别对预处理后的473~1 000 nm范围光谱进行特征波段的提取,对比分析了不同波段下羊肉细菌菌落总数的GA-PLS,CARS-PLS和全波段PLS(W-PLS)模型效果。结果表明,GA-PLS和CARS-PLS的模型效果均优于W-PLS,且CARS-PLS模型效果最好,其校正集的决定系数(R_c^2)和均方根误差(root mean square error,RMSEC)分别为0.96和0.29,交互验证的决定系数(R_(cv)~2)和均方根误差(root mean square errorof cross validation,RMSECV)分别为0.92和0.46,预测集的决定系数(R_p^2)和均方根误差(root mean square error of prediction,RMSEP)分别为0.92和0.47,预测相对分析误差(relative prediction deviation,RPD)为3.58。因此利用高光谱图像技术结合CARS-PLS可以实现羊肉细菌菌落总数快速无损准确检测。展开更多
文摘为提高白酒固态发酵的副产物黄水中淀粉含量预测模型精度和建模效率。采用傅里叶变换近红外光谱仪采集黄水光谱信息,利用一阶导数对光谱进行预处理,并结合偏最小二乘回归(partial least squares regression,PLSR)建立黄水淀粉定量预测模型。使用决定系数(R^(2))和预测均方误差(root mean square error of prediction,RMSEP)评价模型性能。光谱中含有大量冗余信息,为有效提升黄水淀粉含量检测精度和优化模型效率,将不同特征提取方法的优点结合,发现使用竞争性自适应重加权算法(competitive adaptive reweighted sampling,CARS)结合连续投影算法(successive projections algorithm,SPA)提取的光谱特征所建立的PLSR模型,相较于未使用特征提取或仅使用单一特征提取所建立的模型均有明显提升。在单一使用CARS时,模型的R^(2)为0.9654,RMSEP为0.2012%,而结合SPA后,R2为0.9738,RMSEP为0.1748%。此外,光谱维度从2203个减少到了126个,不仅提高了预测精度,也提升了建模效率。本研究提出的方法可作为黄水近红外定量模型优化的有效途径。
基金funded by the Key Science and Technology Program of Henan Province under Grant No.212102110262Science and Technology Plan Project of Henan Provincial Market Supervision and Administration Bureau under Grant No.2021sj40+1 种基金the Key Research Program of Zhejiang Province under Grant No.2020C02018Scientific Research Projects for College Students under Grant No.2020KX0006,China.The authors acknowledge the support.
文摘Objectives:The composition and content of fatty acids are critical indicators of vegetable oil quality.To overcome the drawbacks of traditional detection methods,Raman spectroscopy was investigated for the fast determination of the fatty acids composition of oil.Materials and Methods:Rapeseed and soybean oil at different depths of the oil tank at different storage times were collected and an eighth-degree polynomial function was used to fit the Raman spectrum.Then,the multivariate scattering correction,standard normal variable transformation(SNV),and Savitzky–Golay convolution smoothing methods were compared.Results:Polynomial fitting combined with SNV was found to be the optimal pretreatment method.Characteristic wavelengths were selected by competitive adaptive reweighted sampling.For monounsaturated fatty acids(MUFAs),polyunsaturated fatty acids(PUFAs),and saturated fatty acids(SFAs),44,75,and 92 characteristic wavelengths of rapeseed oil,and 60,114,and 60 characteristic wavelengths of soybean oil were extracted.Support vector regression was used to establish the prediction model.The R^(2)values of the prediction results of MUFAs,PUFAs,and SFAs for rapeseed oil were 0.9670,0.9568,and 0.9553,and the root mean square error(RMSE)values were 0.0273,0.0326,and 0.0340,respectively.The R^(2)values of the prediction results of fatty acids for soybean oil were respectively 0.9414,0.9562,and 0.9422,and RMSE values were 0.0460,0.0378,and 0.0548,respectively.A good correlation coefficient and small RMSE value were obtained,indicating the results to be highly accurate and reliable.Conclusions:Raman spectroscopy,based on competitive adaptive reweighted sampling coupled with support vector regression,can rapidly and accurately analyze the fatty acid composition of vegetable oil.
文摘在光谱建模过程中,采用不同的变量筛选算法进行光谱特征波段的提取已成为提高模型效果的重要方法。以真空包装的冷却羊肉细菌菌落总数作为研究指标,比较了两种变量筛选算法对其高光谱偏最小二乘(partial least squares,PLS)模型效果的影响。研究提取了样品肌肉感兴趣区域(ROIs)的羊肉光谱并进行预处理,进而采用遗传算法(genetic algorithm,GA)和竞争性自适应重加权法(competitive adaptive reweighted sampling,CARS)分别对预处理后的473~1 000 nm范围光谱进行特征波段的提取,对比分析了不同波段下羊肉细菌菌落总数的GA-PLS,CARS-PLS和全波段PLS(W-PLS)模型效果。结果表明,GA-PLS和CARS-PLS的模型效果均优于W-PLS,且CARS-PLS模型效果最好,其校正集的决定系数(R_c^2)和均方根误差(root mean square error,RMSEC)分别为0.96和0.29,交互验证的决定系数(R_(cv)~2)和均方根误差(root mean square errorof cross validation,RMSECV)分别为0.92和0.46,预测集的决定系数(R_p^2)和均方根误差(root mean square error of prediction,RMSEP)分别为0.92和0.47,预测相对分析误差(relative prediction deviation,RPD)为3.58。因此利用高光谱图像技术结合CARS-PLS可以实现羊肉细菌菌落总数快速无损准确检测。