不同制取工艺对大豆油稳定性影响及IPLS- SPA特征波段选择方法应用

摘要/Abstract

摘要： 油脂稳定性关乎其食用安全，为明确大豆油不同生产工艺对其稳定性的影响，本文对压榨及浸出两种工艺生产的大豆油进行煎炸加热处理，根据酸值和羰基值的变化, 探讨煎炸时间及煎炸温度对两种工艺生产的大豆油品质的影响。实验表明：随煎炸时间的延长，浸出大豆油酸值变化程度相对较小；在220℃及240℃煎炸温度下，压榨大豆油的羰基值均高于浸出大豆油，而且在240℃时，压榨和浸出大豆油酸值及羰基值差异显著(P<0.05)，因此在高温煎炸时压榨及浸出两种大豆油相比较而言，浸出油具有较好的稳定性。同时结合近红外光谱分析技术，利用间隔偏最小二乘(interval partial least squares, IPLS)-连续投影算法(successive projections algorithm, SPA)进行特征波段选择，建立一个适于检测油脂中酸值和羰基值变化，且精确度较高的模型。酸值在1150~1315nm和1579~2444nm的组合波段具有最佳建模效果，验证相关系数为0.955，验证均方残差(RMSEP)为0.049；羰基值在1236~2093nm和2187~2594nm的组合波段具有最佳建模效果，验证相关系数为0.946，RMSEP为3.134，实验结果表明采用IPLS-SPA进行特征波段选择可有效提高模型精确度。

关键词: 压榨工艺, 浸出工艺, 酸值, 羰基值, 近红外光谱, 间隔偏最小二乘回归, 连续投影算法

Abstract: The quality stability of oil determines its edible safety，In order to clarify the influence of different soybean oil production processes on its stability, the influence of heating time and temperature on the quality of pressed and extracted soybean oil were investigated according to the changes of acid value and carbonyl value. Experiments results showed that the change degree of acid value of extracted soybean oil is relatively low with heating time. At 220 ℃ and 240 ℃, the carbonyl value of pressed soybean oil was significantly higher than that of extracted soybean oil. Moreover, at 240℃, there were significant differences in the values of acid value and carbonyl value (P<0.05).Therefore, compared with pressed soybean oil, the extracted oil has better stability in high temperature frying. At the same time, the near infrared spectroscopy technology as an employed method, the characteristic band selection is carried out by an interval partial least square (IPLS)-continuous projection algorithm (SPA), and to establish a high accurate and suitable for the rapid detection of acid value and carbonyl value in plant oil. The partial least square regression modeling as modeling method, 182 wavelength points are selected for the four characteristic bands of acid values, and 204 wavelength points in three characteristic bands of carbonyl value. To further select the characteristic wavelength, the continuous projection algorithm to be carried on the selected characteristic band. The analysis shows that the acid value has the best modeling effect in the combined band of 1150~1315nm and 1579~2444nm. The validation correlation coefficient is 0.955 and the verification mean square residual is 0.049. The carbonyl value has the best modeling effect in the combined band of 1236~2093nm and 2187~2594nm. The validation correlation coefficient is 0.946 and the verification mean square residual is 3.134. The experimental results show that the accuracy of the model may be improved effectively by selecting the characteristic bands using IPLS-SPA.

Key words: pressed oil, extracted oil, acid value, carbonyl value, Near Infrared Spectroscopy, Interval Partial Least Squares, Successive Projections Algorithm

中图分类号:

TS225.1

张丙芳王玉林刘成海张丙秀郑先哲刘勇牟艳秋. 不同制取工艺对大豆油稳定性影响及IPLS- SPA特征波段选择方法应用[J]. 食品科学, 0, (): 0-0.

Bing-Fang ZHANG. Effects of Different Processing Technologies on the Stability of Soybean Oil and Application of IPLS-SPA Characteristic Band Selection Method[J]. FOOD SCIENCE, 0, (): 0-0.

参考文献

[1]金俊, 张友峰, 陈金平, 王兴国.几种常见食用植物油溶剂残留量检测和残留溶剂组分分析[J]．[J].粮食与油脂, 2016, 29(12):75-78 [2]MCGLONE V A, MARTINSEN P J, CLARK C J, et al.On-line detection of brown heart in braeburn apples using near infrared transmission measurements[J].Postharvest Biology and Technology, 2005, 37(2):142-151 [3] XiaodongYang, GuanglinLi, JieSong, MingjuGao, ShenglingZhou.Rapid discrimination of Notoginseng powder adulteration of different grades using FT-MIR spectroscopy combined with chemometrics[J]. [J].Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2018, 205( 5):457-464 [4]张玉华, 孟一, 张应龙.近红外光谱技术在食品安全领域应用研究进展[J]. [J].食品科技, 2012, 37(10):283-286 [5]Sun WJ, Zhang X, Zhang ZY, Zhu RH.Data fusion of near-infrared and mid-infrared spectra for identification of rhubarb [ J ]. [J].Spectrochim Acta A, 2017, 171(10):72-79 [6]Zhao YL, Zhang J, Yuan TJ, Shen T, Li W, Yang SH, et al.Discrimination of wild Paris based on near infrared spectroscopy and high performance liquid chromatography combined with multivariate analysis [ J ]. [J].PLoS One, 2014, 9(2):e89-100 [7]Alamprese C, Casale M, Sinelli N, Lanteri S, Casiraghi E.Detection of minced beef adulteration with turkey meat by UV-vis, NIR and MIR spectroscopy [ J ]. [J].LWT Food Sci Technol, 2013, 53(1):225-232 [8]Zhao Y, Zhang J, Yuan T, Shen T, Li W, Yang S, et al.Discrimination of Wild Paris Based on Near Infrared Spectroscopy and High Performance Liquid Chromatography Combined with Multivariate Analysis [ J ][J].PLoS ONE, 2014, 9(2):e89-100 [9]Yun?Li, Jin-Yu?Zhang, Yuan-Zhong?Wang.FT-MIR and NIR spectral data fusion: a synergetic strategy for the geographical traceability of Panax notoginseng [ J ][J].Analytical and Bioanalytical Chemistry, 2018, 410(1):91-103 [10]Bingfang Zhang, Libo Yuan1, Bingxiu Zhang.Optical fiber spectroscopy in near infrared spectral range for rapidly discriminating formulated apple vinegar drinks [ J ]. [J].IEEE, 2013, 10(9):53-55 [11]张丙芳, 苑立波, 张丙秀.近红外光纤光谱技术检测西红柿果浆总酸及可溶性糖的研究[J].光谱学与光谱分析, 2014, 34(2):350-353 [12]Giuliano F.Zagonel, Patricio Peralta-Zamora, Luiz P. Ramos.Multivariate monitoring of soybean oil ethanolysis by FTIR[J]. [J].Talanta, 2004, 63(1):1021-1025 [13]Adewalea P, Mbaa O, Dumonta MJ, Ngadia M, Cocciardi R.Determination of the iodine value and the free fatty acid content of waste animal fat blends using FT-NIR [ J ]. [J].Vib Spectrosc, 2014, 72(1):72-80 [14]MBA O, ADEWALE P, DUMONT M J, et al.Application of near-infrared spectroscopy to characterize binary blends of palm and canola oils[J]. [J].Ind Crops Pro, 2014, 61(2):472-478 [15]ADEWALE P, MBA O, DUMONT M J, et al.Determination of the iodine value and the free fatty acid content of waste animal fat blends using FT-NIR[J].Vib Spec-trosc, 2014, 72(72):72-78 [16]Olusola S.Jolayemi,Mary AAjatta,Abimbola A. Adegeye.Geographical discrimination of palm oils (Elaeis guineensis) using quality characteristics and UV‐visible spectroscopy[J].Food Science& Nutrition, 2018, 6(4):773-782 [17]Gutiérrez S, Tardaguila J, Fernández-Novales J, Diago MP.Support vector machine and artificial neural network models for the classification of grapevine varieties using a portable NIR spectrophotometer[J]. [J].PLoS One, 2015, 10(11):143-197 [18]Márquez C, López MI, Ruisánchez I, Callao MP.FT-Raman and NIR spectroscopy data fusion strategy for multivariate qualitative analysis of food fraud [ J ]. [J].Talanta, 2016, 161(4):80-86 [19]GALTIER O, ABBAS O, DＲEAU Y L, et al.Comparison of PLS1-DA,PLS2-DA and SIMCA for classification by origin of crude petroleum oils by MIR and virgin olive oils by NIR for different spectral regions[J].Vib Spectrosc, 2011, 55(1):132-140 [20]Gorkem Ozulku, Rusen Metin Yildirim, et al.Rapid detection of adulterationof cold pressedsesame oil adulteredwith hazelnut, canola, and sun flower oils using ATR-FTIR spectroscopy combined with chemometric[ J ]. [J].Food Control, 2017, 82(4):212-216 [21]陈锋亮, 魏益民, 钟耕.大豆油高温煎炸质变过程的研究[J].[J].中国油脂, 2006, 6(08):19-22 [22]万重, 黄朦倩, 陈欢, 张清.煎炸不同食物对煎炸油品质的影响[ J ]. [J].中国油脂, 2018, 43(3):54-58 [23]N?rgaard L, Saudland A, Wagner J, et al.Interval Partial Least-Squares Regression (iPLS): A Comparative Chemometric Study with an Example from Near-Infrared Spectroscopy[ J ][J].Applied Spectroscopy, 2000, 54(3):413-419 [24]陈斌, 孟祥龙, 王豪.连续投影算法在近红外光谱校正模型优化中的应用[J].分析测试学报, 2007, 26(1):66-69 [25]ARAùJO M C U, SALDANHA T C B, GALV?O R K H, et al.The successive projections algorithm for variable selection in spectroscopicm ulticom ponent analysis [ J ][J].Chem om etrics and Intelligent Laboratory System s, 2001, 57(2):65-70 [26].?