NIR Analysis of TFAs Content in Oil Based on Kalman Filtering and DBN
WANG Li-qi1, CHEN Ying-shu1, LIU Yu-qi1, SONG Yang2, YU Dian-yu3, ZHANG Na2*
1. School of Computer and Information Engineering, Harbin University of Commerce/Heilongjiang Key Laboratory of E-Commerce and Information Processing, Harbin 150028, China
2. School of Food Engineering, Harbin University of Commerce, Harbin 150076, China
3. College of Food Science, Northeast Agricultural University, Harbin 150030, China
Abstract：In order to control the content of trans fatty acids (TFAs) in the process of oil deodorization, this paper presents a fast method for detecting trans fatty acids (TFAs) content in soybean oil based on near-infrared spectroscopy. First, we prepared 100 soybean oil samples with different TFAs content，and detected precisely the values of TFAs contents by gas chromatography. Then, the near-infrared spectrum of oil samples was scanned and denoised by various methods, and it is found that the denoising effect of MSC was the best. In order to study the characteristic absorption of TFAs in near-infrared region, we used a variety of iPLS methods to select the characteristic band of the spectral data, and the characteristic absorption band of TFAs is selected as 7 258～7 443/6 502～6 691/6 120～6 309 cm-1. On this basis, the Kalman filtering algorithm is used to select the characteristic wavelength variables, and 27 TFAs characteristic wavelength variables are optimized. The deep belief network (DBN) is adopted to construct the correction model, and we found that the performance of the DBN model is the best adopting 3 hidden layers and 50-35-90 hidden layer nodes. Finally, the DBN model with this parameter is compared with the regression model of trans fatty acid content established by PLS. The results show that: when we used the whole denoised spectrum to construct model, the prediction effect of DBN is better than that of PLS，R2 is 0.879 4, RMSEP is 0.060 3 and RSD is 2.18%. When we used the selected characteristic band to model, the prediction effect of the PLS model is better than that of the DBN model. Using the 27 optimized characteristic wavelength variables to construct model, DBN has a good prediction effect, R2 is 0.958 4, RMSEP is 0.035 0 and RSD is 1.31%. It shows that the generalization ability of DBN is better, which achieved better prediction results by using a small number of wavelength variables. The proposed method in this paper can meet the practical needs, and provide technical support for online detecting and regulating TFAs content and producing low/zero TFAs oil products.
Key words：Oil; Trans fatty acids (TFAs); Near-infrared spectroscopy (NIR); Kalman filtering; Deep believe net (DBN)
王立琦，陈颖淑，刘雨琪，宋 旸，于殿宇，张 娜. 基于Kalman滤波与DBN的油脂中TFAs含量近红外光谱分析[J]. 光谱学与光谱分析, 2021, 41(03): 848-852.
WANG Li-qi, CHEN Ying-shu, LIU Yu-qi, SONG Yang, YU Dian-yu, ZHANG Na. NIR Analysis of TFAs Content in Oil Based on Kalman Filtering and DBN. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(03): 848-852.
[1] Albuquerque T G, Costa H S, Castilho M C, et al. Trends in Food Science & Technology, 2011，22(10)：543.
[2] Laroussi-Mezghani S, Vanloot P, Molinet J, et al. Food Chemistry, 2015，173：122.
[3] Zhang M, Yang X, Zhao H T, et al. Food Control, 2015，57：293.
[4] MA Su-min, BI Yan-lan, ZHANG Lin-shang(马素敏, 毕艳兰, 张林尚). Chinese Oil(中国油脂), 2014, 39(3): 15.
[5] Zou Q Q, Tan H R, Ning L U. Packaging & Food Machinery, 2016，32(2)：56，62.
[6] Xu L, Zhu X, Chen X, et al. Food Chemistry, 2015，185：503.
[7] MO Xin-xin, SUN Tong, LIU Mu-hua, et al(莫欣欣, 孙 通, 刘木华, 等). Chemical Analysis(分析化学), 2017, 45(11): 1694.
[8] Mossoba M M, Tyburczy C, Delmonte P, et al. Trans Fats Replacement Solutions. Academic Press and AOCS Press, 2014，89.
[9] ZHANG Rui, DING Xiang-qian, GAO Zheng-xu, et al(张 瑞, 丁香乾, 高政绪, 等). Computer & Digital Engineering(计算机与数字工程), 2019, 47(2): 383.
[10] WANG Jing, DING Xiang-qian, WANG Xiao-dong, et al(王 静, 丁香乾, 王晓东, 等). Infrared and Laser Engineering(红外与激光工程), 2019, 48(4): 31.
[11] Yang H, Hu B, Pan X, et al. Journal of Innovative Optical Health Sciences, 2017, 10(2).
[12] Peng K, Jiao R, Dong J, et al. Neurocomputing, 2019，361：19.
