To address the subjectivity and latency issues inherent in traditional methods of powdered milk quality assessment, this study established a soft sensor model for the quality of powdered milk.This model, constructed based on image analysis technology and artificial intelligence algorithms, utilized the processing conditions and particle morphology of powdered milk.It enabled the accurate and real-time prediction of two crucial quality attributes, including the dispersibility and solubility of instant whole milk powder.With the aid of digital microscopes and image processing techniques, morphological parameters of milk powder particles were obtained.Concurrently, the issue of data imbalance within the original dataset from milk powder factories was rectified by employing resampling techniques.Using the morphological parameters obtained from the experiments and the processing condition data provided by the milk powder plant, this study constructed soft sensor models for dispersibility and solubility based on partial least squares and artificial neural network models.The constructed models were then validated for their accuracy using the original data.Results showed that the partial least squares model, constructed for predicting dispersibility and solubility, had Q2 and R2 values of 0.72 and 0.94, and 0.85 and 0.95, respectively.Additionally, the artificial neural network model designed for the same purpose yielded R2 values of 0.97 and 0.96.The outstanding performance of these models proves their ability to predict the dispersibility and solubility of milk powder accurately and in real time and introduces a new approach for the online quality assessment of instant whole milk powder.
[1] 赵爱利, 王文祥.到底哪个奶营养价值更高？[N].人民政协报, 2021-10-27(7).
ZHAO A L, WANG W X.Which milk has higher nutritional value? [N].People′s Political Consultative Conference News, 2021-10-27(7).
[2] 袁永俊, 陈宝琳. 改善豆奶粉冲调性能的技术研究. 四川工业学院学报, 1999(4):42-46.
YUAN Y J, CHEN B L. Study on the technology for improving soybean milk powder's dissolution properties. Journal of Sichuan University of Science and Technology, 1999(4):42-46.
[3] CROWLEY S V, DESAUTEL B, GAZI I, et al.Rehydration characteristics of milk protein concentrate powders[J].Journal of Food Engineering, 2015, 149:105-113.
[4] LEE J, CHAI C, PARK D J, et al.Novel convenient method to determine wettability and dispersibility of dairy powders[J].Korean Journal for Food Science of Animal Resources, 2014, 34(6):852.
[5] DAVENEL A, SCHUCK P, MARIETTE F, et al.NMR relaxometry as a non-invasive tool to characterize milk powders[J].Le Lait, 2002, 82(4):465-473.
[6] DEPREE N, PRINCE-PIKE A, YOUNG B, et al.Predictive modelling of instant whole milk powder functional performance across three industrial plants[J].Journal of Food Engineering, 2019, 252:1-9.
[7] DING H H, YU W, BOIARKINA I, et al.Effects of morphology on the dispersibility of instant whole milk powder[J].Journal of Food Engineering, 2020, 276:109841.
[8] PÍSECKÝ J.Handbook of Milk Powder Manufacture[M].GEA Process Engineering A/S, 2012.
[9] LLOYD R, STEWART H, BAILEY D.Slowly dissolving particles in instant whole milk powder-characterisation and quantitative analysis[J].International Dairy Journal, 2019, 97:65-70.
[10] MCKENNA A B, LLOYD R J, MUNRO P A, et al.Microstructure of whole milk powder and of insolubles detected by powder functional testing[J].Scanning, 1999, 21(5):305-315.
[11] TOIKKANEN O, OUTINEN M, MALAFRONTE L, et al.Formation and structure of insoluble particles in reconstituted model infant formula powders[J].International Dairy Journal, 2018, 82:19-27.
[12] GAIANI C, BOYANOVA P, HUSSAIN R, et al.Morphological descriptors and colour as a tool to better understand rehydration properties of dairy powders[J].International Dairy Journal, 2011, 21(7):462-469.
[13] BOIARKINA I, YE J, PRINCE-PIKE A, et al. The morphology of instant whole milk powder from different industrial plants, Chemeca 2016: Chemical Engineering-Regeneration, Recovery and Reinvention. Melbourne: Engineers Australia, 2016: 945-953.
[14] SHARMA A, JANA A H, CHAVAN R S.Functionality of milk powders and milk-based powders for end use applications—A review[J].Comprehensive Reviews in Food science and Food safety, 2012, 11(5):518-528.
[15] DHANALAKSHMI K, GHOSAL S, BHATTACHARYA S.Agglomeration of food powder and applications[J].Critical reviews in food science and nutrition, 2011, 51(5):432-441.
[16] KHAN A, MUNIR M T, YU W, et al.Wavelength selection for rapid identification of different particle size fractions of milk powder using hyperspectral imaging[J].Sensors, 2020, 20(16):4645.
[17] DING H H, LI B, BOIARKINA I, et al.Effects of morphology on the bulk density of instant whole milk powder[J].Foods, 2020, 9(8):1024.
[18] 李昂, 韩萌, 穆栋梁, 等. 多类不平衡数据分类方法综述. 计算机应用研究, 2022, 39(12):3534-3545.
LI A, HAN M, MU D L, et al. Survey of multi-class imbalanced data classification methods. Application Research of Computers, 2022, 39(12):3534-3545.
[19] NICKERSON A, JAPKOWiCZ N, MILIOS E E.Using unsupervised learning to guide resampling in imbalanced data sets[C].International Workshop on Artificial Intelligence and Statistics, 2001:224-228.
[20] BATISTA G E A P A, PRATI R C, MONARD M C.A study of the behavior of several methods for balancing machine learning training data[J].Acm Sigkdd Explorations Newsletter, 2004, 6(1):20-29.
[21] 尹延顺, 姜欣, 宋涛, 等. 基于偏最小二乘法分析复合型果酒风味成分与感官属性之间的相关性. 食品与发酵工业, 2021, 47(22):234-240.
YIN Y S, JIANG X, SONG T, et al. Correlation analysis on flavor compositions and sensory attributes of compound fruit wines based on partial least squares regression. Food and Fermentation Industries, 2021, 47(22):234-240.
[22] ZHANG H, MIAO Y L, TAKAHASHI H, et al.Amylose analysis of rice flour using near-infrared spectroscopy with particle size compensation[J].Food Science and Technology Research, 2011, 17(4):361-367.
[23] MUNIR M T, WILSON D I, YU W, et al.An evaluation of hyperspectral imaging for characterising milk powders[J].Journal of Food Engineering, 2018, 221:1-10.
[24] 李冰, 郭祀远, 李琳, 等. 人工神经网络在食品工业中的应用. 食品科学, 2003, 24(6):161-164.
LI B, GUO S Y, LI L, et al. The processing of artificial neural network in food industry. Food Science, 2003, 24(6):161-164.
[25] PIOTROWSKI A P, NAPIORKOWSKI J J.Optimizing neural networks for river flow forecasting-evolutionary computation methods versus the levenberg-marquardt approach[J].Journal of Hydrology, 2011, 407(1-4):12-27.
[26] 孔英会, 景美丽. 基于混淆矩阵和集成学习的分类方法研究. 计算机工程与科学, 2012, 34(6):111-117.
KONG Y H, JING M L. Research of the classification method based on confusion matrixes and ensemble learning. Computer Engineering & Science, 2012, 34(6):111-117.
