关键词: ANN Khlass Sukkary VIS-NIR dates drink milk modeling syrup

来  源:   DOI:10.3390/foods13040524   PDF(Pubmed)

Abstract:
A milk drink flavored with date syrup produced at a lab scale level was evaluated. The production process of date syrup involves a sequence of essential unit operations, commencing with the extraction, filtration, and concentration processes from two cultivars: Sukkary and Khlass. Date syrup was then mixed with cow\'s and camel\'s milk at four percentages to form a nutritious, natural, sweet, and energy-rich milk drink. The sensory, physical, and chemical characteristics of the milk drinks flavored with date syrup were examined. The objective of this work was to measure the physiochemical properties of date fruits and milk drinks flavored with date syrup, and then to evaluate the physical properties of milk drinks utilizing non-destructive visible-near-infrared spectra (VIS-NIR). The study assessed the characteristics of the milk drink enhanced with date syrup by employing VIS-NIR spectra and utilizing a partial least-square regression (PLSR) and artificial neural network (ANN) analysis. The VIS-NIR spectra proved to be highly effective in estimating the physiochemical attributes of the flavored milk drink. The ANN model outperformed the PLSR model in this context. RMSECV is considered a more reliable indicator of a model\'s future predictive performance compared to RMSEC, and the R2 value ranged between 0.946 and 0.989. Consequently, non-destructive VIS-NIR technology demonstrates significant promise for accurately predicting and contributing to the entire production process of the product\'s properties examined.
摘要:
评估了以实验室规模水平生产的用日期糖浆调味的牛奶饮料。日期糖浆的生产过程涉及一系列必要的单元操作,从提取开始,过滤,和两个品种的浓缩过程:Sukkary和Khlass。然后将椰枣糖浆与奶牛和骆驼的牛奶以4%的比例混合,自然,甜,和能量丰富的牛奶饮料。感官,物理,并检查了用日期糖浆调味的乳饮料的化学特性。这项工作的目的是测量用日期糖浆调味的日期水果和牛奶饮料的理化性质,然后利用非破坏性可见近红外光谱(VIS-NIR)对乳饮料的物理性质进行评价。该研究通过采用VIS-NIR光谱并利用偏最小二乘回归(PLSR)和人工神经网络(ANN)分析,评估了用枣糖浆增强的乳饮料的特性。VIS-NIR光谱被证明在估计调味乳饮料的物理化学属性方面非常有效。在这种情况下,ANN模型优于PLSR模型。与RMSEC相比,RMSECV被认为是模型未来预测性能的更可靠指标,R2值介于0.946和0.989之间。因此,非破坏性的VIS-NIR技术在准确预测和促进整个生产过程的产品性能方面显示出巨大的前景。
公众号