• EI
  • Scopus
  • 食品科学与工程领域高质量科技期刊分级目录第一方阵T1
  • DOAJ
  • EBSCO
  • 北大核心期刊
  • 中国核心学术期刊RCCSE
  • JST China
  • FSTA
  • 中国精品科技期刊
  • 中国农业核心期刊
  • CA
  • WJCI
  • 中国科技核心期刊CSTPCD
  • 中国生物医学SinoMed
中国精品科技期刊2020
李裕梅,唐润发,杜芳芳,等. 食用煎炸油理化指标统计分析方法研究进展[J]. 食品工业科技,2021,42(7):416−426. doi: 10.13386/j.issn1002-0306.2020080262.
引用本文: 李裕梅,唐润发,杜芳芳,等. 食用煎炸油理化指标统计分析方法研究进展[J]. 食品工业科技,2021,42(7):416−426. doi: 10.13386/j.issn1002-0306.2020080262.
LI Yumei, TANG Runfa, DU Fangfang, et al. Research Progress of Statistical Analysis Methods for Physicochemical Parameters of Edible Frying Oil[J]. Science and Technology of Food Industry, 2021, 42(7): 416−426. (in Chinese with English abstract). doi: 10.13386/ j.issn1002-0306.2020080262.
Citation: LI Yumei, TANG Runfa, DU Fangfang, et al. Research Progress of Statistical Analysis Methods for Physicochemical Parameters of Edible Frying Oil[J]. Science and Technology of Food Industry, 2021, 42(7): 416−426. (in Chinese with English abstract). doi: 10.13386/ j.issn1002-0306.2020080262.

食用煎炸油理化指标统计分析方法研究进展

Research Progress of Statistical Analysis Methods for Physicochemical Parameters of Edible Frying Oil

  • 摘要: 煎炸食品在愉悦味蕾的同时,也面临着油种类的最优抉择和质量安全问题,众多学者在选择何种类型的油,以及煎炸的时间长短方面进行了研究。本文从统计学的角度出发,概述了常用统计分析方法在煎炸油理化指标数据分析中的研究进展。一是,表格分析方式,大多数是将多次重复实验的数据以均值加减标准差的形式放到表格里进行研究,而且分析较多的理化指标是酸值、脂肪酸和极性化合物这三个指标。二是,作图可视化方式:直条图、折线图和散点图;其中,直条图主要用来展现在不同煎炸油、不同煎炸时间、不同煎炸温度、不同加热功率等条件下理化指标的含量情况;折线图主要用来分析理化指标随着煎炸时间、煎炸温度、不同煎炸油、不同煎炸负荷下理化指标的变化情况;散点图主要用来分析理化指标受煎炸时间、煎炸次数、油样密度等的影响,以及理化指标相互之间的关系和理化指标在不同废弃油样中的分布情况等。三是,拟合方式:线性拟合和非线性拟合;其中,线性拟合主要用来分析理化指标随煎炸时间、煎炸次数、介电常数变化等的线性关系,以及用来分析理化指标和近红外光谱之间的线性关系,分析理化指标实测值和其他模型预测值之间的一致性等;非线性拟合主要涉及到多项式拟合、指数函数拟合和逻辑回归的拟合,用来分析某些理化指标随煎炸时间变化的复杂关系,以及根据理化指标将油进行好坏分类等。四是,理化指标间的相关性分析,绝大多数文献是通过计算指标间的皮尔逊相关系数进行的,少部分文献使用了指标间的线性或者非线性拟合方式来分析。五是,针对理化指标数据进行的方差分析,主要用于研究不同煎炸因素下理化指标的变化情况。六是,针对理化指标进行的主成分分析,主要用来分析各个理化指标对煎炸油质量监控的重要性程度等等。通过以上对于煎炸油理化指标的统计分析方法的进展描述,给研究者们提供对于研究方法和研究手段的选择提供思路。

     

    Abstract: Frying food is not only pleasant to taste buds, but also faced with the optimal choice of oil type and quality and safety issues. Many scholars have studied the choice of oil type and the frying time. From the perspective of statistics, this paper summarizes the research progress of common statistical data analysis methods in the analysis of frying oil’s physicochemical parameters. First, with table analysis, data of repeated experiments are put into the table in the form of mean value plus minus standard deviation in most research cases, and the three common analyzed physicochemical parameters are acid value, fatty acid and total polar compounds. Second, with visualization methods of straight bar charts, line charts and scatter plots, the straight bar charts are mainly used to show the content of physicochemical parameters under different frying oil, frying time, frying temperature or power levels of heating; the line charts are mainly used to analyze the changes of physicochemical parameters with frying time, frying temperature, frying oil or frying load; the scatter plots are mainly used to analyze the influence of frying time, frying times or oil density on physicochemical parameters, as well as the relationship between physicochemical parameters or the distribution of physicochemical parameters in different waste oil samples. Third, with fitting methods of linear fittings and nonlinear fittings, the linear fittings are mainly used to analyze the linear relationships of physicochemical parameters with frying time, frying times or dielectric constants, as well as the linear relationships between physicochemical parameters and near infrared spectrum, and to analyze the consistency between the measured values of physicochemical parameters and the predicted values of other models; The nonlinear fittings mainly involve the methods of polynomial fitting, exponential function fitting and logistic regression and they are used to analyze the complex relationships between some physicochemical parameters and frying time, and to classify oil according to physicochemical parameters. Fourth, about the correlation analysis of physicochemical parameters, most of the literatures are carried out by calculating the Pearson correlation coefficients between physicochemical parameters, and a small number of literatures use the linear or nonlinear fitting methods between physicochemical parameters to analyze the correlation. Fifth, Analysis of variance is mainly used to study the changes of physicochemical parameters under different frying factors. Sixth, the principal component analysis is mainly used to analyze the importance of each physicochemical parameter to the quality control of frying oil, and so on. Through the above description of the progress of the statistical analysis methods for the physicochemical parameters of frying oil, it provides the researchers with some ideas for the selection of research methods and means.

     

/

返回文章
返回