Model and method for predicting sensory acidity by using total acidity and effective acidity

Abstract

The invention discloses a model and a method for predicting sensory acidity by using total acidity and effective acidity, and the method for constructing the prediction model for predicting sensory acidity by using total acidity and effective acidity comprises the following steps: adopting a specified acidic substance as a reference substance; the aqueous solutions of the acidic substances with different concentrations are represented by numerical values of different sensory acidity scales through professional evaluation personnel, and the sensory acidity scales are obtained; the method comprises the following steps: respectively preparing aqueous solutions of different addition amounts of buffer salt, hydrophilic colloid and protein under different concentrations of acidic substances, evaluating the sensory acidity of the aqueous solutions by professional evaluation personnel, and simultaneously detecting the pH value and titration acidity of the aqueous solutions; and performing linear regression analysis by taking the pH value and the titration acidity as independent variables and the sensory acidity as a dependent variable according to the data of the three variables of the sensory acidity, the pH value and the titration acidity in the database to obtain a prediction model. The sensory acidity of the product can be accurately predicted without tasting by sensory personnel.

Description

Technical Field

[0001] The present invention relates to the field of sensory detection, and particularly to a model and method for predicting sensory acidity using total acidity and effective acidity. Background Art

[0002] Sour taste is an important basic taste of humans. The sour taste perception is mainly caused by proton hydrogen ions. Existing research shows that the generation of sour taste perception is due to hydrogen ions or protonated molecules entering the cytoplasm through ion channels on the cell membrane, acidifying it, causing the transmission of cell signals, and further generating taste.

[0003] Food is a complex system of multiple components. The salt ions, proteins, and hydrophilic colloids in it are all charged substances, which will affect the ionization of organic acids and thus affect the acidity of the food system. In the food industry, the acidity of acids is quantitatively evaluated by detecting the total acidity or effective acidity of food. Among them, the total acidity refers to the titratable acidity, which is the total amount of substances that can undergo neutralization reactions with strong bases; the effective acidity refers to the pH value, which is the activity of hydrogen ions in the solution.

[0004] Due to the complexity of the food system, in actual production, there will be problems where the total acidity or effective acidity is the same, but the sensory acidity of the product is not uniform. Neither referring to the total acidity nor the effective acidity can accurately reflect the sensory acidity. Among them, Chinese Patent Application CN 108344835 A has disclosed an attempt to predict sensory acidity using the substance concentrations of different types of sour substances. Its sensory acidity starts from the sour taste detection threshold, with the gradient being the sour taste difference threshold. However, this model is only for predicting in a pure water system and does not consider the factors of the product system. In the actual application of a product system with complex compositions, the detection of the concentration of quantitative substances is complex. Therefore, it does not have practical application value. Summary of the Invention

[0005] Therefore, the problem to be solved by the present invention is that currently, the total acidity, effective acidity, and the concentration of specific substances disclosed cannot accurately reflect the sensory acidity. Thus, a model and method for predicting sensory acidity using total acidity and effective acidity to solve the above problems are provided.

[0006] For this purpose, the present invention provides the following technical solutions:

[0007] A method for constructing a prediction model for predicting sensory acidity using total acidity and effective acidity, comprising:

[0008] Establishing a sensory acidity scale: Using a specified acidic substance as a reference substance, for aqueous solutions of acidic substances with different concentrations, different sensory acidity scale values are used to represent them by professional tasters, and a sensory acidity scale is obtained;

[0009] Database establishment: Aqueous solutions with different addition amounts of buffer salts, hydrocolloids, and proteins under different concentrations of acidic substances are prepared respectively. Professional tasters evaluate the sensory acidity of the aqueous solutions, and at the same time, the pH value and titratable acidity of the aqueous solutions are measured.

[0010] Construction of the prediction model: Based on the data of the three variables of sensory acidity, pH value, and titratable acidity in the database, with the pH value and titratable acidity as independent variables and the sensory acidity as the dependent variable, linear regression analysis is performed to obtain the prediction model.

[0011] The buffer salt is one or more of sodium citrate, potassium citrate, sodium bicarbonate, and sodium tripolyphosphate;

[0012] And / or, the hydrocolloid is one or more of pectin, agar, gellan gum, soluble soybean polysaccharide, xanthan gum, sodium carboxymethyl cellulose, carrageenan, and sodium alginate;

[0013] And / or, the protein source is one or more of raw cow milk, whole milk powder, skim milk powder, and milk protein powder.

[0014] The specified acidic substance is citric acid.

[0015] The corresponding relationship between the citric acid concentration in the aqueous solution and the sensory acidity scale value is as follows:

[0016] The sensory acidity scale corresponding to 0.0125% citric acid is 1, the sensory acidity scale corresponding to 0.025% citric acid is 2, the sensory acidity scale corresponding to 0.050% citric acid is 3, the sensory acidity scale corresponding to 0.075% citric acid is 4, the sensory acidity scale corresponding to 0.100% citric acid is 5, the sensory acidity scale corresponding to 0.150% citric acid is 6, the sensory acidity scale corresponding to 0.200% citric acid is 7, the sensory acidity scale corresponding to 0.250% citric acid is 8, the sensory acidity scale corresponding to 0.300% citric acid is 9, the sensory acidity scale corresponding to 0.400% citric acid is 10, the sensory acidity scale corresponding to 0.500% citric acid is 11, and the sensory acidity scale corresponding to 0.600% citric acid is 12.

[0017] The addition amount of the buffer salt is 0.02 - 0.04%;

[0018] And / or, the addition amount of the hydrocolloid is 0.1 - 0.5%;

[0019] And / or, the addition amount of the protein is 0.5 - 1.5%.

[0020] The hydrocolloid includes acidic hydrocolloids, neutral hydrocolloids, and basic hydrocolloids;

[0021] When the hydrocolloid is pectin, the addition amount is 0.2-0.3%; when the hydrocolloid is soluble soybean polysaccharide, the addition amount is 0.1-0.5%; when the hydrocolloid is sodium carboxymethyl cellulose, the addition amount is 0.3-0.4%.

[0022] A prediction model for predicting sensory acidity using total acidity and effective acidity is obtained by the above construction method.

[0023] The prediction model is: Sensory acidity = -1.6361 * pH value + 0.06291 * Titratable acidity + 8.5259.

[0024] The application of the above prediction model in predicting sensory acidity.

[0025] The prediction process is: Detect the pH value and titratable acidity of the test substance, and substitute the detected pH value and titratable acidity into the prediction model to obtain the sensory acidity of the test substance.

[0026] The technical solution of the present invention has the following advantages:

[0027] The construction method of a prediction model for predicting sensory acidity using total acidity and effective acidity provided by the present invention is to establish a quantitative evaluation standard for sensory acidity for acidic dairy products. The present invention can effectively establish a prediction model for sensory acidity using total acidity and effective acidity. This prediction model can accurately predict the sensory acidity of products by detecting the titratable acidity and pH value of the products without the need for sensory personnel to taste, and has the advantages of efficiently and accurately predicting, comparing and analyzing the taste of products. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0029] Figure 1 It is the residual analysis result diagram of 60 groups of modeling data in Example 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The following embodiments are provided to better further understand the present invention, which is not limited to the best embodiment, and does not limit the content and protection scope of the present invention. Any product obtained by anyone under the inspiration of the present invention or by combining the features of the present invention with other prior art features and being the same or similar to the present invention falls within the protection scope of the present invention.

[0031] For those without specific experimental procedures or conditions noted in the examples, the operations or conditions of the conventional experimental procedures described in the literature in this field can be followed. For reagents or instruments without the manufacturer noted, they are all conventional reagent products that can be obtained commercially.

[0032] The raw materials used in the present invention are all conventional commercially available raw materials, and the raw material percentages are all mass percentages.

[0033] Example 1:

[0034] A method for constructing a prediction model for predicting sensory acidity using total acidity and effective acidity, comprising:

[0035] 1. Definition of sensory acidity standard

[0036] Using citric acid as a reference substance, prepare citric acid solutions with deionized water according to the mass percentage concentration of citric acid in Table 1. Professional tasters taste the citric acid solutions and remember the acid intensity, and detect the pH value and titratable acidity of the citric acid solutions according to GB5009.237 and the first method of GB5009.237.

[0037] Table 1: Sensory acidity scale

[0038]

[0039] 2. Influence of dairy system factors on the sensory acidity and chemical acidity of acid substances

[0040] 2.1 Influence of buffer salts:

[0041] Add 0.02%, 0.03%, and 0.04% sodium citrate to citric acid solutions with different concentrations respectively. After sufficient dispersion, professional tasters evaluate the sensory acidity of the solutions, and at the same time detect the chemical acidity (pH, titratable acidity) of the solutions.

[0042] 2.2 Influence of hydrocolloids:

[0043] Select 3 kinds of colloids (acidic colloid - pectin, neutral colloid - soluble soybean polysaccharide, basic colloid - sodium carboxymethylcellulose), and prepare aqueous solutions of colloids with different concentrations; among them, the addition amounts of pectin are 0.2%, 0.25%, and 0.3% respectively, the addition amounts of soluble soybean polysaccharide are 0.1%, 0.3%, and 0.5% respectively, and the addition amounts of sodium carboxymethylcellulose are 0.31%, 0.35%, and 0.39% respectively. Add different doses of citric acid to the colloidal solutions with different concentrations. After sufficient dispersion, professional tasters evaluate the sensory acidity of the solutions, and at the same time detect the chemical acidity (pH, titratable acidity) of the solutions.

[0044] 2.3 Influence of proteins:

[0045] Prepare solutions with protein contents of 0.5%, 1.0%, and 1.5% using milk protein powder. Use 0.5% soybean polysaccharide as a protein protectant. Add different doses of citric acid to protein solutions with different concentrations. After sufficient dispersion, professional tasters evaluate the sensory acidity of the solutions, and at the same time, measure the chemical acidity (pH, titratable acidity) of the solutions.

[0046] 3. Establish a prediction model for sensory acidity

[0047] Use the data in 2 above to determine the relationship among the three variables of solution sensory acidity, pH value, and titratable acidity. The results shown in Table 2 below are obtained through correlation test analysis.

[0048] Table 2: Correlation matrix (Pearson correlation coefficient, significance level α = 0.05)

[0049]

[0050] It can be seen from the statistical results of the correlation test analysis that the pH value is significantly negatively correlated with the sensory acidity, and the titratable acidity is significantly positively correlated with the sensory acidity.

[0051] Perform linear regression analysis with the pH value and titratable acidity as independent variables and the sensory acidity as the dependent variable, and the regression results shown in Table 3 below are obtained.

[0052] Table 3: Bivariate regression analysis

[0053]

[0054] It can be seen from the above bivariate regression analysis that the correlation coefficient is at a relatively high level, and the Durbin-Watson statistic test is between 0 and 4, indicating that the data is independent; it is proved that using the pH value and titratable acidity together to predict the sensory acidity can obtain relatively accurate results.

[0055] 4. Verify the credibility of the prediction model

[0056] Perform variance analysis on the prediction model for predicting sensory acidity, and the analysis results are shown in Table 4 below.

[0057] Table 4: Model variance analysis

[0058]

[0059] It can be seen from the above analysis results that the P value of the prediction model is less than 0.05, which proves that the model is successfully constructed and has statistical significance.

[0060] 5. Residual analysis to test the prediction results of the model

[0061] Use residual analysis to test the regression effect and the quality of the modeling sample data, and determine whether the model selection is reasonable; specifically, randomly select 60 groups of data for modeling for residual analysis. The results of the residual analysis of these 60 groups of modeling data are shown in Table 5 below and Figure 1 as follows.

[0062] Table 5: Residual Analysis of the Regression Model

[0063]

[0064]

[0065]

[0066] From the results of the above 60 groups of residual analysis, it can be seen that the standardized residuals are between [-2 and 2], and the standardized residuals follow a normal distribution. The regression model describing the variables "pH value", "titratable acidity" and "sensory acidity" is reasonable.

[0067] Example 2:

[0068] An application of a prediction model for predicting sensory acidity using total acidity and effective acidity in evaluating the sensory acidity of milk beverages; specifically, use the prediction model constructed in Example 1 to predict the sensory acidity of 4 milk beverages using citric acid as an ingredient. The 4 milk beverages are Mengniu Yogurt Drink (A), GO Chang Lactic Acid Bacteria Drink (B), Bright LOOK Yogurt Drink (C), and Wahaha Shuangwaiwai (D), and evaluate the sensory acidity through actual measurement, and calculate the relative prediction error.

[0069]

[0070] The detection and prediction results are shown in Table 6 below.

[0071] Table 6: Prediction of Sensory Acidity by the Bivariate Regression Model

[0072]

[0073] From the data in the above table, it can be seen that the relative errors of predicting sensory acidity are all less than 10%, and the prediction results are relatively accurate, and can distinguish the taste differences of products with the same titratable acidity, different pH values, and different titratable acidity and the same pH value.

[0074] The model construction method of the present invention is not only applicable to predicting the sensory acidity of the above-mentioned milk beverages using citric acid as the reference substance, but also applicable to a variety of organic acids. Just replace citric acid with organic acid types such as lactic acid, malic acid, and phosphoric acid in the step of defining the sensory acidity standard in Example 1 to obtain the corresponding prediction models for lactic acid, malic acid, and phosphoric acid respectively, and then select the corresponding model for prediction according to the organic acid type used in the product to be tested.

[0075] Comparative Example 1:

[0076] The difference between this comparative example and Example 1 is that the same data in Example 1 were used to perform linear regression analysis with the pH value or the titratable acidity as the independent variable and the sensory acidity as the dependent variable; the regression results are shown in Table 7 below.

[0077] Table 7: Univariate regression analysis

[0078]

[0079] From the results of the regression analysis, it can be seen that the goodness of fit of the regression equations with the pH value and the titratable acidity as the independent variables is not high, proving that the accuracy of predicting the sensory acidity with a single variable is not high.

[0080] The sensory acidity of the same 4 milk beverages was predicted using the above two single-variable models, and the prediction results are shown in Table 8 and Table 9 below.

[0081] Table 8: Prediction of sensory acidity by the pH value single-variable regression model

[0082]

[0083] Table 9: Prediction of sensory acidity by the titratable acidity single-variable regression model

[0084]

[0085] From the above results, it can be seen that the pH value single-variable regression model cannot distinguish the acid sensation differences of products with the same pH value but different titratable acidities, and the prediction error of the sensory acidity is large, so the single-variable model prediction is inaccurate. The titratable acidity single-variable regression model cannot distinguish the acid sensation differences of products with the same titratable acidity but different pH values, and the prediction error of the sensory acidity is large, so the single-variable model prediction is inaccurate.

[0086] Obviously, the above examples are merely illustrations given for clarity and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all the implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. A method for constructing a prediction model for predicting sensory acidity using total acidity and effective acidity, characterized in that, Including: Establishment of sensory acidity scale: Using a specified acidic substance as a reference substance, the aqueous solutions of acidic substances with different concentrations are represented by the values of different sensory acidity scales by professional tasters, and the sensory acidity scale is obtained. Database establishment: Aqueous solutions of buffer salts, hydrocolloids, and proteins with different addition amounts at different concentrations of acidic substances are prepared respectively. Professional tasters evaluate the sensory acidity of the aqueous solutions, and at the same time, the pH value and titratable acidity of the aqueous solutions are detected. Construction of prediction model: According to the data of three variables, namely sensory acidity, pH value, and titratable acidity, in the database, linear regression analysis is performed with the pH value and titratable acidity as independent variables and the sensory acidity as the dependent variable to obtain the prediction model.

2. The construction method according to claim 1, characterized in that, The buffer salt is one or more of sodium citrate, potassium citrate, sodium bicarbonate, and sodium tripolyphosphate; And / or, the hydrocolloid is one or more of pectin, agar, gellan gum, soluble soybean polysaccharide, xanthan gum, sodium carboxymethyl cellulose, carrageenan, and sodium alginate; And / or, the protein source is one or more of raw cow milk, whole milk powder, skim milk powder, and milk protein powder.

3. The construction method according to claim 1 or 2, characterized in that, The specified acidic substance is citric acid.

4. The construction method according to claim 3, characterized in that, The corresponding relationship between the concentration of citric acid in the aqueous solution and the value of the sensory acidity scale is as follows: The sensory acidity scale corresponding to 0.0125% citric acid is 1, the sensory acidity scale corresponding to 0.025% citric acid is 2, the sensory acidity scale corresponding to 0.050% citric acid is 3, the sensory acidity scale corresponding to 0.075% citric acid is 4, the sensory acidity scale corresponding to 0.100% citric acid is 5, the sensory acidity scale corresponding to 0.150% citric acid is 6, the sensory acidity scale corresponding to 0.200% citric acid is 7, the sensory acidity scale corresponding to 0.250% citric acid is 8, the sensory acidity scale corresponding to 0.300% citric acid is 9, the sensory acidity scale corresponding to 0.400% citric acid is 10, the sensory acidity scale corresponding to 0.500% citric acid is 11, and the sensory acidity scale corresponding to 0.600% citric acid is 12.

5. The construction method according to any one of claims 1-4, characterized in that, The addition amount of the buffer salt is 0.02 - 0.04%; And / or, the addition amount of the hydrocolloid is 0.1 - 0.5%; And / or, the addition amount of the protein is 0.5 - 1.5%.

6. The construction method according to claim 5, characterized in that The hydrocolloid includes acidic hydrocolloids, neutral hydrocolloids, and basic hydrocolloids; And / or, when the hydrocolloid is pectin, the addition amount is 0.2 - 0.3%; when the hydrocolloid is soluble soybean polysaccharide, the addition amount is 0.1 - 0.5%; when the hydrocolloid is sodium carboxymethyl cellulose, the addition amount is 0.3 - 0.4%.

7. A prediction model for predicting sensory acidity using total acidity and effective acidity, characterized in that, Obtained by the construction method according to any one of claims 1 - 6.

8. The prediction model according to claim 7, wherein The prediction model is: Sensory acidity = -1.6361 * pH value + 0.06291 * Titratable acidity + 8.5259.

9. Application of the prediction model according to claim 7 or 8 in predicting sensory acidity.

10. The application according to claim 9, wherein The prediction process is as follows: Detect the pH value and titratable acidity of the test substance, and substitute the detected pH value and titratable acidity into the prediction model to obtain the sensory acidity of the test substance.

Images