Method and system for constructing comprehensive evaluation model of quality of crispy bread

Abstract

The invention relates to the technical field of bread quality evaluation, and discloses a method and system for constructing a comprehensive evaluation model for the quality of crispy bread, and the method comprises the steps of sample collection and grouping, quality index determination, data preprocessing, core index screening, weight determination, model construction, model verification and the like. The method comprises the following steps: collecting short bread samples, measuring multiple quality indexes, and screening out core indexes by using correlation analysis, principal component analysis and clustering analysis; and determining a weight coefficient of each core index by adopting a proportional scale method, and constructing a comprehensive evaluation model for the quality of the crispy bread. And finally, verifying the model by using another group of samples, evaluating the fitting degree between the model score and the sensory score through regression analysis, and ensuring the accuracy of the model. The comprehensive evaluation model provided by the invention is objective, accurate and easy to operate, provides a new thought and method for evaluating the quality of the crispy bread, and is beneficial to promoting the healthy development of the crispy bread industry.

Description

Technical Field

[0001] The present invention relates to the technical field of bread quality evaluation, and in particular to a method and system for constructing a comprehensive evaluation model for puff pastry quality. Background Art

[0002] Bread, as one of the most popular foods in the world, has a huge market size. Among them, puff pastry has become a popular choice for consumers with its distinct layers, soft texture, unique flavor and aroma. However, with the continuous improvement of people's quality of life, consumers' pursuit of puff pastry is no longer limited to satisfying basic satiety needs, but pays more attention to the overall improvement of its flavor and quality. Unfortunately, there are many brands of puff pastry on the market, but the quality is uneven, which undoubtedly brings troubles to consumers' choices.

[0003] At present, the quality evaluation of puff pastry mainly draws on the evaluation indicators and methods of traditional flour products such as bread, steamed buns, and noodles. These evaluation methods mostly rely on sensory evaluation, moisture content, acidity, peroxide value, acid value and other test items specified in national standards. Although these methods can reflect the quality of bread to a certain extent, they have many limitations. In particular, the sensory evaluation method, although it can intuitively reflect the edible quality of bread, its evaluation process requires professional sensory training for an expert group, which is not only time-consuming, labor-intensive and costly, but also the evaluation results are easily affected by personal preferences and regional differences, making it difficult to achieve standardization and unification.

[0004] In addition, there are relatively few in-depth studies on the specific field of puff pastry, and most of them focus on the discussion of a single quality characteristic. There is a lack of comprehensive research on the relationship between multiple indicators and puff pastry quality, and there is a lack of a scientific, reasonable and comprehensive quality evaluation system. Therefore, exploring an objective, accurate and easy-to-operate puff pastry evaluation method has become an urgent problem to be solved. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for constructing a comprehensive evaluation model for puff pastry quality, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for constructing a comprehensive evaluation model for puff pastry quality, the method comprising:

[0007] Step 1: Sample collection and grouping: Collect puff pastry samples and randomly divide them into two groups, one of which is used to build a comprehensive model for puff pastry quality evaluation, and the other is used for model verification;

[0008] Step 2: Determination of quality indicators: Determination of multiple quality indicators of the sample group used to build the model;

[0009] Step 3: Data preprocessing: preprocess the measured quality index data, including data initialization and normalization;

[0010] Step 4: Core index screening: Use correlation analysis, principal component analysis and cluster analysis methods to screen out the core indicators that have a significant impact on the quality evaluation of puff pastry from the multiple quality indicators measured; reveal the relationship between indicators through correlation analysis; extract the principal components that can represent most of the information of the original data through principal component analysis dimensionality reduction; and further verify the results of principal component analysis through cluster analysis;

[0011] Step 5: Determine the weight: Use the proportional scaling method to establish a judgment matrix, and calculate the weight coefficient of each core indicator based on the judgment matrix;

[0012] Step 6: Model construction: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed;

[0013] Step 7. Model verification: Use another group of samples to verify the constructed comprehensive quality evaluation model. The quality indicators of this group of samples are measured and their comprehensive scores are calculated. Sensory evaluation is performed to obtain sensory scores. The model scores are regressed with the sensory scores, and the accuracy of the model is evaluated by the goodness of fit.

[0014] Preferably, the quality indicators of the puff pastry include chewiness, fat content, peroxide value, recovery, moisture, carbohydrates and acidity.

[0015] Preferably, the method of initialization processing is:

[0016] The absolute value of the distance between each quality index value and the ideal value is used to represent the initialization value. The calculation formula is: In the formula, Xi is the original data of the i-th index of the sample; X0 is the ideal value of each index of the sample; X i The value of the sample's i-th indicator after initialization.

[0017] Preferably, the normalization method is:

[0018] The normalized values ​​of each indicator are calculated using the following formula: Where, X i ' is the value of the initialized i-th indicator in the sample, X' imax is the maximum value of the initialized value of the i-th indicator in the sample; Z is the normalized value of the i-th indicator in the sample.

[0019] Preferably, the formula for calculating the comprehensive score of the puff pastry quality comprehensive evaluation model is:

[0020]

[0021] Where Y i The comprehensive score for puff pastry; a ij is the value of each quality index after standardization; ω j For indicator a ij The corresponding weight coefficient.

[0022] Preferably, in step 5, in order to determine the rationality of the judgment matrix, the judgment matrix is ​​further subjected to consistency ratio calculation, and the calculation formula is:

[0023]

[0024] In the formula, CI is the consistency index, CR is the consistency ratio, λ max is the maximum eigenvalue of the judgment matrix; n is the order of the judgment matrix; RI is the average random consistency value corresponding to n.

[0025] Preferably, the cluster analysis uses a K-means clustering method to cluster the quality indicators and verify the results of the principal component analysis.

[0026] Preferably, when CR is less than 0.10, the judgment matrix is ​​considered to have acceptable consistency.

[0027] Preferably, a system for constructing a comprehensive evaluation model for puff pastry quality comprises:

[0028] Sample management and grouping module: used to collect puff pastry samples and randomly divide the samples into model building group and model verification group;

[0029] Quality index determination module: Determine multiple quality indexes of samples in the model building group and obtain original data;

[0030] Data preprocessing module: initialization and normalization of the measured quality index data;

[0031] Core index screening module: Use correlation analysis, principal component analysis and cluster analysis methods to screen out core indicators that have a significant impact on the quality evaluation of puff pastry from multiple quality indicators;

[0032] Weight determination module: Use the proportional scaling method to establish a judgment matrix, and calculate the weight coefficient of each core indicator based on the judgment matrix;

[0033] Model building module: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed;

[0034] Model verification module: The quality indicators of the samples in the verification model group are measured, their comprehensive scores are calculated, and sensory evaluation is performed to obtain sensory scores. The fit between the model score and the sensory score is evaluated through regression analysis to verify the accuracy of the model.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention uses mathematical statistics methods such as correlation analysis, principal component analysis and cluster analysis to screen out core indicators that have a significant impact on the quality evaluation of puff pastry from multiple quality indicators, avoiding the one-sidedness of single indicator evaluation and improving the objectivity and accuracy of the evaluation. At the same time, the hierarchical analysis method is used to determine the weight coefficient of each core indicator, further enhancing the rationality and reliability of the evaluation result.

[0037] The comprehensive quality evaluation model constructed by the present invention is based on objective data analysis and a scientific weight determination method, which avoids the problem of sensory evaluation being easily affected by personal preferences and regional differences, realizes the standardization and unification of the quality evaluation of puff pastry, and provides a reliable basis for the quality comparison of puff pastry of different brands and regions. Compared with the traditional sensory evaluation method, the comprehensive quality evaluation model constructed by the present invention does not require the organization of a professional sensory evaluation team for long-term training. It only needs to measure the quality index of the sample and process the data to obtain the evaluation result, which greatly reduces the evaluation cost and time and improves the evaluation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A step diagram of a method for constructing a comprehensive evaluation model for puff pastry quality according to the present invention;

[0039] Figure 2 Flowchart for selecting core indicators and determining weights. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1-2 The present invention provides a method for constructing a comprehensive evaluation model for puff pastry quality, the method comprising:

[0042] Step 1: Sample collection and grouping: Randomly collect puff pastry samples of different brands and production processes in the market to ensure the diversity and representativeness of the samples. The collected samples are randomly divided into two groups, one as the modeling group for building the quality evaluation model; the other as the verification group for subsequent model verification.

[0043] Step 2: Determination of quality indicators: Multiple quality indicators were measured for each puff pastry sample in the modeling group, including chewiness, fat content, peroxide value, recovery, moisture, carbohydrates and acidity. The determination was carried out using the method specified by the industry standard to ensure the accuracy and reliability of the data.

[0044] Step 3: Data preprocessing: Preprocess the measured quality index data. First, perform data initialization to remove outliers or missing values; then perform normalization to convert each index data to the same dimension for subsequent analysis.

[0045] Step 4: Screening of core indicators: Use correlation analysis to reveal the relationship between various quality indicators and eliminate redundant indicators with strong correlation; then perform principal component analysis to extract the principal components that can represent most of the information of the original data through dimensionality reduction; finally, perform cluster analysis to further verify the results of principal component analysis based on the similarities and differences of the indicators, so as to screen out the core indicators that have a significant impact on the quality evaluation of puff pastry.

[0046] Step 5. Determine the weights: Use the proportional scaling method to establish a judgment matrix, compare the importance of core indicators pairwise based on expert experience or actual data, and obtain the judgment matrix; then use the characteristic root method or the sum-product method to calculate the eigenvector of the judgment matrix, that is, the weight coefficient of each core indicator; finally, perform a consistency test to ensure the rationality and reliability of the weight coefficient.

[0047] Step 6: Model construction: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed using the linear weighted sum method or the fuzzy comprehensive evaluation method. The specific form of the model can be selected according to the actual situation to ensure that the model can accurately reflect the quality characteristics of puff pastry.

[0048] Step 7: Model verification: Use the puff pastry samples from the verification group to verify the constructed comprehensive quality evaluation model. Perform the same quality index test on the samples in the verification group and calculate their comprehensive scores; at the same time, perform sensory evaluation to obtain sensory scores. Perform regression analysis on the model score and the sensory score, and evaluate the accuracy and reliability of the model by the goodness of fit. If the model score is highly correlated with the sensory score, it means that the model has high predictive ability and practical value.

[0049] The present invention will be further described below in conjunction with Examples 1 to 2:

[0050] Embodiment 1:

[0051] In the multi-index evaluation system, since the nature of each evaluation index is different, with different dimensions and orders of magnitude, in order to avoid the influence of these factors on the evaluation results and ensure the reliability of the results, the index data needs to be initialized. The method is to use the absolute value of the distance between each quality index value and the ideal value to represent the initialization value.

[0052]

[0053] Where Xi is the original data of the i-th index of the sample; X0 is the ideal value of each index of the sample; X i The value of the i-th indicator of the sample after initialization.

[0054] After initialization, the range of each evaluation index value is X0. In order to eliminate the influence of different dimensions and orders of magnitude on the comprehensive evaluation and facilitate the subsequent comprehensive quality evaluation, the data needs to be normalized. The forward and normalization processing methods are:

[0055]

[0056] Where, X i ' is the value of the initialized i-th indicator in the sample, X' imax is the maximum value of the initialized value of the i-th indicator in the sample; Z is the normalized value of the i-th indicator in the sample.

[0057] After the 7 core indicators of puff pastry are normalized and normalized, the closer the original indicator value is to the ideal value, the closer the normalized and normalized value is to 1 (0≤Xi≤1).

[0058] Embodiment 2:

[0059] In order to determine the rationality of the judgment matrix, the consistency ratio of the judgment matrix is ​​further calculated, and the calculation formula is:

[0060]

[0061] In the process of constructing a comprehensive evaluation model for puff pastry quality, the selection of core indicators and the determination of weights are crucial steps. The following is a detailed description of these two steps in combination with specific quality indicators and mathematical methods.

[0062] Correlation analysis: Correlation analysis was performed on the measured puff pastry quality indicators (including chewiness, fat content, peroxide value, recovery, moisture, carbohydrates and acidity). The Pearson correlation coefficient or Spearman rank correlation coefficient was used to calculate the correlation coefficients between the indicators to reveal the relationship between the indicators. For indicators with high correlation coefficients (such as greater than 0.8), only one representative indicator was retained, considering that there may be redundancy.

[0063] Principal component analysis: Perform principal component analysis to reduce dimensionality. After standardizing the measured quality index data, construct a covariance matrix or correlation matrix, and calculate its eigenvalues ​​and eigenvectors. According to the size of the eigenvalue, select the principal component that can explain most of the variance of the original data (such as cumulative contribution rate exceeding 85%). Each principal component is a linear combination of the original indicators, and the principal components are independent of each other.

[0064] Cluster analysis: K-means clustering method is used to perform cluster analysis on quality indicators to further verify the results of principal component analysis. First, according to the results of principal component analysis, an appropriate number of cluster centers (K value) are selected, and then cluster centers and sample attribution are iteratively updated until the cluster centers no longer change or the preset number of iterations is reached. Through cluster analysis, the distribution of each indicator in the cluster can be observed to verify whether the principal components extracted by principal component analysis are reasonable.

[0065] Weight determination:

[0066] ① Establish a judgment matrix:

[0067] Use the proportional scaling method (such as 1-9 scale) to establish a judgment matrix. For the selected core indicators, compare the importance of the indicators in pairs based on expert experience or actual data, and assign corresponding scale values. For example, if indicator A is much more important than indicator B, assign a scale value of 9 to B; if A and B are equally important, assign a scale value of 1.

[0068] ②Calculate the weight coefficient:

[0069] According to the judgment matrix, the weight coefficient of each core indicator is calculated. Commonly used methods include characteristic root method, sum product method, etc. Taking the characteristic root method as an example, first calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, then normalize the eigenvector to obtain the weight coefficient of each indicator.

[0070] ③Consistency test:

[0071] In order to verify the rationality of the judgment matrix, a consistency test is performed. The consistency index CI is calculated, and its formula is:

[0072]

[0073] Among them, λ max is the maximum eigenvalue of the judgment matrix, and n is the order of the judgment matrix.

[0074] Next, calculate the consistency ratio CR, the formula is:

[0075]

[0076] Among them, RI is the average random consistency value corresponding to n, which can be obtained by looking up a table or calculating.

[0077] When CR is less than 0.10, the judgment matrix is ​​considered to have acceptable consistency and the calculated weight coefficient is reasonable. Otherwise, the judgment matrix needs to be adjusted and the weight coefficient needs to be recalculated until the consistency requirement is met.

[0078] Through the above steps, the core indicators that have a significant impact on the quality evaluation of puff pastry can be screened out, and the weight coefficient of each indicator can be determined, providing a basis for constructing a comprehensive quality evaluation model.

[0079] Embodiment 3:

[0080] The following is a detailed description of the specific implementation method of quality indicators and comprehensive score calculation.

[0081] Quality index determination:

[0082] The following seven key quality indicators are selected in this implementation:

[0083] Chewability: reflects the ease of chewing puff pastry in the mouth, measured by professional instruments.

[0084] Fat content: The fat content in puff pastry has an important influence on its taste and flavor and is determined by chemical methods.

[0085] Peroxide value: reflects the degree of oxidation of fat in puff pastry and is an important indicator for evaluating its freshness and stability.

[0086] Resilience: A measure of the ability of puff pastry to return to its original shape after being stressed, obtained through physical testing.

[0087] Moisture: The moisture content of puff pastry directly affects its taste and shelf life, and is determined by the drying method.

[0088] Carbohydrate: As one of the main nutrients of puff pastry, its content is determined by chemical method.

[0089] Acidity: reflects the pH value of puff pastry, which is determined by pH meter or titration.

[0090] Comprehensive score calculation formula:

[0091]

[0092] Where Y i The comprehensive score for puff pastry; a ij is the value of each quality index after standardization; ω j For indicator a ij The corresponding weight coefficient.

[0093] Assuming that the effective weights of puff pastry are determined based on hierarchical analysis, the weight coefficients of the core indicators of puff pastry are: chewiness (0.350), fat (0.155), peroxide value (0.061), resilience (0.238), moisture (0.038), carbohydrates (0.098), and acidity (0.061).

[0094] It can be obtained that the comprehensive score Y = 0.350×chewiness+0.155×fat+0.061×peroxide value+0.238×recovery+0.038×water+0.098×carbohydrate+0.061×acidity.

[0095] The present invention also includes a system for constructing a comprehensive evaluation model for puff pastry quality, the system comprising:

[0096] Sample management and grouping module: used to collect puff pastry samples and randomly divide the samples into model building group and model verification group;

[0097] Quality index determination module: Determine multiple quality indexes of samples in the model building group and obtain original data;

[0098] Data preprocessing module: initialization and normalization of the measured quality index data;

[0099] Core index screening module: Use correlation analysis, principal component analysis and cluster analysis methods to screen out core indicators that have a significant impact on the quality evaluation of puff pastry from multiple quality indicators;

[0100] Weight determination module: Use the proportional scaling method to establish a judgment matrix, and calculate the weight coefficient of each core indicator based on the judgment matrix;

[0101] Model building module: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed;

[0102] Model verification module: The quality indicators of the samples in the verification model group are measured, their comprehensive scores are calculated, and sensory evaluation is performed to obtain sensory scores. The fit between the model score and the sensory score is evaluated through regression analysis to verify the accuracy of the model.

[0103] The implementation of the system refers to the above embodiment and will not be described in detail in the specification.

[0104] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0105] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a comprehensive evaluation model for puff pastry quality, characterized in that: The method comprises: Step 1: Sample collection and grouping: Collect puff pastry samples and randomly divide them into two groups, one of which is used to build a comprehensive model for puff pastry quality evaluation, and the other is used for model verification; Step 2: Determination of quality indicators: Determination of multiple quality indicators of the sample group used to build the model; Step 3: Data preprocessing: preprocess the measured quality index data, including data initialization and normalization; Step 4: Core index screening: Use correlation analysis, principal component analysis and cluster analysis methods to screen out the core indicators that have a significant impact on the quality evaluation of puff pastry from the multiple quality indicators measured; reveal the relationship between indicators through correlation analysis; extract the principal components that can represent most of the information of the original data through principal component analysis dimensionality reduction; and further verify the results of principal component analysis through cluster analysis; Step 5: Determine the weight: Use the proportional scaling method to establish a judgment matrix, and calculate the weight coefficient of each core indicator based on the judgment matrix; Step 6: Model construction: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed; Step 7. Model verification: Use another group of samples to verify the constructed comprehensive quality evaluation model. The quality indicators of this group of samples are measured and their comprehensive scores are calculated. Sensory evaluation is performed to obtain sensory scores. The model scores are regressed with the sensory scores, and the accuracy of the model is evaluated by the goodness of fit.

2. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 1, characterized in that: The quality indicators of the puff pastry include chewiness, fat content, peroxide value, recovery, moisture, carbohydrates and acidity.

3. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 2, characterized in that: The initialization method is: The absolute value of the distance between each quality index value and the ideal value is used to represent the initialization value. The calculation formula is: In the formula, Xi is the original data of the i-th index of the sample; X0 is the ideal value of each index of the sample; X i The value of the sample's i-th indicator after initialization.

4. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 2, characterized in that: The normalization method is: The normalized values ​​of each indicator are calculated using the following formula: Where, X i ' is the value of the initialized i-th indicator in the sample, X' imax is the maximum value of the initialized value of the i-th indicator in the sample; Z is the normalized value of the i-th indicator in the sample.

5. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 1, characterized in that: The formula for calculating the comprehensive score of the puff pastry quality comprehensive evaluation model is: Where Y i The comprehensive score for puff pastry; a ij is the value of each quality index after standardization; ω j For indicator a ij The corresponding weight coefficient.

6. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 1, characterized in that: In step 5, in order to determine the rationality of the judgment matrix, the consistency ratio of the judgment matrix is ​​further calculated, and the calculation formula is: In the formula, CI is the consistency index, CR is the consistency ratio, λ max is the maximum eigenvalue of the judgment matrix; n is the order of the judgment matrix; RI is the average random consistency value corresponding to n.

7. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 1, characterized in that: The cluster analysis adopts K-means clustering method to cluster the quality indicators and verify the results of principal component analysis.

8. The method for constructing a comprehensive evaluation model for puff pastry quality according to claim 1, characterized in that: When CR is less than 0.10, the judgment matrix is ​​considered to have acceptable consistency.

9. A system for constructing a comprehensive evaluation model for puff pastry quality, characterized in that: The system comprises: Sample management and grouping module: used to collect puff pastry samples and randomly divide the samples into model building group and model verification group; Quality index determination module: Determine multiple quality indexes of samples in the model building group and obtain raw data; Data preprocessing module: initialization and normalization of the measured quality index data; Core index screening module: Use correlation analysis, principal component analysis and cluster analysis methods to screen out core indicators that have a significant impact on the quality evaluation of puff pastry from multiple quality indicators; Weight determination module: Use the proportional scaling method to establish a judgment matrix, and calculate the weight coefficient of each core indicator based on the judgment matrix; Model building module: Based on the selected core indicators and their corresponding weight coefficients, a comprehensive evaluation model for the quality of puff pastry is constructed; Model verification module: The quality indicators of the samples in the verification model group are measured, their comprehensive scores are calculated, and sensory evaluation is performed to obtain sensory scores. The fit between the model score and the sensory score is evaluated through regression analysis to verify the accuracy of the model.

Images