Multi-dimensional atlas model pericarpium citri reticulatae year identification method and system and storage medium

Abstract

The invention discloses a method and a system for identifying years of dried orange peel by a multi-dimensional atlas model and a storage medium. The method comprises the following steps: obtaining the composition and content of flavonoid components, amino acids, organic acids and volatile components of dried orange peel in different known years; taking the composition and content of flavonoid components, amino acids, organic acids and volatile components of the dried orange peel as samples in the data set; training an orthogonal partial least square discriminant analysis model by using the data set to obtain a year OPLS-DA standard model of the dried orange peel; and inputting the composition and content of the flavonoid components, the amino acids, the organic acids and the volatile components of the to-be-identified pericarpium citri reticulatae into the pericarpium citri reticulatae year OPLS-DA standard model to obtain the year of the to-be-identified pericarpium citri reticulatae. According to the method, the fingerprint spectrum is constructed by utilizing data of flavonoid components, amino acids, organic acids and volatile components of the pericarpium citri reticulatae, and the aging year of the to-be-identified pericarpium citri reticulatae can be accurately predicted in combination with the orthogonal partial least squares discriminant analysis model.

Description

Technical Field

[0001] The present invention relates to the technical field of traditional Chinese medicine quality analysis and detection, and in particular to a method, system, terminal equipment and computer-readable storage medium for identifying tangerine peel years using a multidimensional graph model. Background Art

[0002] Medicinal tangerine peel is the dried mature peel of the Rutaceae plant Citrus reticulate Blanco and its cultivated varieties. It is mainly divided into tangerine peel and Guang tangerine peel, among which Guang tangerine peel has the best quality. Tangerine peel has extremely high edible and medicinal value and is widely used in the fields of food and medicine. Traditional Chinese medicine believes that tangerine peel has the effects of resolving phlegm and relieving cough, and regulating qi and spleen. Modern pharmacology confirms that tangerine peel is rich in flavonoids, volatile oils, alkaloids, polysaccharides and other chemical substances, and has pharmacological effects such as antioxidant, anti-inflammatory, anti-cancer, antibacterial, and anti-pulmonary fibrosis. In addition, the folks believe that the longer the tangerine peel is aged, the higher its quality is, so there is a saying that "the older the tangerine peel, the better" and "a hundred years of tangerine peel is better than gold". However, the appearance characteristics of tangerine peels of different years are similar, and the quality of tangerine peels varies, resulting in the increasingly serious market chaos of low-year tangerine peels pretending to be high-year tangerine peels, which has caused great damage to the tangerine peel industry. Therefore, exploring simple, fast, efficient and accurate methods to identify the age of tangerine peel is extremely necessary to rectify the chaos in the tangerine peel market and safeguard the interests of the tangerine peel industry.

[0003] At present, the identification methods of tangerine peel year are mostly based on spectral technology, molecular biology technology and chromatography technology. Spectroscopic technology includes infrared spectroscopy, ultraviolet spectroscopy and Raman spectroscopy. Although this method can realize non-destructive detection of tangerine peel, the spectral information of tangerine peel of similar years is slightly different, and it needs to be amplified by a variety of data processing methods. The steps are complicated and cumbersome, prone to errors, and the resolution of the constructed model is low. Molecular biology technology includes DNA barcodes, DNA molecular probes and high-throughput sequencing, etc., which use the differences between the DNA genetic information and microbial composition of tangerine peel of different years to realize the classification of tangerine peel year. However, this method is complicated to operate, costly, and has a long cycle. It is also very susceptible to external environmental pollution, which limits its application. Chromatographic technology includes gas chromatography, liquid chromatography, thin layer chromatography, etc. This method has fast analysis speed, high sensitivity, strong stability, and has been widely used in the field of analysis of natural products of Chinese herbal medicine. At present, some studies have established a tangerine peel year identification model based on chromatographic analysis of volatile components or flavonoid components of tangerine peel. However, the characteristic components of tangerine peel are diverse, and only analyzing a certain type of component is poorly representative, and it is impossible to fully reflect the quality differences of different tangerine peels. The identification accuracy of the constructed model is low. Amino acids and organic acids are the main contributors to the taste of tangerine peel and play an important role in evaluating the quality of Xinhui tangerine peel. Amino acids not only provide fresh, sweet, bitter and astringent tastes, but are also important precursors for the synthesis of a variety of aromatic compounds and volatile components, which contribute to the formation of the special aroma and flavor of tangerine peel. Organic acids are the main source of the sour taste of tangerine peel. They can not only enhance the flavor of amino acids, but are also important intermediate components in multiple metabolic pathways. Therefore, it is an effective way to achieve rapid, efficient and accurate identification of tangerine peel years by fully considering the composition of characteristic components that cause the quality differences of tangerine peel in different years, constructing a fingerprint of characteristic components of tangerine peel from multiple dimensions, and exploring the quality differences of different tangerine peels. However, there are no relevant reports yet, and further exploration is urgently needed. Summary of the invention

[0004] In view of this, the present invention provides a method, system, terminal device and computer-readable storage medium for identifying tangerine peel age using a multi-dimensional graph model, which can accurately predict the aging year of the tangerine peel to be identified.

[0005] The first object of the present invention is to provide a method for identifying the age of tangerine peel using a multidimensional graph model.

[0006] The second object of the present invention is to provide a multi-dimensional graph model tangerine peel year identification system.

[0007] The third object of the present invention is to provide a terminal device.

[0008] A fourth object of the present invention is to provide a computer-readable storage medium.

[0009] The first object of the present invention can be achieved by adopting the following technical solutions: A method for identifying tangerine peel years using a multidimensional graph model, the method comprising: Obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; take the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel as samples in the data set; The orthogonal partial least squares discriminant analysis model was trained using the data set, and the OPLS-DA standard model for tangerine peel year was obtained; The flavonoid components, amino acids, organic acids and volatile components of the tangerine peel to be identified and their contents were input into the OPLS-DA standard model of the tangerine peel year to obtain the year of the tangerine peel to be identified.

[0010] Furthermore, the samples in the data set were grouped according to the year of the dried tangerine peel; The sample X is used as the input layer data of the orthogonal partial least squares discriminant analysis model, and the score of the sample X in the model is used as the output layer data of the orthogonal partial least squares discriminant analysis model.

[0011] Furthermore, before the samples are input into the orthogonal partial least squares discriminant analysis model, the samples are standardized; the standardization is at least one of Z-score standardization, maximum and minimum standardization, logarithmic standardization, Sum standardization, centering and normalization.

[0012] Furthermore, the flavonoid composition and content of dried orange peel are obtained by the following process: The crushed dried tangerine peel powder was mixed with 70% ethanol solution in a ratio of 1:10-30, and ultrasonicated for 30-60 min at 50-70°C and 400-600W, and filtered to obtain a crude tangerine peel flavonoid extract; internal standard quercetin was added to the crude tangerine peel flavonoid extract to make its concentration 0.05 mg / mL to obtain a tangerine peel flavonoid test solution; the tangerine peel flavonoid test solution was detected by liquid chromatography-tandem mass spectrometry; based on the detection data, the molecular weight, retention time and characteristic ion fragment data of the compounds were processed using Compound Discoverer software, and the tangerine peel flavonoids were identified using the mzCloud network database and the local high-resolution traditional Chinese medicine component mass spectrum database; for the identified flavonoids, the characteristic flavonoid components were extracted using the traditional Chinese medicine fingerprint software (2010 version) and their contents were calculated.

[0013] Furthermore, the amino acid composition and content of dried tangerine peel are obtained by the following process: Mix the crushed tangerine peel sample powder with distilled water in a ratio of 1:5-20, extract at 70-90°C for 10-20 min, and filter to obtain an amino acid extract; use an amino acid automatic analyzer or high performance liquid chromatography to detect the amino acid extract; import the test data into the traditional Chinese medicine fingerprint software (2010 version), extract the characteristic amino acid components and calculate their contents.

[0014] Further, the composition and content of organic acid components of dried tangerine peel are obtained by the following process: Mix the crushed tangerine peel powder with distilled water in a ratio of 1:5-20, perform ultrasonic extraction at 70-95°C and 200-400W for 20-40min, and filter to obtain an organic acid extract. Detect the organic acid extract using high performance liquid chromatography. Import the detection data into the traditional Chinese medicine fingerprint software (2010 edition), extract the organic acid components and calculate their contents.

[0015] Further, the composition and content of volatile components of dried tangerine peel are obtained by the following process: 0.1 g of crushed dried tangerine peel powder and 20-50 μL of 0.1 g / kg cyclohexanone internal standard solution were mixed in a headspace vial, sealed and stored at 40-60°C for 5-10 min; a 50 μm divinylbenzene / carboxyl or / 30 μm polydimethylsiloxane solid phase microextraction head was inserted into the headspace vial, extracted at 40-60°C for 30-50 min, desorbed at 200-250°C for 6-10 min, and then injected into GC-MS for analysis; according to the analyzed data, the Qualitative Navigator software was used to process the accurate molecular weight, retention time and characteristic ion fragmentation data of the compounds; the volatile organic compounds were identified using the National Institute of Standards and Technology mass spectral library and the Wiley mass spectral database; for the identified volatile organic compounds, the characteristic volatile components were extracted and their contents were calculated using the Traditional Chinese Medicine Fingerprint Software (2010 Edition).

[0016] Furthermore, the component extraction method is: calculate the relative error of the peak area of ​​the same characteristic peak of multiple parallel samples, and when the relative error is less than 5%, the component corresponding to the characteristic peak is the extracted component.

[0017] The second object of the present invention can be achieved by adopting the following technical solutions: A multi-dimensional graph model tangerine peel year identification system, the system comprising: An acquisition module is used to obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel are used as samples in the data set; A training module is used to train the orthogonal partial least squares discriminant analysis model using the data set to obtain the OPLS-DA standard model of tangerine peel year; The identification module is used to input the flavonoid components, amino acids, organic acids and volatile component compositions and contents of the tangerine peel to be identified into the tangerine peel year OPLS-DA standard model to obtain the year of the tangerine peel to be identified.

[0018] The third object of the present invention can be achieved by adopting the following technical solutions: A terminal device includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the multi-dimensional graph model method for identifying the age of tangerine peel is implemented.

[0019] The fourth object of the present invention can be achieved by adopting the following technical solutions: A computer-readable storage medium stores a program, which, when executed by a processor, implements the multi-dimensional graph model method for identifying the age of tangerine peel.

[0020] The present invention has the following beneficial effects compared with the prior art: Compared with the existing tangerine peel year identification method, the present invention provides a supervised multivariate statistical regression modeling method, which comprehensively considers the characteristic component composition that causes the quality difference of tangerine peel in different years, constructs a fingerprint spectrum using tangerine peel flavonoids, amino acids, organic acids and volatile components data, and combines the orthogonal partial least squares discriminant analysis model to establish the tangerine peel year OPLS-DA standard model. The tangerine peel year OPLS-DA standard model has good separation, strong predictability, high accuracy, and can accurately predict the aging year of the tangerine peel to be identified. The method provided by the present invention is simple to operate, fast in analysis speed, and high in sensitivity, avoiding the problem of poor evaluation quality caused by the inaccuracy of existing evaluation indicators, and can provide a reference for the quality assessment of other types of Chinese herbal medicines and functional foods. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0022] Figure 1 It is a flow chart of the method for identifying the year of dried tangerine peel using the multidimensional atlas model according to Example 1 of the present invention; Figure 2 The fingerprints of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel aged for 1 to 15 years according to Example 1 of the present invention are as follows; Figure 3 The score graph of the OPLS-DA standard model of tangerine peel year constructed in Example 1 of the present invention; Figure 4 This is a graph showing the results of 200 random permutation tests of the OPLS-DA standard model for the year of dried tangerine peel constructed in Example 1 of the present invention; Figure 5 The year test result of the validation set tangerine peel in the tangerine peel year OPLS-DA standard model constructed in Example 1 of the present invention; Figure 6 It is a structural block diagram of the multi-dimensional graph model tangerine peel year identification system of embodiment 2 of the present invention; Figure 7 This is a structural block diagram of the terminal device of embodiment 3 of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in 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. It should be understood that the specific embodiments described are only used to explain the present application and are not used to limit the present application.

[0024] It should be noted that the reagents and methods involved in the present invention, unless otherwise specified, are commonly used reagents and methods in the art.

[0025] (1) Materials The dried orange peel can be the dried orange peel of any year. In the present embodiment, dried orange peels aged for 1 year, 4 years, 6 years, 8 years, 10 years and 15 years, a total of 6 groups of years, were collected from Jiangmen Laoshugan Tea Co., Ltd.

[0026] (2) Tangerine peel processing A total of 48 tangerine peel samples from different years (8 samples from each year) were taken, of which 30 were used as training set tangerine peel samples (5 samples from each year) and 18 were used as validation set tangerine peel samples (3 samples from each year), and they were crushed.

[0027] (3) Determination of flavonoid components in dried tangerine peel The 1-15-year-old dried tangerine peel sample powder was mixed with 70% (v / v) ethanol solution in a ratio of 1:20, and ultrasonicated at 60°C and 600W for 30 min, and filtered to obtain a crude tangerine peel flavonoid extract; internal standard quercetin was added to the crude tangerine peel flavonoid extract to a concentration of 0.05 mg / mL to obtain a tangerine peel flavonoid test solution; the tangerine peel flavonoid test solution was detected by liquid chromatography-tandem mass spectrometry to obtain relevant data, and the liquid chromatography-tandem mass spectrometry conditions were in accordance with the provisions of the Chinese Pharmacopoeia (2020 edition) and NY / T 3950-2021; According to the relevant data, Compound Discoverer software (version 3.0) was used to process the accurate molecular weight, retention time and characteristic ion fragment data of the compounds; the mzCloud network database and the local high-resolution mass spectrometry database of traditional Chinese medicine ingredients (mzVault) were used to identify flavonoids; for the identified flavonoids, the Traditional Chinese Medicine Fingerprint Software (2010 version) was used to extract the characteristic flavonoid components and calculate their contents. After standardization, the characteristic flavonoid component training data set A1 and validation data set A2 were obtained.

[0028] (4) Determination of amino acid composition in tangerine peel The powder of 1-15 year old tangerine peel samples was mixed with distilled water in a ratio of 1:10, extracted at 90℃ for 20 min, and filtered to obtain the amino acid extract. The amino acid composition and content were detected by amino acid automatic analyzer or high performance liquid chromatography, and the detection conditions were in accordance with the provisions of GB / T 30987-2020. The original data set of the detected amino acid composition was imported into the traditional Chinese medicine fingerprint software (2010 edition), the characteristic amino acid composition was extracted and its content was calculated, and after standardization, the characteristic amino acid composition data set B1 and the verification data set B2 were obtained.

[0029] (5) Determination of organic acid components in tangerine peel The powder of tangerine peel samples aged 1 to 15 years was mixed with distilled water in a ratio of 1:10, and ultrasonically extracted at 95℃ and 320W for 40 min. The organic acid extract was filtered to obtain the organic acid extract. The content of organic acid components was detected by high performance liquid chromatography, and the detection conditions were in accordance with the provisions of GB5009.157-2016 or GB / T 40179-2021. The original data set of organic acid components detected was imported into the traditional Chinese medicine fingerprint software (2010 version), the characteristic organic acid components were extracted and their contents were calculated, and after standardization, the characteristic organic acid component data set C1 and the verification data set C2 were obtained.

[0030] (6) Determination of volatile components in dried tangerine peel 0.1 g of 1-15 year old dried tangerine peel powder and 20 μL (0.1 g / kg) cyclohexanone internal standard solution were mixed in a headspace vial, sealed and stored at 60°C for 5 min; a 50 μm divinylbenzene / carboxyl / 30 μm polydimethylsiloxane solid phase microextraction head was inserted into the headspace vial, extracted at 60°C for 30 min, desorbed at 250°C for 6 min, and then injected into GC-MS for analysis; GC-MS conditions were in accordance with the provisions of T / GZTPA 0009-2021; Qualitative Navigator software (version B.08.00) was used to process the accurate molecular weight, retention time and characteristic ion fragmentation data of the compounds; the National Institute of Standards and Technology (NIST) mass spectral library and Wiley mass spectral database were used to identify volatile organic compounds; the traditional Chinese medicine fingerprint software (2010 version) was used to extract characteristic volatile components and calculate their contents, and the characteristic volatile component data set D1 and validation data set D2 were obtained after standardization.

[0031] (7) Calculation of year forecast error and year forecast accuracy Year prediction error E=|Y 预测 -Y 实际 ∣,Y 预测 The predicted year for tangerine peel, Y 实际 is the actual year of the dried tangerine peel; Year prediction accuracy T = the number of tangerine peel samples in the validation set with zero year prediction error / the total number of tangerine peel samples in the validation set * 100%.

[0032] Embodiment 1: like Figure 1 As shown, this embodiment provides a method for identifying the year of tangerine peel using a multidimensional graph model, which specifically includes the following steps: S101. Obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peels of different known years and use them as samples in the data set.

[0033] The flavonoid components, amino acids, organic acids and volatile components of dried tangerine peels of different known years are obtained through the above steps (1) to (6).

[0034] Combine data sets A1, B1, C1 and D1 to obtain training data set 1.

[0035] S102, using the data set to train the OPLS-DA model to obtain the tangerine peel year OPLS-DA standard model.

[0036] The orthogonal partial least squares discriminant analysis (OPLS-DA) model is based on the characteristic metabolite expression data (X variable) and sample classification information (Y variable). It can remove data changes unrelated to the X and Y variables through orthogonal signal correction, and filter out noise unrelated to classification information, effectively obtain difference information between sample groups, and improve the classification and discrimination ability of the model.

[0037] The training data set 1 was then imported into the SIMCA-P software, and standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the years of tangerine peel. The scores of the corresponding training data in each year group in the model were used as the output layer for model training, and the standard model 1 of tangerine peel year orthogonal partial least squares discriminant analysis (OPLS-DA) was obtained.

[0038] Model 1 was subjected to 200 random permutation tests to determine whether it had overfitting. Data sets A2, B2, C2, and D2 were combined to obtain validation data set 1, which was imported into standard model 1 for fitting. The year prediction error and year prediction accuracy were calculated based on the model output year and the actual year of tangerine peel in the validation set.

[0039] S103, inputting the flavonoid components, amino acids, organic acids and volatile component compositions and contents of the tangerine peel to be identified into the tangerine peel year OPLS-DA standard model to obtain the year of the tangerine peel to be identified.

[0040] The flavonoid components, amino acids, organic acids and volatile component compositions and contents of the tangerine peel to be identified are input into model 1 to obtain the aging year of the tangerine peel to be identified.

[0041] Comparative Example 1: Dataset A1 was used as training data set 2 and then imported into SIMCA-P software. Standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data set corresponding to each year group in the model were used as the output layer for model training to obtain the tangerine peel year OPLS-DA standard model 2. 200 random permutation tests were performed on model 2 to determine whether it had overfitting. Dataset A2 was used as validation data set 2 and validation data set 2 was imported into standard model 2 for fitting. According to the model output year and the actual year of tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0042] Comparative Example 2: Dataset B1 was used as training data set 3 and then imported into SIMCA-P software. Standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data set corresponding to each year group in the model were used as the output layer for model training to obtain the tangerine peel year OPLS-DA standard model 3. 200 random permutation tests were performed on model 3 to determine whether it had overfitting. Dataset B2 was used as validation data set 3 and validation data set 3 was imported into standard model 3 for fitting. According to the model output year and the actual year of tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0043] Comparative Example 3: Datasets A1 and D1 were used as training data set 4, which were then imported into SIMCA-P software. Standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data sets corresponding to each year group in the model were used as the output layer for model training to obtain the tangerine peel year OPLS-DA standard model 4. 200 random permutation tests were performed on model 4 to determine whether it had overfitting. Datasets A2 and D2 were used as validation data set 4, which were imported into standard model 4 for fitting. According to the model output year and the actual year of tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0044] Comparative Example 4: Datasets A1, B1 and C1 were used as training data set 5, and then imported into SIMCA-P software. Standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data sets corresponding to each year group in the model were used as the output layer for model training to obtain the tangerine peel year OPLS-DA standard model 5. 200 random permutation tests were performed on model 5 to determine whether it had overfitting. Datasets A2, B2 and C2 were used as validation data set 5, and validation data set 5 was imported into standard model 5 for fitting. According to the model output year and the actual year of tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0045] Comparative Example 5: The 1- to 15-year-old dried tangerine peel sample powder was taken and its color parameters L*, a*, b* and △E* were measured using a colorimeter. After standardization, the dried tangerine peel color parameter data set E1 and the validation data set E2 were obtained. The alkaloid content was determined by the method specified in 38571-2020, and the tangerine peel alkaloid data set F1 and validation data set F2 were obtained after standardization; the data sets A1, B1, E1 and F1 were used as training data set 6, and then imported into SIMCA-P software, and standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data sets corresponding to each year group in the model were used as the output layer to train the model and obtain the tangerine peel year OPLS-DA standard model 6; 200 random permutation tests were performed on model 6 to determine whether it had overfitting; the data sets A2, B2, E2 and F2 were used as validation data set 6, and the validation data set 6 was imported into the standard model 6 for fitting, and the year prediction error and year prediction accuracy were calculated according to the model output year and the actual year of the tangerine peel in the validation set.

[0046] Comparative Example 6: The powder of 1-15-year-old dried tangerine peel samples was extracted and the free fatty acid content was determined according to the method specified in GB 5009.168-2016. After standardization, the dried tangerine peel fatty acid dataset G1 and validation dataset G2 were obtained. The mineral element contents of dried tangerine peel were extracted and determined by the method specified in 5009.94-2012, and the mineral element data set H1 and validation data set H2 were obtained after standardization. Data sets E1, F1, G1 and H1 were used as training data set 7, and then imported into SIMCA-P software, and standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of dried tangerine peel. The scores of the training data sets in each year grouping in the model were used as the output layer to train the model, and the OPLS-DA standard model 7 for the year of dried tangerine peel was obtained. 200 random permutation tests were performed on model 7 to determine whether it had overfitting. Data sets E2, F2, G2 and H2 were used as validation data set 7, and validation data set 7 was imported into standard model 7 for fitting. According to the model output year and the actual year of the dried tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0047] Comparative Example 7: The data sets A1, B1, C1, D1, E1 and F1 were used as training data sets 8 and then imported into SIMCA-P software. The standard year groups 1Y, 4Y, 6Y, 8Y, 10Y and 15Y were set according to the year of tangerine peel. The scores of the training data sets corresponding to each year group in the model were used as the output layer for model training to obtain the tangerine peel year OPLS-DA standard model 8. 200 random permutation tests were performed on model 8 to determine whether it had overfitting. The data sets A2, B2, C2, D2, E2 and F2 were used as validation data sets 8, and the validation data set 8 was imported into the standard model 8 for fitting. According to the model output year and the actual year of the tangerine peel in the validation set, the year prediction error and year prediction accuracy were calculated.

[0048] A total of 100 characteristic components with good precision, stability and repeatability were extracted from the active components of tangerine peel of 6 groups of years, including 30 characteristic flavonoid components, 11 characteristic amino acid components, 8 characteristic organic acid components and 51 characteristic volatile components (as shown in Tables 1-4). The model construction test of these components found that reducing the number of these characteristic components would significantly reduce the accuracy and identification effect of the model. Therefore, these 100 components were finally selected as the characteristic components of tangerine peel of different years to construct the identification model of tangerine peel year.

[0049] Table 1 Flavonoid components Table 2 Amino acid composition Table 3 Organic acid components Table 4 Volatile components result: Figure 2 The differences in flavonoids, amino acids, organic acids and volatile components of tangerine peel aged from 1 to 15 years were shown. The results showed that a total of 30 flavonoids, 11 amino acids, 8 organic acids and 51 volatile components were detected in tangerine peel of the six groups of years. With the extension of aging time, the content of flavonoid glycosides in tangerine peel flavonoids decreased, while the content of flavanone aglycones and polymethoxyflavonoids increased; the content of amino acids and organic acids gradually decreased or even disappeared; the content of terpenes in volatile components increased, while the content of alcohols, aldehydes, esters, phenols and ketones decreased. These results show that the flavonoids, amino acids, organic acids and volatile components of tangerine peel aged from 1 to 15 years are very different, and the above differences can be used to construct a tangerine peel year identification model.

[0050] Figure 3 and Figure 4The OPLS-DA standard model score graph of tangerine peel years constructed in Example 1 and the result graph of 200 random permutation tests are shown respectively. Figure 2 It can be seen that the model successfully separated the 6 groups of dried tangerine peels from different years, and the separation degree of each group of dried tangerine peels was good, and there was no overlap. Figure 3 It can be seen that the R obtained by 200 random permutation tests is 2 and Q 2 (indicating the predictive ability of the model) are all lower than the R of the standard model in the upper right corner 2 and Q 2 The intercepts of the fitting lines were 0.182 and -0.559, respectively, indicating that the model does not have overfitting phenomenon and can be used to identify the year of tangerine peel.

[0051] Figure 5 The year prediction results of the validation set tangerine peel in the tangerine peel year OPLS-DA standard model constructed in Example 1 are shown. As can be seen from the figure, the OPLS-DA standard model constructed in Example 1 can correctly classify the 18 validation set tangerine peel samples into the year categories to which they actually belong, and the year identification accuracy rate reaches 100%, which shows that the OPLS-DA standard model constructed based on the four characteristic components has good classification performance, predictability and accuracy for tangerine peel year identification.

[0052] Table 5 shows the quality of the tangerine peel year identification model constructed in Example 1 and Comparative Examples 1 to 7 and the prediction effect of the year of the tangerine peel to be identified. 2 X, R 2 Y and Q 2 The values ​​are 0.943, 0.99 and 0.985 respectively, indicating that the model explains 94.3% of the changes in the X variable and can fit 99% of the Y variable. The predictability of the model is as high as 98.5%, which has good predictability and reliability. In addition, the year prediction accuracy shows that the tangerine peel year identification model constructed in Example 1 can accurately predict the tangerine peel year to be identified, and the year prediction accuracy T is as high as 100%. In contrast, the R values ​​of the models established in Comparative Examples 1 to 4 are as high as 100%. 2 X, R 2 Y and Q 2 The values ​​are only 0.84~0.911, and T is only 72.22%~83.33%, indicating that reducing the types and quantities of characteristic difference components will reduce the quality and accuracy of the constructed model; the R values ​​of the models established in Examples 5 and 6 are 2 X, R 2 Y and Q 2The value is only 0.905~0.954, and T is only 83.33%~88.89%, indicating that changing the types of characteristic difference components of tangerine peel from different years used to construct the model will also reduce the predictability of the model. Only when the quality of the model constructed by using the four types of characteristic difference components of tangerine peel proposed by the present invention can reach the best; in addition, the parameters of the model established in Comparative Example 7 are similar to those in Example 1, indicating that the four types of characteristic difference components selected can fully reflect the differences between tangerine peels from different years. On this basis, increasing the types and quantities of characteristic difference components will only further increase the workload, but will not improve the quality of the model again.

[0053] Table 5 Results obtained from Example 1 and Comparative Examples 1 to 7 Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be completed by instructing related hardware through a program, and the corresponding program may be stored in a computer-readable storage medium.

[0054] It should be noted that although the method operations of the above embodiments are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0055] Embodiment 2: like Figure 6 As shown, this embodiment provides a multi-dimensional graph model tangerine peel year identification system, which includes an acquisition module 601, a training module 602 and an identification module 603, wherein: The acquisition module 601 is used to obtain the flavonoid components, amino acids, organic acids and volatile components of tangerine peel of different known years and their contents; the flavonoid components, amino acids, organic acids and volatile components of tangerine peel and their contents are used as samples in the data set; A training module 602 is used to train the OPLS-DA model using the data set to obtain a standard OPLS-DA model for tangerine peel years; The identification module 603 is used to input the flavonoid components, amino acids, organic acids and volatile components of the tangerine peel to be identified into the tangerine peel year OPLS-DA standard model to obtain the year of the tangerine peel to be identified.

[0056] The specific implementation of each module in this embodiment can refer to the above-mentioned embodiment 1, which will not be described one by one here; it should be noted that the system provided in this embodiment is only illustrated by the division of the above-mentioned functional modules. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure can be divided into different functional modules to complete all or part of the functions described above.

[0057] Embodiment 3: This embodiment provides a terminal device, which can be a computer. Figure 7 As shown, it comprises a processor 702, a memory, an input device 703, a display 704 and a network interface 705 connected via a system bus 701. The processor is used to provide computing and control capabilities. The memory comprises a non-volatile storage medium 706 and an internal memory 707. The non-volatile storage medium 706 stores an operating system, a computer program and a database. The internal memory 707 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the processor 702 executes the computer program stored in the memory, the multi-dimensional atlas model tangerine peel year identification method of the above-mentioned embodiment 1 is implemented as follows: Obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; take the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel as samples in the data set; The OPLS-DA model was trained using the data set to obtain the OPLS-DA standard model for tangerine peel years; The flavonoid components, amino acids, organic acids and volatile components of the tangerine peel to be identified and their contents were input into the OPLS-DA standard model of the tangerine peel year to obtain the year of the tangerine peel to be identified.

[0058] Embodiment 4: This embodiment provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the multi-dimensional graph model method for identifying the age of tangerine peel in the above-mentioned embodiment 1 is implemented as follows: Obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; take the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel as samples in the data set; The OPLS-DA model was trained using the data set to obtain the OPLS-DA standard model for tangerine peel years; The flavonoid components, amino acids, organic acids and volatile components of the tangerine peel to be identified and their contents were input into the OPLS-DA standard model of the tangerine peel year to obtain the year of the tangerine peel to be identified.

[0059] It should be noted that the computer-readable storage medium of the present embodiment may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0060] In summary, the present invention utilizes the data of flavonoid components, amino acids, organic acids and volatile components of tangerine peel from different years to construct its fingerprint, and combines OPLS-DA to establish a method for multidimensional atlas tangerine peel year identification model. This method not only comprehensively analyzes the characteristic components of tangerine peel with high sensitivity, but also has strong predictability and high precision in the constructed model, and can accurately predict the aging year of the tangerine peel to be identified, making up for the deficiency of poor identification ability of the model constructed by a single indicator.

[0061] The above is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solution and inventive concept of the present invention within the scope disclosed by the present invention, which shall fall within the protection scope of the present invention.

Claims

1. A method for identifying the year of dried tangerine peel using a multidimensional atlas model, characterized in that: The method comprises: Obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; take the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel as samples in the data set; The orthogonal partial least squares discriminant analysis model was trained using the data set, and the OPLS-DA standard model for tangerine peel year was obtained; The flavonoid components, amino acids, organic acids and volatile components of the tangerine peel to be identified and their contents were input into the OPLS-DA standard model of the tangerine peel year to obtain the year of the tangerine peel to be identified.

2. The multidimensional atlas model tangerine peel year identification method according to claim 1, characterized in that: Group the samples in the dataset according to the year of the dried tangerine peel; The sample X is used as the input layer data of the orthogonal partial least squares discriminant analysis model, and the score of the sample X in the model is used as the output layer data of the orthogonal partial least squares discriminant analysis model.

3. The multidimensional atlas model method for identifying tangerine peel years according to any one of claims 1 to 2, characterized in that: Before inputting the samples into the orthogonal partial least squares discriminant analysis model, the samples are standardized; the standardization is at least one of Z-score standardization, maximum and minimum standardization, logarithmic standardization, Sum standardization, centering and normalization.

4. The multidimensional atlas model method for identifying tangerine peel years according to claim 1, characterized in that: The flavonoid composition and content of dried orange peel are obtained through the following process: The crushed dried tangerine peel powder was mixed with 70% ethanol solution in a ratio of 1:10-30, and ultrasonicated for 30-60 min at 50-70°C and 400-600W, and filtered to obtain a crude tangerine peel flavonoid extract; internal standard quercetin was added to the crude tangerine peel flavonoid extract to make its concentration 0.05 mg / mL to obtain a tangerine peel flavonoid test solution; the tangerine peel flavonoid test solution was detected by liquid chromatography-tandem mass spectrometry; based on the detection data, the molecular weight, retention time and characteristic ion fragment data of the compounds were processed using Compound Discoverer software, and the tangerine peel flavonoids were identified using the mzCloud network database and the local high-resolution traditional Chinese medicine component mass spectrum database; for the identified flavonoids, the characteristic flavonoid components were extracted using the traditional Chinese medicine fingerprint software (2010 version) and their contents were calculated.

5. The multidimensional graph model tangerine peel year identification method according to claim 1, characterized in that: The amino acid composition and content of dried tangerine peel are obtained through the following process: Mix the crushed tangerine peel sample powder with distilled water in a ratio of 1:5-20, extract at 70-90°C for 10-20 min, and filter to obtain an amino acid extract; use an amino acid automatic analyzer or high performance liquid chromatography to detect the amino acid extract; import the test data into the traditional Chinese medicine fingerprint software (2010 version), extract the characteristic amino acid components and calculate their contents.

6. The multidimensional atlas model method for identifying tangerine peel years according to claim 1, characterized in that: The composition and content of organic acids in dried orange peel are obtained through the following process: Mix the crushed tangerine peel powder with distilled water in a ratio of 1:5-20, perform ultrasonic extraction at 70-95°C and 200-400W for 20-40min, and filter to obtain an organic acid extract. Detect the organic acid extract using high performance liquid chromatography. Import the detection data into the traditional Chinese medicine fingerprint software (2010 edition), extract the organic acid components and calculate their contents.

7. The multi-dimensional graph model tangerine peel year identification method according to claim 1, characterized in that: The composition and content of volatile components of dried tangerine peel were obtained by the following process: 0.1 g of crushed tangerine peel powder and 20-50 μL of 0.1 g / kg cyclohexanone internal standard solution were mixed in a headspace vial, sealed and stored at 40-60°C for 5-10 min; a 50 μm divinylbenzene / carboxyl / 30 μm polydimethylsiloxane solid phase microextraction head was inserted into the headspace vial, extracted at 40-60°C for 30-50 min, desorbed at 200-250°C for 6-10 min, and then injected into GC-MS for analysis; Based on the analyzed data, Qualitative Navigator software was used to process the accurate molecular weight, retention time and characteristic ion fragmentation data of the compounds; The volatile organic compounds were identified using the National Institute of Standards and Technology mass spectral library and the Wiley mass spectral database. The characteristic volatile components of the identified volatile organic compounds were extracted and their contents were calculated using the Traditional Chinese Medicine Fingerprint Software (2010 Edition).

8. The multidimensional atlas model method for identifying tangerine peel years according to any one of claims 4 to 7, characterized in that: The component extraction method is: calculate the relative error of the peak area of ​​the same characteristic peak of multiple parallel samples. When the relative error is less than 5%, the component corresponding to the characteristic peak is the extracted component.

9. A multi-dimensional graph model tangerine peel year identification system, characterized in that: The system comprises: An acquisition module is used to obtain the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel of different known years; the composition and content of flavonoids, amino acids, organic acids and volatile components of dried tangerine peel are used as samples in the data set; A training module is used to train the orthogonal partial least squares discriminant analysis model using the data set to obtain the OPLS-DA standard model of tangerine peel year; The identification module is used to input the flavonoid components, amino acids, organic acids and volatile component compositions and contents of the tangerine peel to be identified into the tangerine peel year OPLS-DA standard model to obtain the year of the tangerine peel to be identified.

10. A computer-readable storage medium storing a program, wherein when the program is executed by a processor, the method for identifying the age of tangerine peel using the multi-dimensional graph model according to any one of claims 1 to 8 is implemented.

Images