Method and system for identifying adulterated tea oil based on machine learning algorithm

Abstract

The present invention discloses a method and system for identifying adulterated tea oil based on a machine learning algorithm, and relates to the technical field of chemical analysis. The method comprises the following steps: using a colorimetric sensor array method to detect a tea oil sample to be tested, analyzing the pre-treated tea oil adulteration related data and the pre-treated tea oil color evaluation related data affected by chemical substances, obtaining a tea oil adulteration evaluation coefficient and a tea oil color evaluation coefficient affected by chemical substances, and comparing the tea oil grade evaluation coefficient with the tea oil grade evaluation threshold to obtain a tea oil grade evaluation threshold comparison result. The present invention compares the tea oil grade evaluation coefficient with the tea oil grade evaluation threshold, thereby performing a quality classification on the source of the tea oil sample to be tested according to the comparison result of the tea oil grade evaluation threshold, thereby achieving the effect of quality rating of the tested tea oil, and solving the problem that the quality rating of the tested tea oil cannot be performed in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical analysis, and in particular to a method and system for identifying adulterated tea oil based on a machine learning algorithm. Background Art

[0002] Traditional tea oil adulteration detection methods include sensory evaluation, physicochemical analysis, and spectroscopy. Sensory evaluation relies on the evaluator's perception of the color, aroma, and taste of the oil, but it is ineffective when the adulterant has similar characteristics to tea oil. In addition, there are physicochemical analysis methods, such as fatty acid content composition analysis, which can accurately identify but are complex and time-consuming.

[0003] For example, the invention patent announcement number: CN113884594B discloses a method for identifying adulterated tea oil based on a machine learning algorithm, including: establishing a multi-classification decision tree model that can directly give a classification conclusion of the adulterated types of camellia oil samples. Based on the fatty acid content and triglyceride experimental data of adulterated camellia oil, the binary and multi-classification decision tree models are applied to the identification of the types of adulterated camellia oil under high and low adulteration gradients.

[0004] For example, the invention patent announcement number CN113406249B discloses a method for predicting the type of adulterated oil in tea oil, including: adding corn oil, rapeseed oil, rice bran oil, sesame oil, and soybean oil in different proportions to tea oil, and performing gas chromatography (GC) and high performance liquid chromatography (HPLC) detection to obtain the corresponding fatty acid content and tocopherol profile. Tea oil, corn oil, rapeseed oil, rice bran oil, sesame oil, and soybean oil can be better distinguished by combining HCA with the fatty acid content ratio and tocopherol composition; and then the adulterated tea oil is discriminated by the PLS-DA discriminant model and / or the SIMCA discriminant model.

[0005] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0006] In the prior art, traditional adulteration detection methods are too simple and unable to distinguish between adulteration of high-quality tea oil and low-quality tea oil and adulteration of tea oil with other oils. There is also a problem of being unable to grade the quality of the tested tea oil. Summary of the invention

[0007] The embodiments of the present application provide a method and system for identifying adulterated tea oil based on a machine learning algorithm, thereby solving the problem in the prior art of being unable to distinguish between adulterated high-quality tea oil and low-quality tea oil, and adulteration of tea oil with other oils, and achieving the effect of quality grading of the tested tea oil.

[0008] The embodiment of the present application provides a method for identifying adulterated tea oil based on a machine learning algorithm, comprising the following steps: collecting tea oil samples to be tested, pre-treating the tea oil samples to be tested, detecting the tea oil samples to be tested using a colorimetric sensor array method, and obtaining tea oil adulteration-related data and tea oil color evaluation-related data due to the influence of chemical substances; pre-treating the tea oil adulteration-related data and the tea oil color evaluation-related data due to the influence of chemical substances, obtaining pre-treated tea oil adulteration-related data and pre-treated tea oil color evaluation-related data due to the influence of chemical substances, analyzing the pre-treated tea oil adulteration-related data and the pre-treated tea oil color evaluation-related data due to the influence of chemical substances, and obtaining tea oil adulteration evaluation coefficients and tea oil color evaluation coefficients due to the influence of chemical substances; and calculating the tea oil adulteration-related data and the tea oil color evaluation coefficients according to the influence of chemical substances from the data. The tea oil adulteration assessment threshold is obtained from the database, the tea oil adulteration assessment coefficient is compared with the tea oil adulteration assessment threshold, and the comparison result of the tea oil adulteration assessment threshold is obtained. The types of tea oil adulteration are divided according to the comparison result of the tea oil adulteration assessment threshold, and the type division result of tea oil adulteration is obtained; the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances are comprehensively evaluated to obtain the tea oil grade assessment coefficient, the tea oil grade assessment threshold is obtained from the database, the tea oil grade assessment coefficient is compared with the tea oil grade assessment threshold, and the comparison result of the tea oil grade assessment threshold is obtained; the quality of the source of the tea oil sample to be tested is divided according to the comparison result of the tea oil grade assessment threshold, and the quality division result of the source of the tea oil to be tested is obtained; the specific method for obtaining the tea oil grade assessment coefficient is as follows: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

[0009] Furthermore, the specific pretreatment process for the tea oil sample to be tested is: the integrated control and decision-making system selects color-sensitive materials that are sensitive to the volatile gas of the oil sample, and the integrated control and decision-making system uses a robotic arm to dissolve the color-sensitive materials in dichloromethane or anhydrous ethanol to obtain a color-sensitive material solution, and titrates the color-sensitive material onto a polyvinylidene fluoride membrane to form a sensor array, and the sensor array is ventilated and stored in a sealed manner.

[0010] Furthermore, the specific detection process of using the colorimetric sensor array method to detect the tea oil sample to be tested is: the integrated control and decision-making system stirs the mixed oil sample evenly through a robotic arm, and places the oil sample in the oil sample chamber, starts the cooling fan, activates the heating platform and the LED ring light source, places the sensor array at the center of the sensor array bracket, and directly contacts the reaction chamber with the heating platform. The image acquisition device collects images of the sensor array before and after the reaction, and directly obtains the saturated fatty acid content of the sample and the monounsaturated fatty acid content of the sample from the image.

[0011] Furthermore, the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances specifically include: the relative molecular mass of the sample, the density of the sample, the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample obtained by mass spectrometry, experiments and colorimetry; the data related to tea oil adulteration include the saturated fatty acid content of the sample, the monounsaturated fatty acid content of the sample, the relative molecular mass of the sample and the density of the sample; the data related to the evaluation of tea oil color due to the influence of chemical substances include the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample.

[0012] Furthermore, the specific preprocessing process for preprocessing the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances is as follows: preprocessing includes data cleaning, data conversion and feature extraction; data cleaning includes filling missing values, deleting outliers and duplicate values; data conversion: normalizing the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances; feature extraction: using the principal component analysis method to perform feature processing on the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances.

[0013] Furthermore, the specific analysis method for analyzing the pretreated tea oil adulteration related data and the pretreated tea oil color evaluation related data due to the influence of chemical substances is as follows: the saturated fatty acid content of the sample, the monounsaturated fatty acid content of the sample, the relative molecular mass of the sample and the density of the sample in the tea oil adulteration related data are analyzed to obtain the tea oil adulteration evaluation coefficient; the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample in the tea oil color evaluation related data due to the influence of chemical substances are analyzed to obtain the tea oil color evaluation coefficient due to the influence of chemical substances.

[0014] Furthermore, the specific comparison process of comparing the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold is as follows: obtain the first tea oil adulteration assessment threshold and the second tea oil adulteration assessment threshold from the database, the first tea oil adulteration assessment threshold is greater than the second tea oil adulteration assessment threshold, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, if the tea oil adulteration assessment coefficient is greater than or equal to the first tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the third level of adulteration, if the tea oil adulteration assessment coefficient is greater than or equal to the second tea oil adulteration assessment threshold and the tea oil adulteration assessment coefficient is less than the first tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the second level of adulteration, if the tea oil adulteration assessment coefficient is less than the second tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the first level of adulteration, if the tea oil source of the tea oil sample is the first level of adulteration, conduct a comprehensive evaluation of the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient.

[0015] Furthermore, the specific comparison process of comparing the tea oil grade assessment coefficient with the tea oil grade assessment threshold is: obtaining the first tea oil grade assessment threshold and the second tea oil grade assessment threshold from the database, the first tea oil grade assessment threshold is greater than the second tea oil grade assessment threshold, if the tea oil grade assessment coefficient is greater than or equal to the first tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a superior grade tea oil, if the tea oil grade assessment coefficient is greater than or equal to the second tea oil grade assessment threshold and the tea oil grade assessment coefficient is less than the first tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a first-grade tea oil, if the tea oil grade assessment coefficient is less than the second tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a second-grade tea oil.

[0016] The embodiment of the present application provides a system for identifying adulterated tea oil based on a machine learning algorithm, including a sample detection module, a data analysis module, and a threshold comparison module:

[0017] Sample detection module: used to collect tea oil samples to be tested, pre-treat the tea oil samples to be tested, and use the colorimetric sensor array method to detect the tea oil samples to be tested, so as to obtain data related to tea oil adulteration and tea oil color evaluation due to the influence of chemical substances;

[0018] Data analysis module: used for preprocessing tea oil adulteration related data and tea oil color evaluation related data due to chemical substances, obtaining preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, analyzing the preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, obtaining tea oil adulteration evaluation coefficient and tea oil color evaluation coefficient due to chemical substances;

[0019] Threshold comparison module: used to obtain the tea oil adulteration assessment threshold from the database, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, obtain the tea oil adulteration assessment threshold comparison result, classify the types of tea oil adulteration according to the tea oil adulteration assessment threshold comparison result, and obtain the tea oil adulteration type classification result;

[0020] Comprehensively evaluate the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, obtain the tea oil grade assessment threshold from the database, compare the tea oil grade assessment coefficient with the tea oil grade assessment threshold to obtain the tea oil grade assessment threshold comparison result, and classify the quality of the source of the tea oil sample to be tested according to the tea oil grade assessment threshold comparison result to obtain the quality classification result of the source of the tea oil to be tested;

[0021] The specific method for obtaining the tea oil grade assessment coefficient is as follows: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

[0022] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0023] 1. By comparing the tea oil grade assessment coefficient with the tea oil grade assessment threshold, the tea oil grade assessment threshold comparison result is obtained, and then the quality of the source of the tea oil sample to be tested is divided according to the tea oil grade assessment threshold comparison result, thereby achieving the effect of quality rating of the tested tea oil, and effectively solving the problem in the prior art that the tested tea oil cannot be rated for quality.

[0024] 2. By comparing the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, the tea oil adulteration assessment threshold comparison result is obtained, and then the types of tea oil adulteration are divided according to the tea oil adulteration assessment threshold comparison result, thereby achieving the effect of dividing the types of tea oil adulteration, and effectively solving the problem that the types of tea oil adulteration cannot be divided in the prior art.

[0025] 3. By collecting tea oil samples to be tested and pre-treating them, the colorimetric sensor array method is used to detect the tea oil samples to obtain data related to tea oil adulteration and data related to the evaluation of tea oil color affected by chemical substances, thereby achieving the accuracy of adulterated tea oil detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flow chart of a method for identifying adulterated tea oil based on a machine learning algorithm provided in an embodiment of the present application;

[0027] Figure 2 It is a structural schematic diagram of a tea oil adulteration detection device;

[0028] Figure 3 This is a schematic diagram of the structure of the box in the tea oil adulteration detection device;

[0029] Figure 4 It is a structural schematic diagram of a heat dissipation device in a tea oil adulteration detection device;

[0030] Figure 5 It is a schematic diagram of the structure of the integrated control and decision-making system in the tea oil adulteration detection device;

[0031] Figure 6 It is a structural schematic diagram of an image acquisition device in a tea oil adulteration detection device;

[0032] Figure 7 It is a schematic diagram of the structure of the reaction chamber in the tea oil adulteration detection device;

[0033] Figure 8 An image of the tea oil color evaluation coefficient affected by chemical substances in the method for identifying adulterated tea oil based on a machine learning algorithm provided in an embodiment of the present application;

[0034] Fig. 9 This is a schematic diagram of the structure of the adulterated tea oil identification system based on the machine learning algorithm provided in the embodiment of the present application. DETAILED DESCRIPTION

[0035] The embodiments of the present application solve the problem in the prior art that it is impossible to grade the quality of the tested tea oil by providing a method and system for identifying adulterated tea oil based on a machine learning algorithm. The tea oil grade assessment coefficient is compared with the tea oil grade assessment threshold to obtain a tea oil grade assessment threshold comparison result, and the source of the tested tea oil sample is divided into quality categories according to the tea oil grade assessment threshold comparison result, thereby achieving the effect of quality grading of the tested tea oil.

[0036] The technical solution in the embodiment of the present application is to solve the above-mentioned problem that the quality of the tea oil tested cannot be graded. The overall idea is as follows:

[0037] A colorimetric sensor array method is used to detect the tea oil samples to be tested, and data related to tea oil adulteration and data related to tea oil color evaluation due to the influence of chemical substances are obtained. The pretreated tea oil adulteration data and the pretreated tea oil color evaluation data due to the influence of chemical substances are analyzed to obtain a tea oil adulteration assessment coefficient and a tea oil color evaluation coefficient due to the influence of chemical substances. The tea oil adulteration assessment threshold is obtained from a database, and the tea oil adulteration assessment coefficient is compared with the tea oil adulteration assessment threshold. The tea oil grade assessment threshold is obtained from the database, and the tea oil grade assessment coefficient is compared with the tea oil grade assessment threshold. According to the comparison results of the tea oil grade assessment threshold, the source of the tea oil sample to be tested is divided into quality categories to obtain the quality classification results of the source of the tea oil to be tested.

[0038] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0039] like Figure 1 As shown, it is a flow chart of the method for identifying adulterated tea oil based on machine learning algorithm provided in an embodiment of the present application, and the method comprises the following steps: collecting tea oil samples to be tested, pre-treating the tea oil samples to be tested, detecting the tea oil samples to be tested using a colorimetric sensor array method, and obtaining tea oil adulteration related data and tea oil color evaluation related data due to the influence of chemical substances; pre-treating the tea oil adulteration related data and the tea oil color evaluation related data due to the influence of chemical substances, and obtaining the pre-treated tea oil adulteration related data and the pre-treated tea oil color evaluation related data due to the influence of chemical substances, and analyzing the pre-treated tea oil adulteration related data and the pre-treated tea oil color evaluation related data due to the influence of chemical substances, and obtaining the tea oil adulteration evaluation coefficient and the tea oil color evaluation coefficient due to the influence of chemical substances. coefficient; obtain the tea oil adulteration assessment threshold from the database, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, obtain the comparison result of the tea oil adulteration assessment threshold, classify the types of tea oil adulteration according to the comparison result of the tea oil adulteration assessment threshold, and obtain the classification result of the types of tea oil adulteration; comprehensively evaluate the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, obtain the tea oil grade assessment threshold from the database, compare the tea oil grade assessment coefficient with the tea oil grade assessment threshold, and obtain the comparison result of the tea oil grade assessment threshold; classify the quality of the source of the tea oil sample to be tested according to the comparison result of the tea oil grade assessment threshold, and obtain the quality classification result of the source of the tea oil to be tested; the specific method for obtaining the tea oil grade assessment coefficient is as follows: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

[0040] In this embodiment, the weight factor of the tea oil adulteration assessment coefficient and the weight factor of the tea oil color assessment coefficient due to the influence of chemical substances can be obtained through the dynamic information database, which represents the proportion of the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient due to the influence of chemical substances in the tea oil grade assessment coefficient. The weight factor of the tea oil adulteration assessment coefficient and the weight factor of the tea oil color assessment coefficient due to the influence of chemical substances can also be obtained through a mapping relationship. For example, a mapping set of the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient due to the influence of chemical substances and their corresponding weights is established through the relationship between the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient due to the influence of chemical substances in historical data and dynamic information. The corresponding weight factors of the tea oil adulteration assessment coefficient and the weight factors of the tea oil color assessment coefficient due to the influence of chemical substances in the mapping set are obtained by inputting the real-time tea oil adulteration assessment coefficient and the tea oil color assessment coefficient due to the influence of chemical substances.

[0041] Furthermore, the specific pretreatment process for the tea oil sample to be tested is: the integrated control and decision-making system selects color-sensitive materials that are sensitive to volatile gases in the oil sample, and the integrated control and decision-making system uses a robotic arm to dissolve the color-sensitive materials in dichloromethane or anhydrous ethanol to obtain a color-sensitive material solution, and titrates the color-sensitive material onto a polyvinylidene fluoride membrane to form a sensor array, and the sensor array is first ventilated and then sealed and stored.

[0042] In this embodiment, color-sensitive materials sensitive to volatile gases of oil samples are selected, mainly including 9 porphyrin compounds and 7 pH indicators. The porphyrin compounds are dissolved in dichloromethane, the pH indicator is dissolved in anhydrous ethanol, and the color-sensitive dyes mainly include 9 porphyrin compounds (tetrachlorophenylporphyrin copper, tetraphenylporphine, tetraphenylporphyrin iron (III) chloride, tetraphenyl-21H, 23H-porphyrin cobalt (II), 5,10,15,20-tetrakis (4-methoxyphenyl)-21H, 23H-porphyrin iron (III) chloride, tetraphenyl-porphyrin zinc, 2,3,7,8,12,13,17,18-octaethyl- 21H,23H-porphine manganese (III) chloride, tetraphenylporphyrin copper, 5,10,15,20-tetrakis (4-methoxyphenyl) porphyrin cobalt (II) methylphenyl porphyrin cobalt) and 7 pH indicators (chlorophenol red, bromocresol green, methyl red, cresol red, phenol red, bromophenol blue, neutral red), the prepared color-sensitive materials were stored in the dark, and 1 μL of the color-sensitive materials were titrated onto a 4m×4m polyvinylidene fluoride (PVDF) membrane using a microcapillary to form a 4×4 sensor array. The prepared sensor array was then placed in a fume hood for 20 minutes, and after stabilization, the sensor array was stored in a sealed bag for later use.

[0043] Furthermore, the specific detection process of using the colorimetric sensor array method to detect the tea oil sample to be tested is: the integrated control and decision-making system stirs the mixed oil sample evenly through a robotic arm, and places the oil sample in the oil sample chamber, starts the cooling fan, activates the heating platform and the LED ring light source, places the sensor array at the center of the sensor array bracket, and directly contacts the reaction chamber with the heating platform. The image acquisition device collects images of the sensor array before and after the reaction, and directly obtains the saturated fatty acid content of the sample and the monounsaturated fatty acid content of the sample from the image.

[0044] In this embodiment, a magnetic stirrer is used to stir the mixed oil sample evenly, the oil sample is placed in the reaction chamber for standby, the cooling fan and the heating platform are turned on, the temperature is set to 100°C, the LED ring light source is started, and after the temperature of the heating platform reaches 100°C, the glass cover of the reaction chamber is opened, the sensor array is placed in the middle of the sensor array bracket, ensuring that the color-sensitive material faces upward and the back faces the oil sample, the cover is closed, the culture dish containing the oil sample is sealed with the sealing ring of the reaction chamber, and the reaction chamber is placed in the positioning partition above the heating platform, ensuring that the reaction chamber is in direct contact with the heating platform and is located directly below the camera. In the integrated control and decision system, a shooting area (ROI) is selected, and the computer controls the camera to shoot the image of the sensor array before the reaction, and the images before and after the reaction are uploaded to the computer, and the RGB value of each color-sensitive point is identified by python, and then the value after the reaction is subtracted from the RGB value before the reaction to generate RGB difference data reflecting different oil sample information. Finally, a 48-dimensional color feature (16 color sensitive points × 3 color features) was obtained, and data related to tea oil adulteration and tea oil color evaluation affected by chemical substances were obtained.

[0045] pass Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 It can be seen that the tea oil adulteration detection device includes a heat dissipation device 1, an integrated control and decision-making system 2, an image acquisition device 3, a reaction chamber 4, a box 5 and a heating device 6, wherein:

[0046] The heat dissipation device 1 includes a heat dissipation fan 16 and a dustproof net 17 .

[0047] The integrated control and decision-making system 2 includes a Raspberry Pi 18 and a display screen 19 .

[0048] The image acquisition device 3 comprises a camera mounting plate 20, a camera 21, bolts 22 and an LED ring light source 23, and is used to acquire images of the colorimetric sensor array.

[0049] The reaction chamber 4 includes a colorimetric sensor array 24 , a colorimetric sensor array support 25 , a square glass cover plate 26 and a sample chamber 27 .

[0050] The box body 5 includes a side cover 7, an upper box body top cover 8, an upper box body 9, a positioning partition 10, a rear plate 11, a lower box body 2 top cover 12, a rear plate 2 13, a lower box body 2 14 and a lower box body 1 15.

[0051] The heating device 6 uses an electric heating plate to provide a heat source.

[0052] Furthermore, the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances specifically include: the relative molecular mass of the sample, the density of the sample, the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample obtained by mass spectrometry, experiments and colorimetry; the data related to tea oil adulteration include the saturated fatty acid content of the sample, the monounsaturated fatty acid content of the sample, the relative molecular mass of the sample and the density of the sample; the data related to the evaluation of tea oil color due to the influence of chemical substances include the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample.

[0053] In this embodiment, the content of saturated fatty acids refers to the content of fatty acids without double bonds in the molecule, the content of monounsaturated fatty acids refers to the content of fatty acids containing one double bond in the molecule, the relative molecular mass refers to the ratio of one molecular mass to 1 / 12 of the mass of a carbon-12 atom, the density is the mass per unit volume, the anisidine value is an indicator for measuring secondary oxidation products in oils and fats, reflecting the degree of oxidation of oils and fats, the thiobarbituric acid reactant value is a measure of the content of secondary oxidation products such as malondialdehyde (MDA) in oils and fats, and the oxidation rate refers to the speed at which oils and fats are oxidized under certain conditions. The content of saturated fatty acids in the above-mentioned samples and the content of monounsaturated fatty acids in the samples can be obtained through images before and after the reaction.

[0054] The saturated fatty acid content and the monounsaturated fatty acid content of the sample can be directly obtained through images, the relative molecular mass of the sample can be obtained by mass spectrometry, the density of the sample can be obtained by the specific gravity bottle method, the anisidine value of the sample and the thiobarbituric acid reactant value of the sample can be obtained by colorimetry, and the oxidation rate of the sample can be obtained by accelerated oxidation experiments.

[0055] Furthermore, the specific preprocessing process for preprocessing the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances is as follows: preprocessing includes data cleaning, data conversion and feature extraction; data cleaning includes filling missing values, deleting outliers and duplicate values; data conversion: normalizing the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances; feature extraction: using the principal component analysis method to perform feature processing on the data related to tea oil adulteration and the data related to the evaluation of tea oil color due to the influence of chemical substances.

[0056] In this embodiment, for example, it was found that oleic acid concentration data was missing in 5 samples, and it was decided to fill these missing values ​​with the average value of the feature. It was found that the oleic acid concentration of 20 samples was much higher than that of other samples. These may be measurement errors, so they were deleted. It was found that the data of 10 samples were completely repeated, and these repeated samples were deleted. The concentration range of oleic acid was converted from the original [30-90] to [0-1]. Data cleaning ensures the integrity and accuracy of the data set and reduces the impact of noise and errors on the model.

[0057] Furthermore, the specific analysis method for analyzing the pretreated tea oil adulteration related data and the pretreated tea oil color evaluation related data due to the influence of chemical substances is as follows: the saturated fatty acid content of the sample, the monounsaturated fatty acid content of the sample, the relative molecular mass of the sample and the density of the sample in the tea oil adulteration related data are analyzed to obtain the tea oil adulteration evaluation coefficient; the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample in the tea oil color evaluation related data due to the influence of chemical substances are analyzed to obtain the tea oil color evaluation coefficient due to the influence of chemical substances.

[0058] In this embodiment, the process of principal component analysis is as follows: through PCA analysis, it is found that the first three principal components can explain 95% of the variance, so these three principal components can replace the original 10 chemical composition features, and feature extraction reduces the dimension of the data, thereby reducing the amount of calculation.

[0059] The specific method for obtaining the tea oil adulteration assessment coefficient is as follows:

[0060] ;

[0061] In the formula, Expressed as tea oil adulteration assessment coefficient, Expressed as the saturated fatty acid content of the sample, Expressed as the content of monounsaturated fatty acids in the sample, Expressed as the relative molecular mass of the sample, Expressed as the standard relative molecular mass of the sample, is the density of the sample, Expressed as the standard density of the sample, Expressed as a natural constant.

[0062] Generally speaking, the higher the content of saturated fatty acids, the greater the relative molecular mass and the higher the density, while the higher the content of monounsaturated fatty acids, the lower the relative molecular mass and the lower the density.

[0063] The standard relative molecular mass and standard density of the sample can be directly obtained through the database.

[0064] The specific method for obtaining the tea oil color evaluation coefficient affected by chemical substances is as follows:

[0065] ;

[0066] In the formula, Expressed as the tea oil color evaluation coefficient affected by chemical substances, Expressed as the anisidine value of the sample, Expressed as the thiobarbituric acid reactant value of the sample, Expressed as the oxidation rate of the sample, Expressed as a natural constant.

[0067] There is a correlation between the anisidine value, thiobarbituric acid reactant value and oxidation rate, which together reflect the degree of lipid oxidation; within a certain period of time, the higher the anisidine value and thiobarbituric acid reactant value, the more oxidation products there are in this fixed time period and the faster the oxidation rate.

[0068] In a specific embodiment, examples of tea oil color evaluation coefficient data affected by chemical substances are shown in the following table.

[0069] Table 1 Example of tea oil color evaluation coefficient data affected by chemical substances

[0070]

[0071] like Figure 8 As shown, it is an image of the tea oil color evaluation coefficient affected by chemical substances in the method for identifying adulterated tea oil based on machine learning algorithm provided in the embodiment of the present application. Figure 8 It can be seen from the data in Table 1 that when the thiobarbituric acid reactant value and the oxidation rate remain unchanged, the larger the anisidine value, the lower the tea oil color evaluation coefficient affected by chemical substances.

[0072] Furthermore, the specific comparison process of comparing the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold is as follows: obtain the first tea oil adulteration assessment threshold and the second tea oil adulteration assessment threshold from the database, the first tea oil adulteration assessment threshold is greater than the second tea oil adulteration assessment threshold, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, if the tea oil adulteration assessment coefficient is greater than or equal to the first tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the third level of adulteration, if the tea oil adulteration assessment coefficient is greater than or equal to the second tea oil adulteration assessment threshold and the tea oil adulteration assessment coefficient is less than the first tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the second level of adulteration, if the tea oil adulteration assessment coefficient is less than the second tea oil adulteration assessment threshold, judge that the tea oil source of the tea oil sample is the first level of adulteration, if the tea oil source of the tea oil sample is the first level of adulteration, conduct a comprehensive evaluation of the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient.

[0073] In this embodiment, if the tea oil adulteration assessment coefficient is greater than or equal to 0.8 (the first threshold), the tea oil from which the tea oil sample is derived is judged to be third-level adulterated; if the tea oil adulteration assessment coefficient is between 0.5-0.8 (between the second threshold and the first threshold), the tea oil from which the tea oil sample is derived is judged to be second-level adulterated; if the tea oil adulteration assessment coefficient is less than 0.5 (less than the second threshold), the tea oil from which the tea oil sample is derived is judged to be first-level adulterated, and it is impossible to determine whether it is adulterated by a machine, and further analysis is required. For example, the tea oil adulteration assessment coefficient is 0.45, which is less than the second threshold (0.5). The tea oil from which the tea oil sample is derived is first-level adulterated, and it may be unadulterated tea oil or highly adulterated, which cannot be distinguished by a machine. A comprehensive assessment is required for the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, and a predetermined threshold is used for judgment, so that the adulteration detection process is standardized and human errors are reduced.

[0074] Furthermore, the specific comparison process of comparing the tea oil grade assessment coefficient with the tea oil grade assessment threshold is: obtaining the first tea oil grade assessment threshold and the second tea oil grade assessment threshold from the database, the first tea oil grade assessment threshold is greater than the second tea oil grade assessment threshold, if the tea oil grade assessment coefficient is greater than or equal to the first tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a superior grade tea oil, if the tea oil grade assessment coefficient is greater than or equal to the second tea oil grade assessment threshold and the tea oil grade assessment coefficient is less than the first tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a first-grade tea oil, if the tea oil grade assessment coefficient is less than the second tea oil grade assessment threshold, it is judged that the tea oil from which the tea oil sample comes is a second-grade tea oil.

[0075] In this embodiment, the first threshold value of tea oil grade evaluation is set to 0.9, and the second threshold value of tea oil grade evaluation is set to 0.7. If the tea oil grade evaluation coefficient is greater than or equal to 0.9, the tea oil sample is judged to be a superior tea oil. If the tea oil grade evaluation coefficient is between 0.7-0.9, the tea oil sample is judged to be a first-class tea oil. If the tea oil grade evaluation coefficient is less than 0.7, the tea oil sample is judged to be a second-class tea oil. Compared with traditional sensory evaluation or complex chemical analysis, the use of machine learning algorithms and preset thresholds can more quickly evaluate the quality of tea oil.

[0076] Superior quality tea oil can be exempted from inspection directly and is judged to be high-quality, unadulterated tea oil. First-quality tea oil has a probability of being adulterated, but it cannot be identified by machine, so relevant personnel will be notified for manual screening. Second-quality tea oil has a high probability of being adulterated. In order to reduce errors, it will be returned to the first step for re-machine inspection.

[0077] like Fig. 9As shown, it is a structural schematic diagram of the adulterated tea oil identification system based on machine learning algorithm provided in the embodiment of the present application. The adulterated tea oil identification system based on machine learning algorithm provided in the embodiment of the present application includes: a sample detection module, a data analysis module and a threshold comparison module:

[0078] Sample detection module: used to collect tea oil samples to be tested, pre-treat the tea oil samples to be tested, and use the colorimetric sensor array method to detect the tea oil samples to be tested, so as to obtain data related to tea oil adulteration and tea oil color evaluation due to the influence of chemical substances;

[0079] Data analysis module: used for preprocessing tea oil adulteration related data and tea oil color evaluation related data due to chemical substances, obtaining preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, analyzing the preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, obtaining tea oil adulteration evaluation coefficient and tea oil color evaluation coefficient due to chemical substances;

[0080] Threshold comparison module: used to obtain the tea oil adulteration assessment threshold from the database, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, obtain the tea oil adulteration assessment threshold comparison result, classify the types of tea oil adulteration according to the tea oil adulteration assessment threshold comparison result, and obtain the tea oil adulteration type classification result;

[0081] Comprehensively evaluate the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, obtain the tea oil grade assessment threshold from the database, compare the tea oil grade assessment coefficient with the tea oil grade assessment threshold to obtain the tea oil grade assessment threshold comparison result, and classify the quality of the source of the tea oil sample to be tested according to the tea oil grade assessment threshold comparison result to obtain the quality classification result of the source of the tea oil to be tested;

[0082] The specific method for obtaining the tea oil grade assessment coefficient is as follows: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

[0083] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0087] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0088] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for identifying adulterated tea oil based on a machine learning algorithm, characterized in that: The following steps are involved: Collect tea oil samples to be tested, pre-treat the tea oil samples to be tested, and use a colorimetric sensor array method to detect the tea oil samples to be tested, so as to obtain data related to adulteration of tea oil and data related to evaluation of tea oil color affected by chemical substances; Preprocessing the tea oil adulteration related data and the tea oil color evaluation related data due to chemical substances to obtain the preprocessed tea oil adulteration related data and the preprocessed tea oil color evaluation related data due to chemical substances, analyzing the preprocessed tea oil adulteration related data and the preprocessed tea oil color evaluation related data due to chemical substances to obtain the tea oil adulteration evaluation coefficient and the tea oil color evaluation coefficient due to chemical substances; Acquire the tea oil adulteration assessment threshold from the database, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, obtain the tea oil adulteration assessment threshold comparison result, classify the types of tea oil adulteration according to the tea oil adulteration assessment threshold comparison result, and obtain the type classification result of tea oil adulteration; Comprehensively evaluate the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, obtain the tea oil grade assessment threshold from the database, compare the tea oil grade assessment coefficient with the tea oil grade assessment threshold to obtain the tea oil grade assessment threshold comparison result, and classify the quality of the source of the tea oil sample to be tested according to the tea oil grade assessment threshold comparison result to obtain the quality classification result of the source of the tea oil to be tested; The specific method for obtaining the tea oil grade assessment coefficient is: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

2. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific pretreatment process of pretreating the tea oil sample to be tested is: The integrated control and decision-making system selects color-sensitive materials that are sensitive to volatile gases in oil samples, dissolves the color-sensitive materials in dichloromethane or anhydrous ethanol through a robotic arm to obtain a color-sensitive material solution, and titrates the color-sensitive material onto a polyvinylidene fluoride membrane to form a sensor array, which is then ventilated and stored in a sealed manner.

3. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific detection process of using the colorimetric sensor array method to detect the tea oil sample to be tested is: The integrated control and decision-making system stirs the mixed oil sample evenly through a robotic arm, places the oil sample in an oil sample chamber, starts a cooling fan, activates a heating platform and an LED ring light source, places the sensor array in the center of a sensor array bracket, brings the reaction chamber into direct contact with the heating platform, and uses an image acquisition device to capture images of the sensor array before and after the reaction, and directly obtains the saturated fatty acid content and the monounsaturated fatty acid content of the sample from the image.

4. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The data related to tea oil adulteration and the data related to tea oil color evaluation affected by chemical substances specifically include: The relative molecular mass of the sample, the density of the sample, the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample were obtained by mass spectrometry, experimental and colorimetric methods; The data related to tea oil adulteration include the content of saturated fatty acids, the content of monounsaturated fatty acids, the relative molecular mass and the density of the samples; The relevant data for the evaluation of tea oil color affected by chemical substances include the anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample.

5. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific preprocessing process of preprocessing the tea oil adulteration related data and the tea oil color evaluation related data affected by chemical substances is as follows: Preprocessing includes data cleaning, data transformation, and feature extraction; Data cleaning includes filling missing values, removing outliers and duplicate values; Data conversion: normalize the data related to tea oil adulteration and the data related to the evaluation of tea oil color affected by chemical substances; Feature extraction: The principal component analysis method is used to perform feature processing on the data related to tea oil adulteration and the data related to the evaluation of tea oil color affected by chemical substances.

6. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific analysis method for analyzing the data related to adulteration of pretreated tea oil and the data related to color evaluation of pretreated tea oil affected by chemical substances is as follows: The saturated fatty acid content, monounsaturated fatty acid content, relative molecular mass and density of the samples in the tea oil adulteration related data were analyzed to obtain the tea oil adulteration assessment coefficient; The anisidine value of the sample, the thiobarbituric acid reactant value of the sample and the oxidation rate of the sample in the relevant data of the tea oil color evaluation due to the influence of chemical substances are analyzed to obtain the tea oil color evaluation coefficient due to the influence of chemical substances.

7. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific comparison process of comparing the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold is as follows: The first threshold value and the second threshold value of tea oil adulteration assessment are obtained from the database. The first threshold value of tea oil adulteration assessment is greater than the second threshold value of tea oil adulteration assessment. The tea oil adulteration assessment coefficient is compared with the tea oil adulteration assessment threshold. If the tea oil adulteration assessment coefficient is greater than or equal to the first threshold value of tea oil adulteration assessment, the tea oil from which the tea oil sample is derived is judged to be third-level adulterated. If the tea oil adulteration assessment coefficient is greater than or equal to the second threshold value of tea oil adulteration assessment and the tea oil adulteration assessment coefficient is less than the first threshold value of tea oil adulteration assessment, the tea oil from which the tea oil sample is derived is second-level adulterated. If the tea oil adulteration assessment coefficient is less than the second threshold value of tea oil adulteration assessment, the tea oil from which the tea oil sample is derived is first-level adulterated. If the tea oil from which the tea oil sample is derived is first-level adulterated, the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances are comprehensively evaluated to obtain the tea oil grade assessment coefficient.

8. The method for identifying adulterated tea oil based on a machine learning algorithm as claimed in claim 1, characterized in that: The specific comparison process of comparing the tea oil level assessment coefficient with the tea oil level assessment threshold is as follows: The first threshold value and the second threshold value of tea oil level assessment are obtained from the database. The first threshold value of tea oil level assessment is greater than the second threshold value of tea oil level assessment. If the tea oil level assessment coefficient is greater than or equal to the first threshold value of tea oil level assessment, the tea oil from which the tea oil sample comes is judged to be a superior grade tea oil. If the tea oil level assessment coefficient is greater than or equal to the second threshold value of tea oil level assessment and the tea oil level assessment coefficient is less than the first threshold value of tea oil level assessment, the tea oil from which the tea oil sample comes is judged to be a first grade tea oil. If the tea oil level assessment coefficient is less than the second threshold value of tea oil level assessment, the tea oil from which the tea oil sample comes is judged to be a second grade tea oil.

9. The adulterated tea oil identification system based on machine learning algorithm is characterized by: Including sample detection module, data analysis module and threshold comparison module: Sample detection module: used to collect tea oil samples to be tested, pre-treat the tea oil samples to be tested, and use the colorimetric sensor array method to detect the tea oil samples to be tested, so as to obtain data related to tea oil adulteration and tea oil color evaluation due to the influence of chemical substances; Data analysis module: used for preprocessing tea oil adulteration related data and tea oil color evaluation related data due to chemical substances, obtaining preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, analyzing the preprocessed tea oil adulteration related data and preprocessed tea oil color evaluation related data due to chemical substances, obtaining tea oil adulteration evaluation coefficient and tea oil color evaluation coefficient due to chemical substances; Threshold comparison module: used to obtain the tea oil adulteration assessment threshold from the database, compare the tea oil adulteration assessment coefficient with the tea oil adulteration assessment threshold, obtain the tea oil adulteration assessment threshold comparison result, classify the types of tea oil adulteration according to the tea oil adulteration assessment threshold comparison result, and obtain the tea oil adulteration type classification result; Comprehensively evaluate the tea oil adulteration assessment coefficient and the tea oil color assessment coefficient affected by chemical substances to obtain the tea oil grade assessment coefficient, obtain the tea oil grade assessment threshold from the database, compare the tea oil grade assessment coefficient with the tea oil grade assessment threshold to obtain the tea oil grade assessment threshold comparison result, and classify the quality of the source of the tea oil sample to be tested according to the tea oil grade assessment threshold comparison result to obtain the quality classification result of the source of the tea oil to be tested; The specific method for obtaining the tea oil grade assessment coefficient is: ; In the formula, Eco represents the tea oil grade assessment coefficient, Expressed as tea oil adulteration assessment coefficient, Expressed as the weight factor of the tea oil adulteration assessment coefficient, Expressed as the tea oil color evaluation coefficient affected by chemical substances, It is expressed as the weight factor of the tea oil color evaluation coefficient affected by chemical substances, Expressed as a natural constant.

Images