Method for identifying adulterated rapeseed oil in linseed oil

Abstract

The invention discloses a method for identifying adulterated rapeseed oil in linseed oil. The method comprises the following steps: preparing a standard substance, a sample and a reagent; carrying out fatty acid analysis pretreatment by utilizing a full-automatic dilution standard preparation instrument; analyzing the content distribution of each fatty acid in the to-be-detected solutions of the standard sample, the linseed oil sample, the rapeseed oil sample and the adulterated sample by using a gas chromatographic method; analyzing component data by using a principal component analysis method; and establishing a chemometrics-optimization algorithm discrimination model, training by using the component data, and screening out an optimal discrimination model. And identifying the to-be-detected sample based on the optimal discrimination model. According to the method, automatic pretreatment can be realized, the characteristics of automation, high efficiency and accurate reagent moving and taking of a full-automatic standard configuration instrument are utilized, the operation error in the pretreatment process of a large batch of samples is solved, and the efficiency and the accuracy are improved. The method is combined with chemometrics, so that the adulteration condition of the linseed oil can be quickly and accurately judged.

Description

Technical Field

[0001] The invention belongs to the technical field of food detection and analysis, and specifically relates to a method for determining the fatty acid composition of edible vegetable oil by using automatic pre-processing technology, and then analyzing the data by chemometrics, combined with principal component analysis and optimization algorithm to realize identification of adulterated rapeseed oil in linseed oil. Background Art

[0002] Flaxseed oil is rich in unsaturated fatty acids, especially α-linolenic acid, which has the functions of regulating blood lipids, lowering cholesterol, anti-oxidation, and preventing cardiovascular diseases. At present, the flaxseed oil on the market is relatively niche among conventional edible vegetable oils, but because of its unique nutritional value, flaxseed oil is still popular among consumers. However, compared with traditional edible vegetable oils such as soybean oil and rapeseed oil, the unit price of flaxseed oil is higher. In order to pursue profits, some unscrupulous merchants began to sell inferior products as good ones and mix them with other low-cost oils. These adulteration behaviors not only harm the interests of consumers, but are also likely to have a negative impact on health. Rapeseed oil and flaxseed oil are both plant seed oils with similar properties, and are easily regarded as the source of flaxseed oil adulteration. However, the fatty acid composition of rapeseed oil is quite different from that of flaxseed oil, and the content of α-linolenic acid is less, and the nutritional content is lower than that of flaxseed oil. Consumers cannot achieve the expected health effects by taking such oils. Therefore, identifying the adulteration of flaxseed oil has become an urgent problem to be solved.

[0003] At present, the main methods for identifying adulterated edible oils at home and abroad are: physical and chemical testing, spectral analysis, chromatography and rapid detection methods based on electronic nose. Among them, the physical and chemical testing method is less efficient and suitable for the analysis of a small amount of samples; near-infrared spectroscopy, Raman spectroscopy, fluorescence spectroscopy and other spectral analysis techniques are fast and simple, but the spectrogram signals of edible vegetable oils with similar properties are prone to overlap, and the characteristic parameters are difficult to distinguish; the electronic nose method requires a considerable amount of sample results to establish a spectral library, and the accuracy is limited by the instrument status, the detection limit is low, and it is suitable as a supplementary inspection method; chromatographic analysis techniques include liquid chromatography, gas chromatography and mass spectrometry, among which gas chromatography is the method specified in GB 5009.168 "National Food Safety Standard Determination of Fatty Acids in Food", and is also a conventional method for fatty acid testing in edible oils. Among them, the ester exchange method is simple and quick in pretreatment, and is suitable for identifying edible oil adulteration through the difference in the content of different characteristic fatty acids.

[0004] In the prior art, the ester exchange method is often used to methylate the fatty acids in oils and fats. This method is convenient and fast, but it is time-consuming and laborious to process large batches of samples (such as more than 100 batches). In addition, the pre-treatment operation of the ester exchange method requires accurate pipetting of organic reagents and control of reaction time, which is difficult to accurately control in large batches of samples.

[0005] The distribution of fatty acids in edible vegetable oils is relatively complex. Some fatty acid categories of flaxseed oil and rapeseed oil specified in the national standards are not sufficient to serve as a basis for identifying adulteration. There is varying degrees of overlap in their fatty acid content, making it difficult to determine.

[0006] Therefore, there is an urgent need for a method for identifying linseed oil adulterated with rapeseed oil with high automation, fast detection speed and good accuracy. Summary of the invention

[0007] In view of the above technical problems, the present invention provides a method for identifying linseed oil adulterated with rapeseed oil. The method adopts automated pre-processing to avoid human operational errors, and combines gas chromatography analysis and machine learning algorithms to achieve automated rapid detection, providing operational reference for large-scale detection and identification of bulk oil adulteration, and also providing technical support for quality control and supervision of the oil industry.

[0008] The technical solution provided by the present invention is as follows:

[0009] The method for identifying linseed oil adulterated with rapeseed oil comprises the following steps:

[0010] Preparation of standards, samples, and reagents;

[0011] Using a fully automatic dilution and labeling instrument to perform fatty acid analysis pretreatment, to obtain test solutions of standard samples, linseed oil samples, rapeseed oil samples and adulterated samples;

[0012] Analyzing the content distribution of each fatty acid in the test solution of the standard sample, the linseed oil sample, the rapeseed oil sample and the adulterated sample by gas chromatography to obtain component data including the content distribution of the linseed oil sample, the rapeseed oil sample and the adulterated sample;

[0013] The principal component analysis method was used to analyze the component data and confirm the characteristic components;

[0014] Establish a chemometrics-optimization algorithm discrimination model, use component data for training, and screen out the optimal discrimination model;

[0015] Identify the samples to be tested based on the optimal discriminant model.

[0016] In one possible implementation, the standard is a mixed standard of 37 fatty acid methyl esters; the samples include linseed oil, rapeseed oil and adulterated samples; and the reagents include isooctane, potassium hydroxide-methanol solution and sodium bisulfate.

[0017] In one possible implementation, the method for pre-processing fatty acid analysis using a fully automatic dilution and labeling instrument comprises the following steps:

[0018] Place each sample and reagent in a different area;

[0019] A fully automatic dilution and labeling instrument program was written to realize the methyl esterification reaction steps of automated liquid collection and vortex mixing to obtain the sample solution.

[0020] In one possible implementation, the adulterated sample is linseed oil mixed with rapeseed oil of gradient mass concentrations.

[0021] Further, linseed oil was weighed; and pure rapeseed oil was added in an amount of 5%, 10%, 15%, 20%, 50% and 80% by mass of linseed oil respectively.

[0022] In a possible implementation, the chromatographic conditions of the gas chromatography are as follows:

[0023] Detector: Hydrogen flame ionization detector FID;

[0024] Capillary column: SP-2560 capillary column;

[0025] Inlet temperature: 260°C;

[0026] Detector temperature: 260°C;

[0027] Flow rate: 1 mL / min;

[0028] Program temperature rise: initial temperature 100℃, hold for 1 min; heat to 176℃ at 8℃ / min, hold for 5 min; heat to 200℃ at 2℃ / min, hold for 51 min; heat to 240℃ at 28℃ / min, hold for 8.5 min;

[0029] Carrier gas: N2; split ratio: 100:1; injection volume: 1.0 μL.

[0030] In a possible implementation, the method of analyzing component data using principal component analysis to identify characteristic components includes:

[0031] The samples were set as linseed oil group, rapeseed oil group and adulterated group;

[0032] Principal component analysis was performed on the component data to obtain the three factors that contributed most to the differences;

[0033] The characteristic components that differ between linseed oil and rapeseed oil were confirmed by two-dimensional scatter plots and three-dimensional distribution plots.

[0034] Furthermore, the factors contributing to the difference include C18:1n9c, C18:3n3 and C18:2n6c.

[0035] In one possible implementation, the chemometrics-optimization algorithm discrimination model includes a partial least squares model, a support vector machine model, a Hackney Bayes model, a decision tree model, and a neural network model.

[0036] Furthermore, the method for selecting the optimal discriminant model includes:

[0037] Divide the component data into a training group and a prediction group;

[0038] Import the data of the training group into the chemometrics-optimization algorithm discrimination model for training;

[0039] After the training is completed, the chemometrics-optimization algorithm discrimination model is used to judge the samples in the prediction group, and the prediction accuracy and detection limit of each model are compared to determine the optimal discrimination model.

[0040] The beneficial effects of the present invention are as follows:

[0041] (1) Pre-processing automation: The use of a fully automatic dilution and labeling instrument can take advantage of the device's accurate pipetting and time control, reduce operational errors and improve experimental efficiency.

[0042] (2) The gas chromatography analysis method adopted in the present invention can achieve complete separation of 37 chromatographic peaks in the fatty acid mixed standard by controlling the chromatographic conditions, thereby improving the accuracy of component analysis.

[0043] (3) The present invention adopts chemometric analysis, which can perform cluster analysis on multi-component data of large batches of samples, and identify the adulteration of oils and fats through multi-dimensional analysis. In addition, it combines autonomous learning to establish an identification model, which can achieve rapid and accurate determination of oil and fat adulteration. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a physical picture of the fully automatic dilution and labeling instrument;

[0045] Figure 2 This is the gas chromatogram of fatty acid composition of a representative sample of pure rapeseed oil;

[0046] Figure 3 This is the gas chromatogram of the fatty acid composition of a representative sample of pure linseed oil;

[0047] Figure 4 This is a two-dimensional principal component analysis diagram of pure rapeseed oil, pure linseed oil and adulterated oil samples; red represents pure rapeseed oil, yellow represents pure linseed oil, and blue represents adulterated oil samples;

[0048] Figure 5This is a three-dimensional principal component analysis diagram of pure rapeseed oil, pure linseed oil and adulterated oil samples; red represents pure rapeseed oil, yellow represents pure linseed oil, and blue represents adulterated oil samples;

[0049] Figure 6 This is the distribution diagram of C18:1n9c and C18:3n3 content in linseed oil and rapeseed oil samples. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and beneficial effects of the present invention clearer, the contents of the present invention are further explained and illustrated in conjunction with specific embodiments below.

[0051] A method for identifying linseed oil adulterated with rapeseed oil comprises the following steps:

[0052] S100, preparation of standards, samples, and reagents;

[0053] S200, using a fully automatic dilution and labeling instrument to perform fatty acid analysis pretreatment to obtain test solutions of standard samples, linseed oil samples, rapeseed oil samples and adulterated samples;

[0054] S300, analyzing the content distribution of each fatty acid in the test solution of the standard sample, the linseed oil sample, the rapeseed oil sample and the adulterated sample by gas chromatography, and obtaining component data including the content distribution of the linseed oil sample, the rapeseed oil sample and the adulterated sample;

[0055] S400, analyzing component data using principal component analysis to identify characteristic components;

[0056] S500, establish a chemometrics-optimization algorithm discrimination model, use component data for training, and select the optimal discrimination model;

[0057] S600, identifying the sample to be detected based on the optimal discriminant model.

[0058] In one possible implementation, the standard is a mixed standard of 37 fatty acid methyl esters; the samples include linseed oil, rapeseed oil and simulated adulterated samples; and the reagents include isooctane, potassium hydroxide-methanol solution and sodium bisulfate.

[0059] Furthermore, the 37 fatty acid methyl ester mixed standard contains the following components: C4:0 methyl ester, C6:0 methyl ester, C8:0 methyl ester, C10:0 methyl ester, C11:0 methyl ester, C12:0 methyl ester, C13:0 methyl ester, C14:0 methyl ester, C14:1n5 methyl ester, C15:0 methyl ester, C15:1n5 methyl ester, C16:0 methyl ester, C16:1n7 methyl ester, C17:0 methyl ester, C17:1n7 methyl ester, C18:0 methyl ester, C18:1n9t methyl ester, C18:1n9c methyl ester. ester, C18:2n9t,12t methyl ester, C18:2n6c methyl ester, C20:0 methyl ester, C18:3n6 methyl ester, C20:1 methyl ester, C18:3n3 methyl ester, C21:0 methyl ester, C20:2 methyl ester, C22:0 methyl ester, C20:3n6 methyl ester, C22:1n9 methyl ester, C20:3n3 methyl ester, C20:4n6 methyl ester, C23:0 methyl ester, C22:2 methyl ester, C20:5n3 methyl ester, C24:0 methyl ester, C24:1 methyl ester and C22:6n3 methyl ester. The 37 kinds of fatty acid methyl ester mixed standards are the fatty acid methyl ester mixed standards specified in GB 5009.168.

[0060] In one possible implementation, the pretreatment method for fatty acid analysis using a fully automatic dilution and labeling instrument comprises the following steps:

[0061] Place each sample and reagent in a different area;

[0062] A fully automatic dilution and labeling instrument program was written to realize the methyl esterification reaction steps of automated liquid collection and vortex mixing to obtain the sample solution.

[0063] In a possible implementation, the chromatographic conditions of the gas chromatography are as follows:

[0064] Detector: Hydrogen flame ionization detector FID;

[0065] Capillary column: SP-2560 capillary column;

[0066] Inlet temperature: 260°C;

[0067] Detector temperature: 260°C;

[0068] Flow rate: 1 mL / min;

[0069] Program temperature rise: initial temperature 100℃, hold for 1 min; heat to 176℃ at 8℃ / min, hold for 5 min; heat to 200℃ at 2℃ / min, hold for 51 min; heat to 240℃ at 28℃ / min, hold for 8.5 min;

[0070] Carrier gas: N2; split ratio: 100:1; injection volume: 1.0 μL.

[0071] In one possible implementation, the adulterated sample is linseed oil mixed with rapeseed oil of gradient concentrations.

[0072] Furthermore, the preparation method of the adulterated sample is: weigh linseed oil; and add 5%, 10%, 15%, 20%, 50%, and 80% of pure rapeseed oil by weight of linseed oil respectively.

[0073] In a possible implementation, the method of analyzing component data using principal component analysis to identify characteristic components includes:

[0074] The samples were set as linseed oil group, rapeseed oil group and adulterated group;

[0075] Principal component analysis was performed on the component data to obtain the three factors that contributed most to the differences;

[0076] The characteristic components that differ between linseed oil and rapeseed oil were confirmed by two-dimensional scatter plots and three-dimensional distribution plots.

[0077] Furthermore, the factors contributing to the difference include C18:1n9c, C18:3n3 and C18:2n6c.

[0078] In one possible implementation, the chemometrics-optimization algorithm discrimination model includes a partial least squares model, a support vector machine model, a Hackney Bayes model, a decision tree model, and a neural network model.

[0079] Furthermore, the method for selecting the optimal discriminant model includes:

[0080] Divide the component data into a training group and a prediction group;

[0081] Import the data in the training group into the chemometrics-optimization algorithm discrimination model for training;

[0082] After the training is completed, the chemometrics-optimization algorithm discrimination model is used to judge the samples in the prediction group, and the prediction accuracy and detection limit of each model are compared to determine the optimal discrimination model.

[0083] This will be described below in conjunction with more specific embodiments.

[0084] The method for identifying linseed oil adulterated with rapeseed oil comprises the following steps:

[0085] 1. Analysis of fatty acid composition of pure rapeseed oil

[0086] 1. Preparation of standards, samples and reagents:

[0087] Standards: 37 kinds of fatty acid methyl esters mixed standards;

[0088] Samples: linseed oil, rapeseed oil and simulated adulterated samples;

[0089] Reagents: isooctane (chromatographic grade), potassium hydroxide-methanol solution (2 mol / L) and sodium bisulfate.

[0090] 2. See Figure 1 , use the fully automatic dilution and standardization instrument to prepare the sample solution with gradient concentration to realize automatic pretreatment. The operating procedures for setting up the fully automatic dilution and standardization instrument are as follows:

[0091] ① Basic settings: Set the A, B, and C areas of the calibration instrument to three specifications: 30mL×32, 2mL×96, and 12mL×72. Add about 60mg of rapeseed oil sample (about 3 to 4 drops) to the C140 bottle to be tested in area C, add 12mL of 2mol / L potassium hydroxide-methanol solution to the C189 bottle, weigh about 1g of sodium bisulfate in the A1 bottle, and connect the pipe in the SO1 solvent bottle to isooctane. Set the vortex mixing speed to 500rpm, 3 times. Bottle B44 is the final injection vial.

[0092] ② Write a fully automatic dilution and labeling instrument program to realize the methyl esterification reaction steps of automated liquid collection and vortex mixing to obtain the sample solution.

[0093] The details are as follows:

[0094] Serial number Function Liquid collection location Liquid volume Target location Target volume Remove solvent Mix time 1 Transfer solvent SO1 4.0mL C140 4.0mL Isooctane Vortex mixing 2 Transfer solution C189 0.2mL C140 0.2mL Vortex mixing 3 Transfer solution C140 4.2mL A1 4.2mL Vortex mixing 4 pause none 4min 5 Transfer solution A1 1.0mL B44 1.0mL none

[0095] The program can be set up cyclically according to the test requirements to obtain the sample solution required for the test.

[0096] 3. Analysis of fatty acid composition of pure rapeseed oil samples by gas chromatography:

[0097] The specific chromatographic conditions are:

[0098] Detector: Hydrogen flame ionization detector FID;

[0099] Capillary column: SP-2560 capillary column (100 m × 0.25 mm × 0.25 μm, Supelco);

[0100] Inlet temperature: 260°C;

[0101] Detector temperature: 260°C;

[0102] Program temperature rise: initial temperature 100℃, hold for 1 min; heat to 176℃ at 8℃ / min, hold for 5 min; heat to 200℃ at 2℃ / min, hold for 51 min; heat to 240℃ at 28℃ / min, hold for 8.5 min;

[0103] Carrier gas: N2 (purity 99.999%); split ratio: 100:1; injection volume: 1.0 μL.

[0104] Chromatographic analysis was performed on the test standard and rapeseed oil sample solutions. The retention time of 37 fatty acid methyl esters was used to qualitatively analyze the fatty acid composition in the pure rapeseed oil sample using the normalization method, and the content distribution of each fatty acid in the rapeseed oil sample was obtained by the conversion coefficient. Figure 2 Gas chromatogram of fatty acid composition of a representative sample of pure rapeseed oil.

[0105] 2. Analysis of fatty acid composition of pure linseed oil

[0106] 1. Preparation of standards, samples and reagents:

[0107] Standards: 37 kinds of fatty acid methyl esters mixed standards;

[0108] Sample: Pure linseed oil sample;

[0109] Reagents: isooctane (chromatographic grade), potassium hydroxide-methanol solution (2 mol / L) and sodium bisulfate.

[0110] 2. Set up the operating procedures of the fully automatic dilution and labeling instrument:

[0111] ① Basic settings: Set the A, B, and C areas of the calibration instrument to three specifications: 30mL×32, 2mL×96, and 12mL×72. Add about 60mg of pure rapeseed oil sample (about 3 to 4 drops) to the C141 bottle to be tested in area C, add 12mL of 2mol / L potassium hydroxide-methanol solution to the C189 bottle, weigh about 1g of sodium bisulfate in the A2 bottle, and connect the pipe in the SO1 solvent bottle to isooctane. Set the vortex mixing speed to 500rpm, 3 times. Bottle B45 is the final injection vial.

[0112] ② Editing operation method: (The program can be set up cyclically according to the actual experimental samples)

[0113] Serial number Function Liquid collection location Liquid volume Target location Target volume Remove solvent Mix time 1 Transfer solvent SO1 4.0mL C141 4.0mL Isooctane Vortex mixing 2 Transfer solution C189 0.2mL C141 0.2mL Vortex mixing 3 Transfer solution C141 4.2mL A2 4.2mL Vortex mixing 4 pause none 4min 5 Transfer solution A2 1.0mL B45 1.0mL none

[0114] 3. Analysis of fatty acid composition of pure linseed oil samples by gas chromatography:

[0115] The specific chromatographic conditions are:

[0116] Detector: Hydrogen flame ionization detector FID;

[0117] Capillary column: SP-2560 capillary column (100 m × 0.25 mm × 0.25 μm, Supelco);

[0118] Inlet temperature: 260°C;

[0119] Detector temperature: 260°C;

[0120] Program temperature rise: initial temperature 100℃, hold for 1 min; heat to 176℃ at 8℃ / min, hold for 5 min; heat to 200℃ at 2℃ / min, hold for 51 min; heat to 240℃ at 28℃ / min, hold for 8.5 min;

[0121] Carrier gas: N2 (purity 99.999%); split ratio: 100:1; injection volume: 1.0 μL.

[0122] Chromatographic analysis was performed on the test standard and pure linseed oil sample solutions. The retention time of 37 fatty acid methyl esters was used to qualitatively analyze the fatty acid composition in the pure linseed oil sample using the normalization method, and the content distribution of each fatty acid in the oil sample was obtained by the conversion coefficient. Figure 3 Gas chromatogram of the fatty acid composition of a representative sample of pure linseed oil.

[0123] It should be noted that due to the wide variety of fatty acids and their complex composition, during the gas chromatography analysis, many chromatographic peaks overlap, interfering with the analysis results and taking too long to run. However, the gas chromatography program described in this embodiment can achieve complete separation of 37 chromatographic peaks by improving and optimizing the heating program. Figure 2 , Figure 3 As shown, the chromatographic analysis method described in this embodiment can achieve complete separation of 37 chromatographic peaks.

[0124] 3. Identification of flaxseed oil adulterated with rapeseed oil based on chemometrics

[0125] Pure linseed oil and pure rapeseed oil samples were selected for simulated adulteration experiments: pure rapeseed oil with contents of 5%, 10%, 15%, 20%, 50%, and 80% was added to the pure linseed oil, and the fatty acid composition was analyzed by the methods in step 1 and step 2 to obtain the content data of each fatty acid in the adulterated samples.

[0126] The content data of each fatty acid in pure linseed oil sample, pure rapeseed oil sample and adulterated sample were combined to obtain the component data.

[0127] Principal component analysis: The above component data were imported into Agilent Mass Profiler Professional software, and the data were divided into linseed oil group, rapeseed oil group and adulteration group. Among them, the adulteration group is the data of the adulterated sample. All fatty acid data in each sample were included in the analysis, and principal component analysis was performed after One-way ANOVA processing to obtain the three factors that contributed the most to the difference. Figure 4 and Figure 5It can be seen that the three factors with the largest contribution to the difference are C18:1n9c, C18:3n3 and C18:2n6c, and their contribution rates are 29.8%, 17.11% and 11.36% respectively.

[0128] Establishment of adulteration model and optimization algorithm: Randomly select 3 / 4 of the data from all the above pure rapeseed oil samples, pure linseed oil samples, and adulterated samples of different proportions as training groups, and the rest of the data as prediction groups, and import them into Agilent MassProfiler Professional software. After marking the data in the training group as linseed oil and non-linseed oil, machine learning is performed. The acquired information is processed and learned through compound deconvolution, compound search and chemometric analysis, and five models are established: partial least squares, support vector machine, Hackberry Bayes, decision tree and neural network. The prediction group is imported into each model, and the type of oil is determined by the established model analysis. The prediction accuracy and detection limit of the five discrimination methods are compared to determine the optimal algorithm. After testing, the accuracy of the five discrimination models are: partial least squares 87.50%, support vector machine 83.33%, Hackberry Bayes 91.67%, decision tree 95.83%, neural network 87.50%. It can be seen that the decision tree has the highest discrimination accuracy (≥95%), confirming that it is the optimal discrimination model. When the optimal discrimination model takes the sample selected in this experiment as an example, its detection limit of adulteration ratio is 10% (credibility is 100%), which has a low detection limit and high credibility, and thus has a good identification accuracy.

[0129] The above description is only a preferred specific implementation manner of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention should be included in the scope of protection of the invention.

Claims

1. A method for identifying linseed oil adulterated with rapeseed oil, characterized in that: The following steps are involved: Preparation of standards, samples, and reagents; Using a fully automatic dilution and labeling instrument to perform fatty acid analysis pretreatment, to obtain test solutions of standard samples, linseed oil samples, rapeseed oil samples and adulterated samples; Analyzing the content distribution of each fatty acid in the test solution of the standard sample, the linseed oil sample, the rapeseed oil sample and the adulterated sample by gas chromatography to obtain component data including the content distribution of the linseed oil sample, the rapeseed oil sample and the adulterated sample; The principal component analysis method was used to analyze the component data and confirm the characteristic components; Establish a chemometrics-optimization algorithm discrimination model, use component data for training, and screen out the optimal discrimination model; Identify the samples to be tested based on the optimal discriminant model.

2. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The standard substance is a mixed standard substance of 37 fatty acid methyl esters; the samples include linseed oil, rapeseed oil and adulterated samples; and the reagents include isooctane, potassium hydroxide-methanol solution and sodium bisulfate.

3. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The method for pre-processing fatty acid analysis using a fully automatic dilution and labeling instrument comprises the following steps: Place each sample and reagent in a different area; A fully automatic dilution and labeling instrument program was written to realize the methyl esterification reaction steps of automated liquid collection and vortex mixing to obtain the sample solution.

4. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The adulterated sample is linseed oil mixed with rapeseed oil of gradient mass concentration.

5. The method for identifying linseed oil adulterated with rapeseed oil according to claim 4, characterized in that: The preparation method of the adulterated sample is as follows: weigh linseed oil; and add 5%, 10%, 15%, 20%, 50% and 80% of pure rapeseed oil respectively by mass of the linseed oil.

6. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The chromatographic conditions of the gas chromatography are as follows: Detector: Hydrogen flame ionization detector FID; Capillary column: SP-2560 capillary column; Inlet temperature: 260°C; Detector temperature: 260°C; Flow rate: 1 mL / min; Program temperature rise: initial temperature 100℃, hold for 1 min; heat to 176℃ at 8℃ / min, hold for 5 min; heat to 200℃ at 2℃ / min, hold for 51 min; heat to 240℃ at 28℃ / min, hold for 8.5 min; Carrier gas: N2; split ratio: 100:1; injection volume: 1.0 μL.

7. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The method of analyzing component data using principal component analysis to confirm characteristic components includes: The samples were set as linseed oil group, rapeseed oil group and adulterated group; Principal component analysis was performed on the component data to obtain the three factors that contributed most to the differences; The characteristic components that differ between linseed oil and rapeseed oil were confirmed by two-dimensional scatter plots and three-dimensional distribution plots.

8. The method for identifying linseed oil adulterated with rapeseed oil according to claim 7, characterized in that: The factors contributing to the difference include C18:1n9c, C18:3n3 and C18:2n6c.

9. The method for identifying linseed oil adulterated with rapeseed oil according to claim 1, characterized in that: The chemometrics-optimization algorithm discrimination model includes a partial least squares model, a support vector machine model, a Hackney Bayes model, a decision tree model and a neural network model.

10. The method for identifying linseed oil adulterated with rapeseed oil according to claim 9, characterized in that: The method for screening out the optimal discriminant model comprises: Divide the component data into a training group and a prediction group; Import the data of the training group into the chemometrics-optimization algorithm discrimination model for training; After the training is completed, the chemometrics-optimization algorithm discrimination model is used to judge the samples in the prediction group, and the prediction accuracy and detection limit of each model are compared to determine the optimal discrimination model.

Images