CN104198529A - Method for distinguishing donkey-hide gelatin by utilizing electronic nose technology - Google Patents

Abstract

The invention discloses a method for identifying donkey-hide gelatin using electronic nose technology. Select donkey-hide gelatin products from specific manufacturers as samples; test each sample in the sample separately, crush the sample into pieces, weigh a certain amount and place it in a sealed sample bottle, and stir After a period of time, a part of the gas in the sample bottle is extracted and injected into the electronic nose air chamber, and the sensor array in the electronic nose air chamber collects sample data; the sample data in the sample is divided into calibration set data and verification set data, and the calibration set data is used to The discriminant model is established from the data, and the established discriminant model is verified through the verification set data; the sample data of the donkey-hide gelatin sample to be detected is collected, and the principal component data of the donkey-hide gelatin sample to be detected is calculated. The principal component center of the principal component analysis model of the training set Distance, to obtain the F value of the donkey-hide gelatin sample to be tested, according to the set confidence interval of the homogeneity of variance test, judge whether there is a significant difference in the precision, and the donkey-hide gelatin of the specific manufacturer does not have a significant difference.

Description

A method for identifying donkey-hide gelatin using electronic nose technology

technical field

The invention relates to the field of identification of donkey-hide gelatin, in particular to a method for identifying donkey-hide gelatin using electronic nose technology.

Background technique

Ejiao, which has been used for 3,000 years, has always been known as the "Holy Medicine for Blood" and "National Treasure of Nourishment". Together with ginseng and antler, it is known as the "Three Treasures of Traditional Chinese Medicine". The origin of donkey-hide gelatin is "Dong'e County" in Shandong. Dong'e Ejiao has always been famous all over the world. Until today, "Dong'e Ejiao" has almost become synonymous with Ejiao. At the same time, it has also evolved into the abbreviation of Shandong Dong'e Ejiao Co., Ltd., the largest manufacturer of Ejiao and its series of products in China.

At present, there are many donkey-hide gelatin manufacturers. Due to different raw materials and production processes, the product quality is uneven. However, the current pharmacopoeia standard setting is relatively low, and it is impossible to judge the authenticity and quality level of donkey-hide gelatin. Selling poor-quality donkey-hide gelatin in the name of donkey-hide gelatin produced by well-known trademark manufacturers disrupts market order, adversely affects the reputation of famous trademark manufacturers, and at the same time deceives and hurts consumers. In this context, there is an urgent need for an objective, fast and effective method to identify donkey-hide gelatin products, so that consumers can distinguish donkey-hide gelatin products from well-known manufacturers from other manufacturers' donkey-hide gelatin products.

Contents of the invention

The invention provides a method for identifying donkey-hide gelatin using electronic nose technology, which is used to identify donkey-hide gelatin products objectively, quickly and effectively, so as to distinguish donkey-hide gelatin products from specific manufacturers from other manufacturers' donkey-hide gelatin products.

In order to achieve the above object, the present invention provides a method for identifying donkey-hide gelatin using electronic nose technology, comprising the following steps:

a) Select donkey-hide gelatin products from specific manufacturers as samples;

b) detecting each sample in the samples, the detection process is:

b1) Weigh 0.5-2.0g of the sample powder after crushing and sieving the sample, place it in a sealed sample bottle with a volume of 5-20ml, stir for 1000s-3000s, take 1-5ml of the gas in the sample bottle and inject it The needle is injected into the electronic nasal air chamber; the electronic nasal air chamber is provided with 3 chambers, and the 3 chambers are provided with a sensor array composed of a total of 18 sensors;

b2) The sensor array collects the sample data in the sample bottle; wherein, the sample data of each sample is collected 3 times and averaged; the collected sample data is the characteristic response spectrum of the sample odor;

c) dividing all sample data in the sample into calibration set data and verification set data, using the calibration set data to establish a discriminant model, and verifying the established discriminant model through the verification set data;

Wherein the method for establishing a discriminant model using the calibration set data is:

c1) Preprocessing the data of a single sample, extracting feature information and performing the following transformations:

${\mathrm{<span\; class="notranslate">\; SNVs</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}{\sqrt{\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}<span\; class="notranslate">\; ,</span>$

Among them, SNV _i is the standard normal variable of the i-th sensor response value in the electronic nose signal of a single sample; x _i is the response value of the i-th sensor in the single sample; is the average value of all sensor response values in the electronic nose signal of the single sample; p is the number of sensors in the electronic nose air chamber; i is a natural number, and its value ranges from 1 to p;

c2) Obtain the preprocessed data of all samples in the calibration set, use the SIMCA method to establish a principal component analysis discriminant model for the sample, first use the samples in the calibration set as a training set, and use the sample data matrix of the training set Carrying out principal component analysis respectively, establishing the principal component analysis discriminant model of the training set and optimizing the principal component analysis discriminant model by leave-one-out method; wherein the sample data matrix is composed of the peak values of the response values of p sensors;

d) collecting the sample data of the donkey-hide gelatin sample to be detected according to the method of step b, and preprocessing the sample data of the donkey-hide gelatin sample to be detected according to the method of step c1, substituting the sample data of the donkey-hide gelatin sample to be detected into the principal component analysis discriminant model, Obtain the principal component data of the donkey-hide gelatin sample to be detected and the distance of the principal component center of the principal component analysis discriminant model of the training set, and _calculate the F calculation value F calculation of the variance homogeneity test of the donkey-hide gelatin sample to be detected according to this distance, according to design Determine the confidence interval of the homogeneity of variance test classification, compare the F _calculation with the critical _value F critical at the set significance level α, and judge whether the precision of the donkey-hide gelatin sample to be detected and the discriminant model of the principal component analysis has significant If F _calculation < F _critical , there is no significant difference between the donkey-hide gelatin sample to be detected and the donkey-hide gelatin sample of the specific manufacturer, and it is determined that the donkey-hide gelatin sample to be detected is the donkey-hide gelatin of the specific manufacturer. If F _calculation > F _critical , then If there is a significant difference between the Ejiao sample to be tested and the Ejiao sample from the specific manufacturer, it is determined that the Ejiao sample to be tested is not from the specific manufacturer.

Wherein, the sample selected in step a) has at least 20 samples.

Wherein, the sensor array includes gas sensors with strong oxidizing ability, toxic gas sensors, organic compound sensors, flammable gas sensors and aromatic compound sensors.

Wherein, the temperature of the injection needle is 80°C-150°C, and the injection speed is 1-5ml/s.

Wherein, the collection time for the sensor array to collect the sample data in the sample bottle is 60-600s.

Wherein, the stirring speed of the sample in the sample bottle is 300-900rpm.

Wherein, the temperature at which the sample generates gas in the sample bottle is 50°C-150°C.

Among them, the set significance level α is 0.05.

Compared with the prior art, the beneficial effects of the present invention are reflected in:

The method for discriminating donkey-hide gelatin using electronic nose technology provided by the present invention, through the sensor response of the electronic nose, establishes a discriminant model according to the standard normal variable of the sample, and utilizes the F test, it is objective, accurate and fast, the analysis process is simple and convenient, and the sample consumption is small , and can be extended to other fields, with very high practicability.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is the flow chart of the method for identifying donkey-hide gelatin utilizing electronic nose technology in an embodiment of the present invention;

Fig. 2 is the sample data when utilizing the electronic nose technology to identify Dong-E-E-Jiao according to an embodiment of the present invention;

Fig. 3 is the sample data that the donkey-hide gelatin sample to be detected is a counterfeit donkey-hide gelatin when the electronic nose technology is used to identify donkey-hide gelatin according to an embodiment of the present invention;

Fig. 4 is the sample data of donkey-hide gelatin samples to be detected are other brands of donkey-hide gelatin when the electronic nose technology is used to identify donkey-hide gelatin according to an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Fig. 1 is a flowchart of a method for identifying donkey-hide gelatin using electronic nose technology according to an embodiment of the present invention. As shown in Figure 1, a kind of method utilizing electronic nose technology of the present invention to differentiate donkey-hide gelatin comprises the following steps:

a) Select donkey-hide gelatin products from specific manufacturers as samples;

b) detecting each sample in the samples, the detection process is:

b1) Weigh 0.5-2.0g of the sample powder after crushing and sieving the sample, place it in a sealed sample bottle with a volume of 5-20ml, stir for 1000s-3000s, take 1-5ml of the gas in the sample bottle and inject it The needle is injected into the electronic nasal air chamber; the electronic nasal air chamber is provided with 3 chambers, and the 3 chambers are provided with a sensor array composed of a total of 18 sensors;

b2) The sensor array collects the sample data in the sample bottle; wherein, the sample data of each sample is collected 3 times and averaged; the collected sample data is the characteristic response spectrum of the sample odor;

c) dividing all sample data in the sample into calibration set data and verification set data, using the calibration set data to establish a discriminant model, and verifying the established discriminant model through the verification set data;

Wherein the method for establishing a discriminant model using the calibration set data is:

c1) Preprocessing the data of a single sample, extracting feature information and performing the following transformations:

${\mathrm{<span\; class="notranslate">\; SNVs</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}{\sqrt{\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}<span\; class="notranslate">\; ,</span>$

Among them, SNV _i is the standard normal variable of the i-th sensor response value in the electronic nose signal of a single sample; x _i is the response value of the i-th sensor in the single sample; is the average value of all sensor response values in the electronic nose signal of the single sample; p is the number of sensors in the electronic nose air chamber; i is a natural number, and its value ranges from 1 to p;

c2) Obtain the preprocessed data of all samples in the calibration set, use the SIMCA method to establish a principal component analysis discriminant model for the sample, first use the samples in the calibration set as a training set, and use the sample data matrix of the training set Carrying out principal component analysis respectively, establishing the principal component analysis discriminant model of the training set and optimizing the principal component analysis discriminant model by leave-one-out method; wherein the sample data matrix is composed of the peak values of the response values of p sensors;

d) collecting the sample data of the donkey-hide gelatin sample to be detected according to the method of step b, and preprocessing the sample data of the donkey-hide gelatin sample to be detected according to the method of step c1, substituting the sample data of the donkey-hide gelatin sample to be detected into the principal component analysis discriminant model, Obtain the principal component data of the donkey-hide gelatin sample to be detected and the distance of the principal component center of the principal component analysis discriminant model of the training set, and _calculate the F calculation value F calculation of the variance homogeneity test of the donkey-hide gelatin sample to be detected according to this distance, according to design Determine the confidence interval of the homogeneity of variance test classification, compare the F _calculation with the critical _value F critical at the set significance level α, and judge whether the precision of the donkey-hide gelatin sample to be detected and the discriminant model of the principal component analysis has significant If F _calculation < F _critical , there is no significant difference between the donkey-hide gelatin sample to be detected and the donkey-hide gelatin sample of the specific manufacturer, and it is determined that the donkey-hide gelatin sample to be detected is the donkey-hide gelatin of the specific manufacturer. If F _calculation > F _critical , then If there is a significant difference between the Ejiao sample to be tested and the Ejiao sample from the specific manufacturer, it is determined that the Ejiao sample to be tested is not from the specific manufacturer.

In one embodiment of the present invention, the sample selected in step a) has at least 20 samples.

In one embodiment of the present invention, the sensor array includes a strong oxidizing gas sensor, a toxic gas sensor, an organic compound sensor, a flammable gas sensor and an aromatic compound sensor.

In one embodiment of the present invention, the temperature of the injection needle is 80°C-150°C, and the injection speed is 1-5ml/s.

In one embodiment of the present invention, the collection time for the sensor array to collect the sample data in the sample bottle is 60-600s.

In one embodiment of the present invention, the agitation speed of the sample in the sample bottle is 300-900 rpm.

In one embodiment of the present invention, the temperature at which the sample generates gas in the sample bottle is 50°C-150°C.

In one embodiment of the present invention, the ratio of the number of samples in the calibration set to the verification set is about 2:1.

In one embodiment of the present invention, the set significance level α is 0.05.

Taking the identification of donkey-hide gelatin products (hereinafter referred to as Dong-E-E-Jiao) produced by Shandong Dong-E-E-Jiao Co., Ltd. as an example, the implementation process of the present invention will be specifically described.

Select 30 Dong-E-E-Jiao samples with batch number 120433 as samples;

Each Dong'e Ejiao sample in the sample is detected, and the detection process is:

b1) Pulverize the donkey-hide gelatin sample and sieve it to obtain the sample powder. Weigh 1.0 g of the sample and place it in a sealed sample bottle with a volume of 10 ml. Stir at a temperature of 90° C. for 1800 s at a stirring speed of 500 rpm. Take 2.0ml of the gas in the sample bottle and inject it into the air chamber of the electronic nose with an injection needle; the temperature of the injection needle is 100°C, the total volume is 5.0ml, and the injection speed is 2.0ml/s. The air chamber of the electronic nose is provided with 3 chambers, and the 3 chambers are provided with a sensor array composed of 18 sensors in total. The sensor array includes gas sensors with strong oxidizing ability, toxic gas sensors, organic compound sensors, flammable gas sensors and aromatic compound sensors. In this embodiment, the FOX4000 electronic nose of Alpha MOS Company in France is used, which is equipped with HS100 automatic sampler, air compressor, and air purifier, and some of the sensors required in the 18 sensors can be selected according to the needs, which are effective Please refer to Table 1 for the detected gases.

Table 1 Sensor array and its sensitive gas in FOX4000 electronic nose

In this embodiment, all 18 sensors are selected, and the sensor array collects the sample data in the sample bottle; wherein, the sample data of each sample is collected 3 times, and each collection time is 120s, and the average value is taken; the collected samples The data is the characteristic response spectrum of the odor of Dong-E-E-Jiao samples. The sample data of Dong-E-E-Jiao collected in this example are shown in Figure 2.

The sample data of 30 samples in this sample are divided into calibration set data and verification set data, and the response peak values of 18 sensors are used as a sample data matrix. Among them, the calibration set data includes 20 Dong-E-E-Jiao sample data, and the verification set data includes 10 Dong-E-E-Jiao sample data. A discriminant model is established by using the calibration set data, and the established discriminant model is verified by the verification set data.

Wherein the method for establishing a discriminant model using the calibration set data is:

Preprocess the data of a single sample, extract feature information and perform the following transformations:

${\mathrm{<span\; class="notranslate">\; SNVs</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}{\sqrt{\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 17</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 18</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}<span\; class="notranslate">\; ,</span>$

Among them, SNV _i is the standard normal variable of the i-th sensor data in the electronic nose signal of a single sample; x _i is the response value of the i-th sensor in this sample; is the average value of all sensor response values of the single sample signal; 18 is the number of sensors in the electronic nose air chamber; i is a natural number, and its value range is from 1 to 18;

Obtain the preprocessed data of all samples in the calibration set, and use the SIMCA method to establish a PCA discriminant analysis model for the sample. First, the samples in the calibration set are used as a training set, and the sample data matrix of each training set is respectively Perform principal component analysis, and establish a principal component analysis mathematical model of each training set; optimize the model by using calibration set samples, and the optimization content includes signal selection, principal component numbers, and preprocessing methods. When optimizing, the result of the leave-one-out method cross-validation is used as the criterion, that is, one sample is removed from the calibration set each time, and other samples are used for modeling, and the model is used to predict the removed sample to determine whether it is Dong-A Ejiao, in this way, all samples in the calibration set will be predicted once, and the total misjudgment rate is used as the criterion to optimize the modeling parameters. In this embodiment, it is finally determined that the signals on 17 sensors are selected for modeling, and the number of principal components is 6. The confidence interval of the F-test is set to 95% (that is, 1-α, α=0.05).

Collect the sample data of the donkey-hide gelatin sample to be detected according to the above-mentioned steps of collecting sample data, preprocess the sample data according to the above transformation formula, substitute the sample data of the donkey-hide gelatin sample to be detected into the principal component analysis discriminant model, and obtain the donkey-hide gelatin to be detected The principal component data of the sample and the distance of the principal component center of the principal component analysis discriminant model of the training set, and _calculate the F calculation value F calculation of the variance homogeneity test of the donkey-hide gelatin sample to be detected according to this distance, according to the variance homogeneity of setting Confidence interval for the classification of sex _test , compare the F _calculation with the critical value F critical at the set significance level α, the significance level α=0.05 in this embodiment, if _{the calculated value of} F<F _critical , then the donkey-hide gelatin to be detected If there is no significant difference between the sample and the Dong-E-E-Jiao sample, it is determined that the E-Jiao sample to be detected is Dong-E-E-Jiao. If _{the calculated value of} F > F _{is critical} , there is a significant difference between the E-Jiao sample to be detected and the Dong-E-E-Jiao sample, and it is determined to be detected Sample donkey-hide gelatin is not Dong'e donkey-hide gelatin. As shown in Figure 3, it is the sample data of a counterfeit donkey-hide gelatin; Figure 4 is the sample data of another brand of donkey-hide gelatin.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of an embodiment, and the modules or processes in the accompanying drawing are not necessarily necessary for implementing the present invention.

Those of ordinary skill in the art can understand that: the modules in the device in the embodiment may be distributed in the device in the embodiment according to the description in the embodiment, or may be changed and located in one or more devices different from the embodiment. The modules in the above embodiments can be combined into one module, and can also be further split into multiple sub-modules.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. a method utilizing electronic nose technology to differentiate donkey-hide gelatin, is characterized in that, comprises the following steps: a) Select donkey-hide gelatin products from specific manufacturers as samples; b) detecting each sample in the samples, the detection process is: b1) Weigh 0.5-2.0g of the sample powder after crushing and sieving the sample, place it in a sealed sample bottle with a volume of 5-20ml, stir for 1000s-3000s, take 1-5ml of the gas in the sample bottle and inject it The needle is injected into the electronic nasal air chamber; the electronic nasal air chamber is provided with 3 chambers, and the 3 chambers are provided with a sensor array composed of a total of 18 sensors; b2) The sensor array collects the sample data in the sample bottle; wherein, the sample data of each sample is collected 3 times and averaged; the collected sample data is the characteristic response spectrum of the sample odor; c) dividing all sample data in the sample into calibration set data and verification set data, using the calibration set data to establish a discriminant model, and verifying the established discriminant model through the verification set data; Wherein the method for establishing a discriminant model using the calibration set data is: c1) Preprocessing the data of a single sample, extracting feature information and performing the following transformations: ${\mathrm{<span\; class="notranslate">\; SNVs</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}{\sqrt{\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}<span\; class="notranslate">\; ,</span>$ Among them, SNV _i is the standard normal variable of the i-th sensor response value in the electronic nose signal of a single sample; x _i is the response value of the i-th sensor in the single sample; is the average value of all sensor response values in the electronic nose signal of the single sample; p is the number of sensors in the electronic nose air chamber; i is a natural number, and its value ranges from 1 to p; c2) Obtain the preprocessed data of all samples in the calibration set, use the SIMCA method to establish a principal component analysis discriminant model for the sample, first use the samples in the calibration set as a training set, and use the sample data matrix of the training set Carrying out principal component analysis respectively, establishing the principal component analysis discriminant model of the training set and optimizing the principal component analysis discriminant model by leave-one-out method; wherein the sample data matrix is composed of the peak values of the response values of p sensors; d) collecting the sample data of the donkey-hide gelatin sample to be detected according to the method of step b, and preprocessing the sample data of the donkey-hide gelatin sample to be detected according to the method of step c1, substituting the sample data of the donkey-hide gelatin sample to be detected into the principal component analysis discriminant model, Obtain the principal component data of the donkey-hide gelatin sample to be detected and the distance of the principal component center of the principal component analysis discriminant model of the training set, and _calculate the F calculation value F calculation of the variance homogeneity test of the donkey-hide gelatin sample to be detected according to this distance, according to design Determine the confidence interval of the homogeneity of variance test classification, compare the F _calculation with the critical _value F critical at the set significance level α, and judge whether the precision of the donkey-hide gelatin sample to be detected and the discriminant model of the principal component analysis has significant If F _calculation < F _critical , there is no significant difference between the donkey-hide gelatin sample to be detected and the donkey-hide gelatin sample of the specific manufacturer, and it is determined that the donkey-hide gelatin sample to be detected is the donkey-hide gelatin of the specific manufacturer. If F _calculation > F _critical , then If there is a significant difference between the Ejiao sample to be tested and the Ejiao sample from the specific manufacturer, it is determined that the Ejiao sample to be tested is not from the specific manufacturer. 2. the method utilizing electronic nose technology to differentiate donkey-hide gelatin according to claim 1, is characterized in that, the sample that chooses in step a) has at least 20 samples. 3. the method utilizing electronic nose technology to discriminate donkey-hide gelatin according to claim 1, is characterized in that, described sensor array comprises strong oxidizing ability gas sensor, toxic gas sensor, organic compound sensor, flammable gas sensor and aromatic compound sensor . 4. The method for identifying donkey-hide gelatin using electronic nose technology according to claim 1, characterized in that the temperature of the injection needle is 80°C-150°C, and the injection speed is 1-5ml/s. 5. The method for identifying donkey-hide gelatin using electronic nose technology according to claim 1, characterized in that the collection time for the sensor array to collect the sample data in the sample bottle is 60-600s. 6. the method for utilizing electronic nose technology to differentiate donkey-hide gelatin according to claim 1, is characterized in that, the speed that sample stirs in described sample bottle is 300-900rpm. 7. The method for identifying donkey-hide gelatin using electronic nose technology according to claim 1, characterized in that the temperature at which the sample generates gas in the sample bottle is 50°C-150°C. 8. The method for identifying donkey-hide gelatin using electronic nose technology according to claim 1, characterized in that the set significance level α is 0.05.