CN102692488A - Jinhua ham grading and identifying method based on electronic nose technology - Google Patents

Abstract

The invention discloses a Jinhua ham grading and identifying method based on an electronic nose technology. The method comprises the following steps of: putting standard samples into a sealed container, and sealing the standard samples in a thermostat at the temperature of 30 degrees for 30 minutes, wherein special grade, first grade and second grade Jinhua hams which are identified by a sensing method according to national recommended standard of GB/T19088-2008 original producing area products of Jinhua hams are taken as standard samples; pumping headspace gas into an electronic nose through a sampling channel, adsorbing a certain amount of volatile substances by a sensor so as to change the conductivity, acquiring signals by a data acquisition system, and storing the signals in a computer; and performing principal component analysis, linear discriminant analysis and partial least square analysis on the acquired data by using Unscrambler software, an WinMuster software and the like, establishing discriminant models of the hams in various grades, and predicting the samples to be tested by using the models. The method provided by the invention is convenient to operate and accurate and objective in result, and the traditional sensory grading can be replaced..

Description

Jinhua ham classification and discrimination method based on Electronic Nose Technology

Technical field

The present invention relates to a kind of Jinhua ham of different brackets, and the discrimination method of Jinhua ham and butcher's meat, more particularly, relate to a kind of different brackets Jinhua ham based on Electronic Nose Technology, and the discrimination method of Jinhua ham and butcher's meat.

Background technology

Jinhua ham is the most well-known special product in Jinhua, Zhejiang Province city and some areas, Quzhou City, and with a long history, unique flavor is described as " Zhejiang Triratna " with Longjing tea, Shaoxing rice wine.Jinhua ham starts from the Tang Dynasty, and the history in existing more than 1200 year is just classified as tribute during the Qing Dynasty so far, is the elite in the Chinese pickle cured meat product.This kind ham color of the leather Huang is bright, likeness in form Chinese lute, yellowish pink ruddy, give off a strong fragrance, nutritious, tasty, plain with color, shape " four is exhausted " is well-known, win a high admiration in the world.Only Jinhua municipalization ham manufacturing enterprise more than 100 is tame at present, and annual production reaches three, 4,000,000, and the output value surpasses 1,000,000,000 yuan, and is as the delicacies of traditional gift and family and catering industry, very popular.Jinhua ham also exports to countries and regions such as Japan, Hong Kong, Association of South-east Asian Nations simultaneously, and annual export volume has reached more than 3,000 ton.The Jinhua has obtained " township of Chinese Jinhua ham " title, and Jinhua ham was protected product by State General Administration for Quality Supervision's approval for originating in the region in 2007, becomes the 5th product (at present being called the geographical sign protection product) that obtains the protection of place of origin in Zhejiang Province.

Jinhua ham is to be that the bright pig back leg of the hybridization market pig of female parent is a raw material with the Jinhua pig and with it; Require to carry out processing and fabricating traditionally; The experience fresh ham checks and accepts, repair cut the leg base, pickle, soak leg, wash leg, shine leg, shaping, fermentation keeping, the frame classification that falls, pile up, ten multiple working procedures such as finished product, lasted 9-10 individual month.The physics and the chemical change of a series of complicacies having taken place in the ham curing process, formed a large amount of flavor substances, thereby gives Jinhua ham unique flavor characteristics.In Jinhua ham quality and ranking, fragrance always is a considerable quality index.By national standard, Jinhua ham is divided into superfine ham, one-level ham, secondary ham.The tradition stage division adopts " three label methods ", is differentiated with bamboo let by experienced ham technician.Three label down in being divided into according to difference that insert to sign the position, good sign (head label) is near knee joint; Near middle label the (two sign) inclined to one side leg dorsal part hip joint; Following label (three sign) are between vertebra and hipbone.Must bamboo let vertically be inserted ham from the ham flesh noodles when playing label and beat the label position, the degree of depth is the 1/3-1/2 of das Beinfleisch face thickness.Superfine ham requires three label all good fragrance will be arranged; The one-level ham also requires three to sign perfume, but outward appearance is not as superfine ham; Two label perfume and any one were signed all frees from extraneous odour during secondary ham requirement three was signed, and are as shown in Figure 2.The ham quality is not only relevant with processing technology thereof, and is also relevant with making season, can be divided into early winter leg, positive winter leg, early spring leg, late spring leg season by pickling, and wherein the quality with positive winter leg is the best.The anti-season leg that can produce has throughout the year also appearred now.

Except Jinhua ham, domestic more famous Yunnan Xuanwei ham, Rugao, Jiangsu ham in addition comprise many hams of inferior brand.The ham quality is to guarantee the guarantee of this millennium brand lasting prosperity of Jinhua ham, but along with the manufacturer of Jinhua ham is more and more, very different, mix the genuine with the fictitious and wait product quality problem also to get more and more.Some dealer pretend to be Jinhua ham to sell in order to reap staggering profits with the ham of other son of inferior brand, even pretend to be ham with butcher's meat.

In addition, domestic and international many scholars have done many-sided research to the fragrance forming process of Jinhua ham.Zhang Jianhao etc. through after the solarization to Jinhua ham traditional handicraft process, the trace analysis of the volatile flavor substance of different phase samples such as mesophilic digestion, high-temperature maturation, ripe undercarriage; The formation and the Changing Pattern of volatile flavor substance in the Jinhua ham process have been studied; The result shows: have 78 kinds of volatile flavor substances and detect, principal ingredient comprises aldehyde, carboxylic acid, alcohol, ester, ketone, aliphatics and aromatic hydrocarbons, sulfocompound, pyrazine etc.; The relative content of aldehyde and ketone reduces with the rising of temperature gradually in the process, and the relatively then content of carboxylic acid, ester, pyrazine and sulfocompound increases (Zhang Jianhao etc., 2004. Food Science .25 (11): 221-226) gradually.Song Huan high post application of dynamic head space sample preparation (DHS) and gas chromatography-suction is heard (GC-0) method the odor compound of Jinhua ham is analyzed; Detect 56 kinds of odor compounds; But fail to confirm that which kind of simplification compound can give characteristic perfume (Song Huanlu, the 2006. Chinese food journal .6 (1): 48-51) of Jinhua ham.Zhao Gaiming etc. are raw material with 6O hybridized pig back leg; Processing traditionally; Employing SPME (SPME) technology is separated with gas chromatograph-mass spectrometer (GCMS) (GC-MS), authenticate technology, the formation Changing Pattern of volatile flavor compound in the biceps muscle of thigh of having analyzed and researched.The result shows, compares more (Zhao Gaiming etc., 2007. food and fermentation industries .33 (5): 120-124) of carboxylic acid, branched compound, aromatics and heterocyclic compound species and content in the Jinhua ham finished product with external dry-cured ham.Zu Daohai etc. concentrate it with the Vigreux post with the volatile flavor compounds in the Jinhua ham of distillation-extraction apparatus (SDE) extraction simultaneously, and concentrate separates discriminating with GC-MS.The result shows, main meat perfume compound be (E, E)-2,4-decadienal, 2-ethyl furan and 2-amyl furan; And isolated the compound of not seeing in the research report of multiple Jinhua ham in the past; Like 3-methyl-2-(2-carbonyl propyl group)-furans, methyl amyl two sulphur, 2-octen-1-ol (Z)-(Zu Daohai etc., 2004. Food Science .25 (12): 139-142) such as 2-heptenal.

Up to now, traditional Jinhua ham standard, as in " GB/T 19088-2008 originate in region product Jinhua ham " with fragrance as important classification and discriminating project, and empirical sensory evaluation method is adopted in classification and discriminating.But there are many drawbacks in fact sensory evaluation method, and identification result is identified many artificial factor such as technician's experience, health, degree of having a sharp sense of smell to a great extent.And conventional gas analysis equipment though can measure the composition and the concentration of smell, can't be set up the relation between they and the aroma effect like gas chromatography-mass spectrum (GC-MS).

At present, also do not see the report that Electronic Nose Technology is applied to Jinhua ham classification and discriminating aspect.

Summary of the invention

The present invention is in order to overcome the deficiency of traditional Jinhua ham sense organ stage division, a kind of Jinhua ham classification and discrimination method based on Electronic Nose Technology to be provided, and this method is accurate, objective, sensitive.

The objective of the invention is to realize through following technical scheme: a kind of Jinhua ham classification and discrimination method based on Electronic Nose Technology, this method may further comprise the steps:

(1) Jinhua ham to be measured and butcher's meat sample are put into beaker, rim of a cup with the preservative film sealing, leaves standstill 30min immediately then in 30 ℃ constant temperature oven.

(2) earlier Electronic Nose is carried out initialization 30 min, with the good Jinhua ham of the sample introduction needle adsorption treatment of Electronic Nose and the head space gas of butcher's meat sample; Employed Electronic Nose is the PEN2 type portable electric nose of German Airsense company, and this electric nasus system is a kind of analytical instrument of being made up of one group of composite chemical sensor and identification software; The PEN2 Electronic Nose comprises 10 metal oxide sensor arrays, and the title and the performance specification of each sensor see the following form:

The standard transducer array of PEN2:

(3) data of the sensor array of Electronic Nose being collected are carried out pre-service earlier; Carry out principal component analysis (PCA), linear discriminate analysis and PLS analysis again; Set up a data model that detects Jinhua ham according to process result; In order to identifying the Jinhua ham of different brackets, and can realize the prediction of different brackets Jinhua ham, Jinhua ham and butcher's meat are differentiated.

The present invention has following technique effect:

1, detection method of the present invention is to adopt the Electronic Nose of being made up of 10 dissimilar sensors that Jinhua ham is carried out classification and discriminating, and its qualification result is accurate, objective, sensitive, good reproducibility.

2, the present invention does not need sample is carried out complicated pre-treatment, does not need organic solvent, and is pollution-free.

3, the present invention is simple to operate, and Measuring Time is short, is convenient to realize promoting the use on a large scale.

Description of drawings

Fig. 1 is the process flow diagram that the present invention is based on the Jinhua ham authentication method of Electronic Nose Technology;

Fig. 2 is traditional Jinhua ham graded index;

Fig. 3 is the response curve of different sealing time and temperature lower sensor;

Principal component analysis (PCA) (PCA) result that Fig. 4 differentiates for the different brackets Jinhua ham, horizontal ordinate is the first principal component variable quantity, ordinate is the Second principal component, variable quantity;

Linear discriminate analysis (LDA) result that Fig. 5 differentiates for the different brackets Jinhua ham, horizontal ordinate is linear discriminant function LD1, ordinate is linear discriminant function LD2;

Fig. 6 is partial least squares analysis (PLS) result that the different brackets Jinhua ham is differentiated, horizontal ordinate is a sample number, 10 sample numbers of each grade ham wherein, and ordinate is predicted value (reference value);

Fig. 7 is the principal component analysis (PCA) result that Jinhua ham and butcher's meat are differentiated, horizontal ordinate is the first principal component variable quantity, and ordinate is the Second principal component, variable quantity;

Fig. 8 is the linear discriminate analysis result that Jinhua ham and butcher's meat are differentiated, horizontal ordinate is linear discriminant function LD1, and ordinate is linear discriminant function LD2.

Embodiment

The present invention is based on the Jinhua ham classification and the discrimination method of Electronic Nose Technology, may further comprise the steps:

1, Jinhua ham to be measured and butcher's meat sample are put into beaker, rim of a cup with the preservative film sealing, leaves standstill 30 min immediately then in 30 ℃ constant temperature oven.

2, earlier Electronic Nose is carried out initialization 30 min, with the good Jinhua ham of the sample introduction needle adsorption treatment of Electronic Nose and the head space gas of butcher's meat sample.Employed Electronic Nose is the PEN2 type portable electric nose of German Airsense company, and this electric nasus system is a kind of analytical instrument of being made up of one group of composite chemical sensor and identification software.The PEN2 Electronic Nose comprises 10 metal oxide sensor arrays, and the title and the performance specification of each sensor are seen table 1.

  

The standard transducer array of table 1:PEN2

3, the data of the sensor array of Electronic Nose being collected are carried out pre-service earlier; Carry out principal component analysis (PCA) (Principal Component Analysis again; PCA), linear discriminate analysis (Linear Discriminant Analysis, LDA) and PLS (Partial Least Squares PLS) analyzes; Set up a data model that detects Jinhua ham according to process result; In order to identifying the Jinhua ham of different brackets, and can realize the prediction of different brackets Jinhua ham, Jinhua ham and butcher's meat are differentiated.

Through being carried out analyzing and processing, the data of collecting find that along with the prolongation of acquisition time, each signal of sensor strengthens earlier gradually, then tends towards stability.The reacting condition of each sensor output signal be the sensitivity of this sensor to a certain smell, the signal that sensor array obtains has then comprised the Global Information of this working sample.

The present invention is a standard specimen with superfine, the firsts and seconds Jinhua ham of differentiating according to national proposed standard " GB/T 19088-2008 originate in region product Jinhua ham " organoleptic method, and sample is put into airtight container, and 30 ℃ of lower seals are 30 minutes in constant temperature oven.Then head space gas is pumped in the Electronic Nose through sampling channel, sensor is because of having adsorbed a certain amount of volatile matter, and conductivity changes, and this signal is obtained and is stored in the computing machine by data acquisition system (DAS).The data that collect are carried out principal component analysis (PCA) (PCA), linear discriminate analysis (LDA) and partial least squares analysis (PLS) with Unscrambler software and WinMuster software etc.; Each rank all selects 30 values between the 78s-80s of the phase of tending to be steady to carry out PCA and LDA analyzes; One group of data choosing between 71s-80s are carried out the PLS analysis as the data of pattern-recognition; Set up the discriminating model of each grade ham, this model is used to predict testing sample.

Take the inventive method, easy to operate, the result is accurately objective, can substitute traditional sense organ classification.The present invention helps to improve the Jinhua ham standard, and is all significant for quality control in the Jinhua ham production and sales and original producton location geographical sign product protection.

Below with specific embodiment the present invention is elaborated, but be not limited to the present invention.

The Electronic Nose that the present invention utilizes is made up of gas sensor array, data acquisition system (DAS) and PRS for the PEN2 type portable electric nose of German Airsense company.Before testing, earlier Electronic Nose is carried out preheating, it is best to confirm that each parameter makes Electronic Nose measure state.

At first confirm sampling and scavenging period.Sampling time is meant in the one-time detection process, and Electronic Nose is an important parameters to the adsorption time of sample gas.Show that according to repeatedly testing each sensor response curve tends towards stability about 80s, expression has reached a stable status, therefore will the sampling time be made as 80s.Electronic Nose needs to eliminate the residual gas of last experiment with pure air cleaning sensor array after having detected a sample, each sensor is restPosed, thereby detect next time.Through test of many times, confirm that 60s is the scavenging period of Electronic Nose in this experiment.

Secondly confirm sealing time and temperature.Generation head space gas in the time of before experiment, need sample being left standstill in the container of sealing.Sealing time and temperature have directly determined the volatilization degree of sample smell, and then have determined the detection accuracy and the effect of Electronic Nose.In order to confirm an experiment condition preferably, sample is divided into two groups (concrete disposal route is seen table 2), one group is positioned in 10 ℃ the constant temperature oven, and another group is positioned in 30 ℃ the constant temperature oven.Following 4 appearance of each condition, every duplicate samples takes by weighing 10 g.Relatively two temperature results comparison (see figure 3)s show, in the time of 10 ℃ because temperature is lower, and gaseous volatilization is incomplete, the response that records low than 30 ℃.Three comparative result demonstrations standing time, the sample under the 10 min conditions is because the time is shorter, and response is lower; Sample under the 60 min conditions is compared no significant change with 30 min; The gaseous volatilization of sample was near state of saturation when 30 min were described, and along with the prolongation of standing time, fluctuation appears in response curve; So, confirm that best sealing time and temperature conditions are: leave standstill 30 min under 30 ℃ of conditions.

Table 2: sample packet processing method

The main flow process of measuring is: sample is after sealing a period of time; Its head space gas is pumped in the Electronic Nose through sampling channel; Sensor is because of having adsorbed a certain amount of volatile matter, and conductivity changes, and this signal is obtained and is stored in the computing machine by data acquisition system (DAS).After sampling was accomplished, the pure air behind activated carbon filtration was pumped to Electronic Nose, sensor is cleaned and makes it return to original state, for measurement next time ready.

Sensor acquisition to data carry out principal component analysis (PCA) (PCA), linear discriminate analysis (LDA) and partial least squares analysis (PLS) with Unscrambler software and WinMuster software etc.PCA carries out data-switching with the information of the many indexs of being extracted of sensor, replaces original multi-C vector with less major component.Experimental result is chosen preceding two major components and is analyzed, and has comprised the first principal component that in PCA conversion, obtains and the contribution rate of Second principal component, on PC1 and the PC2.Contribution rate (variable quantity) is big more, explains that principal ingredient can reflect the information of original many indexs preferably.Total contribution rate can be used above 70%～85% method.It is a kind of statistical method of type under the study sample that LDA analyzes.When LDA analyzed, the signal that has utilized all the sensors was to improve the accuracy of classification.The discriminant function of being constructed is drawn through the linearity combination by original variable, can distinguish different sample sets to greatest extent, when reducing the data space dimension, reduces information dropout to greatest extent.PLS is a kind of expansion to multiple linear regression model, is the regression modeling method of a kind of many dependent variables to many independents variable.It can carry out regression modeling under there is the condition of serious multiple correlation in independent variable, and allows under the sample point number is less than the condition of variable number, to carry out regression modeling.Under an algorithm, can realize simultaneously that regression modeling (multiple linear regression), data structure simplify the correlation analysis (canonical correlation analysis) between (principal component analysis (PCA)) and the two groups of variablees.

Jinhua ham sample used in following examples is required and tradition winter leg production technology in strict accordance with " GB/T 19088-2008 originates in region product Jinhua ham " by Jinhua Han Bang Food Co., Ltd of Jinhua ham professional production enterprise; With the bright back leg of Jinhua pig is raw material; Wash in the winter time and shine and pickle; Through 70 multiple working procedures, last and process the 9-10 month.By having more than 30 year working experience; Senior ham judge teacher with ham technician certificate differentiates by this standard, adopts empirical sensory evaluation method, is main foundation with the fragrance that distributes after extracting in the bamboo let insertion ham muscle; In conjunction with physical and chemical indexs such as lean meat ratio, moisture; Select superfine (three sign perfume), one-level (three sign perfume) and secondary 3 different brackets such as (two sign perfume, and one signs free from extraneous odour), recheck by ham quality testing center, Jinhua, Zhejiang Province technician at last.Used butcher's meat is the new greatly healthy butcher's meat of buying from the supermarket among the embodiment.

Embodiment 1: the discriminating of different brackets Jinhua ham

(1) superfine, the one-level and the Jinhua ham of secondary are cut into the fritter of 2 mm * 2 mm * 2 mm, place the 150mL beaker of cleaning, rim of a cup seals with preservative film rapidly.In 30 ℃ constant temperature oven, leave standstill 30 min.Every duplicate samples takes by weighing 10g, and each rank is done 10 parallel laboratory tests.

(2) behind Electronic Nose initialization 30 min, sample introduction needle is seen through preservative film insert in the beaker, adsorb its head space gas.It is 80s that adsorption time is set, and scavenging period is 60s.

(3) data that collect of collecting sensor array, with Unscrambler software and WinMuste software etc. carry out principal component analysis (PCA) (PCA, Fig. 4), linear discriminate analysis (LDA, Fig. 5) and partial least squares analysis (PLS, Fig. 6).

For PCA and LDA; Each rank all selects 30 values between the 78s-80s of the phase of tending to be steady to analyze as the data of pattern-recognition; Can find out from Fig. 4 and Fig. 5; The ham of each grade all has been distributed in three different zones preferably, explains that the present invention can distinguish the Jinhua ham of different brackets.

For PLS, one group of Electronic Nose data that the ham of each grade is all chosen between 71s-80s are used as modeling, and the data of selecting one group of same time period else are as prediction.Give superfine ham assignment 1 respectively, one-level ham assignment 2, secondary ham assignment 3 is as the reference value of prediction.The error threshold that allows is made as 0.3 (it is as shown in Figure 6 to predict the outcome).Visible from Fig. 6, the ham predicted value of each grade and the error of reference value all are controlled within 0.3, and it is comparatively desirable to predict the outcome, and explains that the present invention can predict the Jinhua ham of different brackets.

Visible through above embodiment, utilize Electronic Nose not only can detect the Jinhua ham of different brackets, can also realize the ham of unknown grade is carried out the target of classification in conjunction with certain analysis software.Therefore, at first Electronic Nose is carried out the pattern-recognition of training set, set up the Electronic Nose discriminating model of Jinhua ham with this.Table 3 is the recognition result to Three Estate ham training set.

Table 3: the Jinhua ham level evaluation model training collection recognition result of Electronic Nose

Can find out that from table 3 the overall discrimination of training set is 95%.Then this model of setting up is verified that to the recognition result demonstration of Three Estate ham checking collection, the overall discrimination of checking collection is 88.33%, recognition effect is better.

Embodiment 2: the discriminating of Jinhua ham and butcher's meat

(1) Jinhua ham and butcher's meat all are cut into the fritter of 2mm * 2mm * 2 mm, place clean 150mL beaker respectively, rim of a cup seals with preservative film rapidly.In 30 ℃ constant temperature oven, leave standstill 30 min.Every duplicate samples takes by weighing 10g, and each other ham of level and butcher's meat are all done 4 parallel laboratory tests.

(2) behind Electronic Nose initialization 30 min, sample introduction needle is seen through preservative film insert in the beaker, adsorb its head space gas.It is 80s that adsorption time is set, and scavenging period is 60s.

(3) data that sensor array collected with WinMuster software carry out principal component analysis (PCA) (PCA, Fig. 7) and linear discriminate analysis (LDA, Fig. 8).

For PCA and LDA; Each rank all selects 30 values between the 78s-80s of the phase of tending to be steady to analyze as the data of pattern-recognition; Visible from Fig. 7 and Fig. 8; The Jinhua ham of each grade all is distributed in the different zones with butcher's meat well, explains that this invention can differentiate Jinhua ham and bacon.

Claims (2)

1. Jinhua ham classification and discrimination method based on an Electronic Nose Technology is characterized in that this method may further comprise the steps:

(1) Jinhua ham to be measured and butcher's meat sample are put into beaker, rim of a cup with the preservative film sealing, leaves standstill 30min immediately then in 30 ℃ constant temperature oven.

2. initialization 30 min are earlier carried out to Electronic Nose in (2), with the good Jinhua ham of the sample introduction needle adsorption treatment of Electronic Nose and the head space gas of butcher's meat sample; Employed Electronic Nose is the PEN2 type portable electric nose of German Airsense company, and this electric nasus system is a kind of analytical instrument of being made up of one group of composite chemical sensor and identification software; The PEN2 Electronic Nose comprises 10 metal oxide sensor arrays, and the title and the performance specification of each sensor see the following form:

The standard transducer array of PEN2:

The array sequence number Sensor name Performance specification 1 W1C Fragrance ingredient 2 W5S Sensitivity is big, and is very sensitive to oxynitrides 3 W3C Ammoniacal liquor is sensitive to fragrance ingredient 4 W6S Mainly selective to hydrogen 5 W5C Alkane, fragrance ingredient 6 W1S Sensitive to methane 7 W1W Sensitive to sulfide 8 W2S Sensitive to ethanol 9 W2W Fragrance ingredient is sensitive to organic sulfide 10 W3S Sensitive to alkane

(3) data of the sensor array of Electronic Nose being collected are carried out pre-service earlier; Carry out principal component analysis (PCA), linear discriminate analysis and PLS analysis again; Set up a data model that detects Jinhua ham according to process result; In order to identifying the Jinhua ham of different brackets, and can realize the prediction of different brackets Jinhua ham, Jinhua ham and butcher's meat are differentiated.

Images