CN107607585A - A kind of method that vinegar semi-manufactured goods quality is monitored using electronic nose sensor combinations optimization - Google Patents

Abstract

The invention belongs to food liquid monitoring technology field, a kind of method that vinegar semi-manufactured goods quality is monitored using electronic nose sensor combinations optimization is provided, using normal vinegar as reference substance, vinegar to be measured is sample, detection identification is carried out using electronic nose sensor respectively, sensor curve figure corresponding to foundation, sensitive sensor is selected by curve map analysis, by sensitive sensor response become the steady time be chosen to be comparison judge the time, then the sensitive sensor combination of selection energy intuitive judgment carries out interpretation, response when judging time progress radar map analysis and identification is being compared to selected sensitive sensor combination.Select the effective sensor of sensitivity, numerical value change is analyzed to differentiate whether vinegar quality changes, judge whether vinegar comes into metamorphic epoch, just can carry out property determination from without carrying out the strain separating work substantial amounts of, cumbersome, the cycle is longer, foundation is provided for the subsequent treatment in product.

Description

It is a kind of to monitor vinegar semi-manufactured goods quality using electronic nose sensor combinations optimization Method

Technical field

The invention belongs to food liquid monitoring technology field, and in particular to one kind utilizes electronic nose sensor combinations optimization The method for monitoring vinegar semi-manufactured goods quality, vinegar semi-finished product refer to that vinegar finishes in the ongoing product of acetic fermentation or leaching vinegar , production on-site be stored in all kinds of containers, will enter allotment, sterilizing, filling process edible vinegar.

Background technology

In recent years, in shelf life and stagnant deposit aerogenesis, the kettle that rises, tacky iso-metamorphism occur for edible vinegar in China in the phase Phenomenon, normal operation and the consumption of the product of enterprise are seriously affected, have been the general character and bottleneck problem of enterprise development, It has its source in lacks to monitoring of the vinegar in product key point.The technical procedure of vinegar production is from raw material to product export, entirely Cycle is longer, and main fermentation process is divided into alcoholic fermentation and acetic fermentation, and semi-finished product storage process is mainly in the ageing stage.Enterprise The ageing stage is concentrated mainly on when pinpointing the problems, the stage is also the vacuum zone of production process monitoring, i.e., without detection behavior and Index request, and the quality control of fermented product focuses on process.

Electronic nose is a kind of chemical sensor array by with partial selective and appropriate PRS forms, The instrument of simple or complicated smell can be identified, is had a wide range of applications in food industry, Aero-Space, environmental monitoring etc. Prospect, this is that the field receives significant attention and the reason for most attention, is hot research problem at present in the world, in recent years Quickly grow, be widely used in industrial each department, such as tobacco, cosmetics, food industry, petrochemical industry, grain storage Deposit and process, drinks and beverage, environment monitoring, clinical diagnosis.In addition, electronic nose is additionally operable to commodity inspection, the discriminating of microorganism, medicine Classification and the department such as differentiation, aerospace, and check explosive, firearms and ammunition, drugs and other prohibited items etc..It is external Electronic Nose Technology production link has been applied to successfully at present.The technology can be used for the making vinegars such as Shanxi mature vinegar, mature vinegar In production process, also can popularization and application other provinces and area making vinegar production in.

The fermentation process of vinegar is a complicated microbial metabolism, the use of Daqu is diastatic fermentation in production process Agent, and Daqu is the set of numerous kinds microorganism, the metabolite species and mechanism of every kind of microorganism so far still can not be clear and definite, Some bacterial strains are even difficult to pure culture, or are difficult to reproduce metabolic process.16SPCR gene surveys such as are carried out to tacky vinegar Sequence, shown in obtained result figure 1.

Therefrom it can be seen that advantageous bacterial strain, i.e. CY values are 62480 entitled Lactobacillus Hokkaidonensis bacterial strain, and this bacterial strain inhereditary feature after isolation and unstable, after normal vinegar is infected with it The tacky phenomenon of identical is not shown.The result figure 2 being sequenced by fungi ITS.As can be seen from Figure 2 there are a variety of advantages Bacterial strain, and these dominant strains can not be separated, therefore cannot also carry out checking test.It is existing that micro-image can also prove this As.Micrograph is shown in Fig. 3, in figure：A shows that rotten vinegar spreads all over organisms in field of microscope, there is difference from profile Species, but based on the dominant strain being dominant with a kind of quantity；B shows two kinds of thalline that Gram's staining shows, i.e. gram Positive bacteria and Gram-negative bacteria；C shows the diversity of microbial cells in rotten vinegar.Carried out using rotten vinegar as sample micro- Bio-separation, the bacterium bacterial strain of 1-2 kind amounts can be only obtained, similar Fungiform bacterial strain is not separated, from strain isolation Angle also illustrate to obtain the very difficult of pure bacterial strain.D shows one plant of isolated bacterium micro-image.

Although the difficulty for isolating pure bacterial strain is larger, it is the fact that all kinds of bacterial strains participate in later stage fermentation in vinegar, Since there is the phenomenon of fermentating metabolism, then the generation of metabolite is necessarily had, and then the smell of rotten vinegar can be made with correlation Feature.

CN201310706197 inventions provide a kind of preparation method for the gas sensor array for detecting Shanxi mature vinegar, Including step：（1）With vacuum coating equipment respectively by nylon 6, polyacrylamide, polystyrene, polyacrylonitrile, five kinds of polyvinyl chloride Polymer is vaporized on five different quartz substrates；（2）By the quartz substrate being deposited welding on circuit boards；Using this hair Gas sensor array size prepared by bright method is small, and cost is low, reproducible, energy long-term stable operation, co-existing gases production Raw influence is small, high to the resolution performance of mixed gas.And with it with reference to made of piezoelectric system electric nasus system, detection mountain Sample cumbersome pretreatment process is not needed during the western mature vinegar true and false, can directly be carried out, and fast response time, the degree of accuracy are high.

A kind of method of non-demolition identification Watermelon Maturity of CN201010154584 disclosure of the invention：By electronic nose to The smell that the ripe watermelon and raw watermelon known volatilize is sampled, using the PCA analysis modules of electronic nose to known ripe watermelon Analyzed with the sampled data of raw watermelon, raw watermelon standard analysis template and ripe watermelon standard analysis are established according to analysis result Template.The smell volatilized by electronic nose to unknown watermelon samples, and is utilized respectively the Euclidean distance analysis mould of electronic nose The hits of block, mahalanobis distance analysis module, correlation analysis analysis module and difference decision analysis module to unknown watermelon According to being analyzed, analysis result and the template established contrast are identified into unknown watermelon, and combine this four analysis modules Qualification result draws final qualification result, i.e. has at least three identifications to be unanimously just considered to unknown watermelon in four analysis modules Qualification result, therefore the watermelon authentication method of the present invention is realized without destroying identification and the identification degree of accuracy is high.

The technology neck that a kind of vinegars of CN201510558300.6, the classification discrimination method category food liquid classification of soy sauce differentiate Domain, there is provided a kind of vinegar, the classification discrimination method of soy sauce, established in terms of electronic nose standard form database and visual templates two The uniqueness of standardized product is identified and judged, foundes vinegar and soy sauce each between brand and production technology（Brewage or prepare）'s Identify identification, so as to stable prod inherent quality, distinguish personation and it is of poor quality.Detected and identified using electronic nose, establish standard Sample Electron nasil feels template, using micro- sem observation, establishes the visual templates of standard sample, then according to standard sample electronics Nasil feels that the visual templates of template and standard sample carry out analysis mirror to the electronic nose smell template and visual templates of testing sample Not.Discrimination increases, and carrying out identification for different classes of sample lays a good foundation, and not only can such as brewage food to the classification of vinegar Vinegar, preparation of vinegar, the brand of different vinegars can also make a distinction outside making a distinction to the classification of soy sauce.

A kind of analysis methods for differentiating brown rice tea flavouring essence quality of CN201510967876.8, there is provided one kind differentiates brown rice tea perfume The analysis method of gas quality, belong to technical field of analysis and detection, including a brown rice stir-frying-technology step；One utilizes electronic nose The step of flavouring essence quality for the brown rice tea that different stir-frying-technologies brew, is quickly distinguished and differentiated to technology；One utilizes GC-MS technologies The step of analyzing brown rice tea fragrance component；One carries out smelling using subjective appreciation method to brown rice tea, and to the sense of brown rice tea Official's attribute carries out the step of smelling describes.Electronic nose analysis is mainly using principal component analysis and radar finger-print point in the present invention The flavouring essence quality for analysing the brown rice tea brewed to different stir-frying-technologies differentiates, using GC-MS technologies to the preferable brown rice of quality The fragrance component of tea is measured.The inventive method has that objective result, reproducible, high sensitivity, analyze speed be fast, operation Simply, the advantages that economical and efficient.

CN201510537111.0 discloses discriminating tea using the method for electronic nose quick discriminating tealeaves radiation absorber amount Whether leaf irradiates and the method for radiation absorber amount and a kind of new application of electronic nose, discrimination method include：Using known difference The Tea Samples of dose delivery processing；The Tea Samples of radiation absorber amount known to electronic nose detection；Electronic nose is drawn to tealeaves Volatile materials response diagram；It is determined that differentiate Selecting time；Calculate the variance contribution of each principal component electronic nose sensor characteristic values Rate and the accuracy for returning initial packet case differentiation, preferably effective sensor；The classification of different irradiation doses is calculated Function coefficients, obtain differentiating the classification function of Tea Samples irradiation dose；Unknown Tea Samples are detected with electronic nose, substitute into classification In function, the irradiation dose of tealeaves is obtained.The more existing method for differentiating tealeaves of the discrimination method has that method is simple, step It is few, quickly, the features such as identification efficiency is high, and the degree of accuracy is high, suitable for differentiating that whether tealeaves irradiates or tealeaves radiation absorber amount.

The method that CN200810152157.0 detects shrimp freshness using electronic nose discloses one kind using electronic nose detection The method of shrimp freshness, it is desirable to provide it is a kind of fast and convenient, it is easy to use, shadow of the storage requirement to testing result can be reduced Ring, the objective reliable method using electronic nose detection shrimp freshness of measurement result.Comprise the steps：By shrimp sample to be detected It is put into sealing container, water bath with thermostatic control heating；The gas of flow speed stability is passed through in test cabinet, the shrimp sample after heating is put into In test cabinet, air-flow is carried the smell that sample gives out and enter by the array being made up of 5 different types of gas sensors Row detection；The data of gas sensor array collection are collected, data processing are carried out to the data of collection, according to data processed result Evaluate shrimp freshness.The present invention detection method use by 5 different types of sensor groups into electronic nose shrimp samples enter Row detection, testing result is objective, accurate, quick, reproducible.

From existing disclosed document report, it can be seen that to authenticity of products, category classification etc., electronic nose plays Great function.But the electronic nose of existing literature report does not have to sensor progress various combination progress properties of samples judgement Narration.

The content of the invention

It is an object of the invention to provide one kind to utilize electronic nose sensor combinations optimization monitoring vinegar semi-manufactured goods quality Method, i.e., deposit this critical period of phase vinegar semi-finished product are stagnant, by this advanced instrument of applying electronic nose, aid in special reality Material and method are tested, detects key point parameter, the information whether product is stablized is obtained, offer section is moved towards to the next step of semi-finished product Learn foundation.

The present invention adopts the following technical scheme that realization：One kind utilizes electronic nose sensor combinations optimization monitoring vinegar half The method of final product quality, using normal vinegar as reference substance, vinegar to be measured is sample, is detected respectively using electronic nose sensor Identification, sensor curve figure corresponding to foundation, by the sensitiveness of sensor curve map analysis sensor, select sensitive sensor Device, become the steady time by sensor curve map analysis sensitive sensor response, become steady time of sensitive sensor response is selected It is set to comparison and judges the time, the then sensitive sensor combination of selection energy intuitive judgment carries out interpretation, i.e., to selected sensitive biography Sensor combination is comparing the progress radar map analysis and identification of response when judging the time.

The sensor curve figure is the measure of sample using the ratio of response initial resistance and last resistance as ordinate Time is abscissa.First cleaned during practical measurement, at this moment show a resistance value, as initial resistivity value, then sample introduction Measure, the value of each second and the ratio of initial value are ordinate.Sample can gather the semi-finished product stored in tank, detect in tank The semi-finished product of storage can shift to an earlier date control point, be advantageous to take measures before filling the container, stabilized product quality.

Specific detecting step is as follows：

（1）Sample pre-treatments：Choose normal specimens and each 0.1-1.0ml of testing sample is moved into ml headspace bottle, screw bottle cap immediately Room temperature places 0.5-1.0 hours；

（2）The sample feeding handled well is determined by electronic nose conventional practices using German PEN3 electronic noses instrument system, Sample is entered into line sensor response signal intensive analysis, sensor curve figure corresponding to foundation；

（3）According to step（2）When the sensitiveness and sensor response of the sensor curve map analysis sensor of acquisition become steady Between, sensitive sensor is selected, selected sensitive sensor response becomes the steady time to compare the judgement time, reads normal specimens with treating The sensitive sensor of test sample product judges the response of time comparing, and passes through the exhausted of the response value difference of normal specimens and testing sample To being worth analysis and identification testing sample；

（4）The sensitive sensor combination of selection energy intuitive judgment carries out interpretation：The single sensitivity of time is judged selected comparison The response of sensor carries out radar map analysis and identification, or carries out radar map analysis mirror to the sensitive sensor response of combination Not；

（5）The sound of the ratio between the response of 2 sensitive sensors by normal specimens 2 sensitive sensors corresponding with testing sample It the ratio between should be worth and to be compared, determine product critical modification value, analyze testing sample accordingly.

The present invention is tested using German PEN3 electronic noses equipment.The equipment has 10 sensors, in comparison process, Sensor is accepted or rejected, the sensor of identity difference is removed, makes sample itself characteristic easy to identify.Normal specimens and testing sample Sampling amount according to the size of ml headspace bottle select sampling amount.Using normal vinegar and vinegar to be measured as sample, using to smell very Sensitive electronic nose is measured, and is identified and compared by the sensor of electronic nose, selects the effective sensor of sensitivity, point Analysis numerical value change judges whether vinegar comes into metamorphic epoch, from without carrying out largely to differentiate whether vinegar quality changes , the strain separating that cumbersome, the cycle is longer work just can carry out property determination, provide foundation for the subsequent treatment in product.

Brief description of the drawings

Fig. 1 is to carry out 16SPCR gene sequencing result figures to tacky vinegar；Fig. 2 be CY values be 62480 it is entitled Lactobacillus hokkaidonensis bacterial strain ITS sequencing result figures；Fig. 3 is the micrograph of microorganism in rotten vinegar Picture；Fig. 4 is normal specimens and the sensor curve figure for becoming quality sample, and A be normal specimens in figure, and B is change quality sample；Fig. 5 is just Normal sample and when becoming 75 seconds of quality sample sensitive sensor response radar map, A be normal specimens in figure, and B is change quality sample； Fig. 6 is normal specimens and becomes response of the quality sample at 75 seconds, and A is normal specimens in figure, and B is change quality sample；Fig. 7 is normal Sample and become radar map of the quality sample in 1,2,6,7,8, No. 9 sensor, A be normal specimens in figure, and B is change quality sample；Fig. 8 is 1st, 2,6,7, No. 9 sensor combinations to normal specimens and become the radar map of quality sample, and A be normal specimens in figure, and B is the sample that goes bad Product；Fig. 9 is 2,6,7,8, No. 9 sensor combinations to normal specimens and becomes the radar map of quality sample, and A is normal specimens in figure, and B is Become quality sample；Figure 10 is 2,6,7, No. 9 sensor combinations to normal specimens and becomes the radar map of quality sample, and A is normal sample in figure Product, B are change quality sample；Figure 11 is 2,6,7, No. 8 sensor combinations to normal specimens and becomes the radar map of quality sample, and A is in figure Normal specimens, B are change quality sample；Figure 12 is detects the normal specimens of East Lake vinegar and becomes the sensor curve figure of quality sample, A in figure For normal specimens, B is change quality sample；Figure 13 is East Lake vinegar normal specimens and becomes response of the quality sample at 75 seconds, and A is in figure Normal specimens, B are change quality sample；Figure 14 is each sensor to East Lake vinegar normal specimens and becomes the radar map of quality sample, A in figure For normal specimens, B is change quality sample.

Embodiment

It is a kind of using electronic nose sensor combinations optimization monitor vinegar semi-manufactured goods quality method, using normal vinegar as pair According to product, vinegar to be measured is sample, and detection identification is carried out respectively using electronic nose sensor, sensor curve figure corresponding to foundation, By the sensitiveness of sensor curve map analysis sensor, sensitive sensor is selected, is passed by the way that sensor curve map analysis is sensitive Sensor response becomes the steady time, is chosen to be comparison and judges the time sensitive sensor response steady time of becoming, then selection can be straight See the sensitive sensor combination judged and carry out interpretation, i.e., response when judging the time is being compared to selected sensitive sensor combination Value carries out radar map analysis and identification.Sensor curve figure be using the ratio of response initial resistance and last resistance as ordinate, The minute of sample is abscissa.First cleaned during practical measurement, at this moment show a resistance value, as initial resistance It is worth, then sample introduction determines, and the value of each second and the ratio of initial value are ordinate.

Comprise the following steps that：

（1）Sample pre-treatments：Choose normal specimens and become quality sample and moved into as each 0.1-1.0ml of testing sample in ml headspace bottle, Specific sampling amount selectes sampling amount according to the size of ml headspace bottle, screws bottle cap room temperature immediately and places 0.5-1.0 hours；Normal sample Product and rotten sample collection finish from certain vinegar factory leaching vinegar, are stored in all kinds of containers in production on-site, will enter allotment, go out Bacterium, the semi-finished product vinegar of filling process.

（2）The sample feeding handled well is surveyed by electronic nose conventional practices using German PEN3 electronic noses instrument system It is fixed, sample is entered into line sensor response signal intensive analysis, sensor curve figure corresponding to foundation and becomes quality sample at normal specimens Sensor curve figure see Fig. 4；

（3）According to step（2）When the sensitiveness and sensor response of the sensor curve map analysis sensor of acquisition become steady Between, sensitive sensor is selected, selected sensitive sensor response becomes the steady time to compare the judgement time, reads normal specimens with treating The sensitive sensor of test sample product judges the response of time comparing, and passes through the exhausted of the response value difference of normal specimens and testing sample To being worth analysis and identification testing sample；

From sensor values curve map analysis, 1,2,6,7,8, No. 9 sensor sensing, and numerical value is larger, for normal specimens, 10 sensor responses steady time of becoming is forward, about after 10 seconds；For abnormal sample, 10 sensor responses are 10 Also have repeatedly after second, until numerical value change just tends towards stability after about 50 seconds.Therefore it is selected to compare the time judged in 60- Between 90 seconds, this experimental selection judges the time in 75 seconds to compare.

（4）The sensitive sensor combination of selection energy intuitive judgment carries out interpretation：The single of time is judged to selected comparison The response of sensitive sensor carries out radar map analysis and identification, or carries out radar map point to the sensitive sensor response of combination Analysis differentiates.

Sensitive sensor 1, the response of 2,6,7,8, No. 9 when judging that the time is 75 seconds to comparing carry out radar map point Analysis, the radar map of making are shown in Fig. 5；Can intuitively find out that the response of No. 6 sensors dramatically increases from figure, thus form 6, Shaded area of No. 7 sensors in radar map changes greatly.

Fig. 6 is normal specimens and becomes response of the quality sample at 75 seconds, makees table according to response, takes each sensor pair Answer the absolute value of two sample response value difference values to change to observe 10 sensor values, the results are shown in Table 1.

The sensor values of table 1 change table

Sensor number	Normal specimens	Abnormal sample	The absolute value of difference
				1	3.864	9.737	5.873
2	90.025	62.032	27.993
				3	1.815	2.686	0.871
4	1.206	1.795	0.589
				5	2.262	3.647	1.385
6	22.548	72.135	49.587
				7	80.528	67.366	13.162
8	12.439	31.208	18.769
				9	75.742	69.359	6.383
10	1.532	2.233	0.701

From table from the point of view of data, most obvious change is No. 6 sensors, and normal specimens are 22.548, and abnormal sample reaches 72.135, poor absolute value is 49.587；The changing value of sensor 3,4,5,10 is less than 5；1st, 2,6,7,8, No. 9 sensor changes Value is between 5-50.

Sample classification is different or vinegar brand is different, and its numerical value is different, but rule is similar, is exactly 1,2,6,7,8, No. 9 biography Sensor is sensitive, and the response of defective in quality No. 6 sensors of vinegar can all increase.From the rich of side illustration abnormal sample smell Degree is big compared with normal sample.Conventional commercial East Lake vinegar is also have detected simultaneously, its its normal product and deteriorating itemses detection sensor Curve map is shown in Figure 12, and response of each sensor in 75s is shown in Figure 13, and radar map of each sensor is shown in Figure 14 corresponding to it

（5）The sound of the ratio between the response of 2 sensitive sensors by normal specimens 2 sensitive sensors corresponding with testing sample It the ratio between should be worth and to be compared, analyze testing sample.

In table 1, for normal specimens, the ratio between No. 6, No. 7 sensor values 22.548/80.528=28%；For abnormal sample The ratio between product, No. 6, No. 7 sensor values 72.135/67.366=107%.Ratio can be understood as belonging to normal specimens less than 28%, Belong to abnormal sample higher than 28%, numerical value is higher, and the problem of showing sample is more serious.Sensed in the present embodiment using 6 and No. 72 Device is adjacent, and the amplitude of fluctuation of No. 6 is big, it is easily observed that change, relatively more directly perceived.For sample DH, its normal specimens 6, No. 7 biographies The ratio between sensor numerical value 31.851/87.550=36%；For abnormal sample, the ratio between No. 6, No. 7 sensor values 128.279/ 88.023=146%。

According to above-mentioned rule, realization of the invention can be：

1. in the case where sensor does not make trade-offs, unknown sample is determined according to a conventional method（The semi-finished product stored in tank）.Measure Afterwards referring initially to curve and radar map, if close to Fig. 1, Fig. 2 form, and the ratio between No. 6, No. 7 sensor values are between 15%-50% （This numerical value be certain brand product repeatedly measure after decision content or critical value, different brands and product be possible to change）, then table Sample product belong to normal, are otherwise exception or problem sample.

2. due to 1,2,6,7,8, No. 9 sensor sensing, thus be excluded that remaining 3,4,5, No. 10 sensor, is only passed with 6 Sensor carries out interpretation, and result of this method on radar map is more than you know.Such as Fig. 7.

As seen from Figure 7, normal specimens are in the polarity in 3 directions in 6 sensor identification systems, and abnormal sample There is the 4th pole, i.e. No. 6 sensors.So first can tentatively be judged with radar map by this method, then carry out No. 6, No. 7 ratio in judgement.

3. the combination that can similarly carry out 1,2,6,7,8, No. 9 sensor difference number carries out interpretation.As 5 combinations are sentenced Read, 4 combination interpretations, 3 combination interpretations.

For same sample, the combination of different sensors, the figure shown differs, several sides of this in this experiment Formula can more clearly judge difference.For the product of different brands, it is necessary to select sensitive sensor in advance, reselection is easy In the sensor combinations form of intuitive judgment.

This method was differentiated by the methods of Euclidean distance, mahalanobis distance, correlation analysis, but used radar Figure, the leading interpretation of response curve, then carry out 6,7 sensor ratio in judgement.Method is directly perceived, easy, quick, is more beneficial for giving birth to The control of production process.

It is to embody the referential data of the critical point of sample quality, can also select other to carry out sensor ratio in judgement Sensor, such as more inert sensor 3,4,5,10, the ratio of so fine or not two samples is compared, although difference is bigger, The change for being the shaded area that measure is formed can not be matching, therefore testing result is not directly perceived, thus selects testing result Intuitively 6,7 sensor.Other sensitive sensors can also do ratio, it is same because shaded area change intuitive not as Option, but combination display change can be selected.

, can also be to micro- life using its principle in this method practical operation in addition to it can carry out production process control to vinegar Thing yeast making process, wine fermentation process, soy sauce brewing etc. carry out process control, avoid quality fluctuation and produce the product of inferior quality, side Method and step are the same as foregoing.

Claims (2)

1. A kind of 1. method that vinegar semi-manufactured goods quality is monitored using electronic nose sensor combinations optimization, it is characterised in that：With just Often feeding vinegar is reference substance, and vinegar to be measured is sample, and detection identification is carried out respectively using electronic nose sensor, establishes corresponding sense Device curve map, by the sensitiveness of sensor curve map analysis sensor, sensitive sensor is selected, pass through sensor curve figure point Analysis sensitive sensor response become the steady time, by sensitive sensor response become the steady time be chosen to be comparison judge the time, then The sensitive sensor combination of selection energy intuitive judgment carries out interpretation, i.e., judges the time in comparison to selected sensitive sensor combination When response carry out radar map analysis and identification.
2. A kind of 2. side that vinegar semi-manufactured goods quality is monitored using electronic nose sensor combinations optimization according to claim 1 Method, it is characterised in that：Comprise the following steps that：

   （1）Sample pre-treatments：Choose normal specimens and each 0.1-1.0ml of testing sample is moved into ml headspace bottle, screw bottle cap immediately Room temperature places 0.5-1.0 hours；

   （2）The sample feeding handled well is determined by electronic nose conventional practices using German PEN3 electronic noses instrument system, Sample is entered into line sensor response signal intensive analysis, sensor curve figure corresponding to foundation；

   （3）According to step（2）When the sensitiveness and sensor response of the sensor curve map analysis sensor of acquisition become steady Between, sensitive sensor is selected, selected sensitive sensor response becomes the steady time to compare the judgement time, reads normal specimens with treating The sensitive sensor of test sample product judges the response of time comparing, and passes through the exhausted of the response value difference of normal specimens and testing sample To being worth analysis and identification testing sample；

   （4）The sensitive sensor combination of selection energy intuitive judgment carries out interpretation：The single sensitivity of time is judged selected comparison The response of sensor carries out radar map analysis and identification, or carries out radar map analysis mirror to the sensitive sensor response of combination Not；

   （5）The sound of the ratio between the response of 2 sensitive sensors by normal specimens 2 sensitive sensors corresponding with testing sample It the ratio between should be worth and to be compared, determine product critical modification value, analyze testing sample accordingly.