CN103592256A - Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform - Google Patents
Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform Download PDFInfo
- Publication number
- CN103592256A CN103592256A CN201310629904.6A CN201310629904A CN103592256A CN 103592256 A CN103592256 A CN 103592256A CN 201310629904 A CN201310629904 A CN 201310629904A CN 103592256 A CN103592256 A CN 103592256A
- Authority
- CN
- China
- Prior art keywords
- oil
- sample
- spectrum
- unknown
- spectrogram
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a mid-infrared spectroscopic method for rapidly distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform. The method comprises the following steps: taking a known normal edible vegetable oil sample and a known refined hogwash oil sample, scanning the samples in a spectral wavelength range of 4000 to 450/cm, acquiring spectrums, subjecting the spectrums to spectral pretreatment and establishing PLS-DA analysis model standards for distinguishing normal edible vegetable oil from refined hogwash oil on the basis of combination of a partial least square discrimination method; and then scanning an unknown oil sample, acquiring the spectrum of the unknown oil sample, subjecting the spectrum of the unknown oil sample to same spectral pretreatment, carrying out analysis by using the partial least square discrimination method and comparing analysis results with the analysis model standards so as to determine whether the unknown oil sample is normal edible vegetable oil or refined hogwash oil. The method provided by the invention is easy and convenient to operate; with the method, suspicious oil samples can be rapidly screened; and the method is applicable to on-site screening of hogwash oil during production and circulation of edible oil.
Description
Technical field
The present invention relates to a kind of detection method of edible oil, relate in particular to a kind of method based on the Fourier transform middle infrared spectrum normal edible vegetable oil of quick identification and refining hogwash fat.
Background technology
Hogwash fat also claims waste oil or swill oil, refers to the greasy floating thing in the leftovers of hotel, restaurant, leftovers (common name swill) or sewer is collected through grease trap, then adds sulfuric acid, the grease that thermal dehydration, de-slag, decolouring produce.Because the chemical reactions such as hydrolysis, oxidation, condensation have occurred hogwash fat grease in process, the poisonous and harmful substances such as benzene, pyrene, naphthalene, anthracene, nitrate and nitrite have been produced, even some strong carcinogen.Edible hogwash fat there will be the toxicity symptoms such as dizziness, headache, nausea,vomiting,diarrhea at first, long-term edible the lighter can make people lack nutrition, accelerate aging, severe one will cause enteron aisle and angiocardiopathy, destroy alimentary canal mucous membrane, and internal organ are badly damaged even carcinogenic.But the raw material that contains the chemical articles such as a large amount of production biodiesel, biological demulsifying agent, commercial grease, surfactant in hogwash fat, can develop at industrial hogwash fat, turns waste into wealth.
In recent years, hogwash fat backflow dining table problem is the focus of vast media and public attention.Because waste grease preparation is simple, with low cost, some lawless persons are in order to reap staggering profits, in the dark using it directly as edible oil or added to and sell in other edible vegetable oils and use, bring huge potential safety hazard to the life and health of numerous people.In prior art, the detection method of hogwash fat has been carried out to a large amount of research, as the application number patent of invention that is 201010177232.6, reported that a kind of rapid screening goes out to contain the method for the edible oil of hogwash fat, first with conductivity measuring instrument, directly measure qualified edible oil, be mixed with the conductivity of edible oil and the hogwash fat of hogwash fat, according to measuring the data obtained, set up examination standard, measure again the direct conductivity of edible oil to be measured, and will measure institute's value and the comparison of examination standard, thereby in judgement edible oil, whether be mixed with hogwash fat, and further set up following standard: the conductivity in the time of 25 ℃ is greater than in 100pS/m preliminary judgement edible oil and is mixed with hogwash fat.In < < Food Science > > the 28th volume o. 11th of and for example publishing for 2007 " application conductivity detects the research of a hogwash fat method " literary composition, researcher is by oil sample dissolution with solvents, add redistilled water through mix and blend layering, water is carried out to conductance measurement, getting under oil mass equal conditions, the conductivity of edible oil is at 7.47 μ S/cm, minimum only has 3.58 μ S/cm, and the conductivity of hogwash fat is the highest at 18.55 μ S/cm, the minimum 12.23 μ S/cm that also reach.This method be owing to measuring water, before namely measuring, must grease to be measured be dissolved, then with redistilled water or deionized water, be extracted with solvent, operates loaded down with trivial detailsly, is unfavorable for on-the-spotly differentiating fast.Also have scholar to propose to apply vapor-phase chromatography and differentiate, this method is to utilize in hogwash fat, to have multiple fatty acid spectrogram feature and have obvious difference with the sour spectrogram of single fat, thereby tells in edible oil, whether to contain hogwash fat.This method is because will use this valuable checkout equipment of gas chromatograph, and analytic process is more complicated also, and can not get rid of the impact of the compound lard that mediation wet goods is qualified, and therefore this method does not still have actual application value.
Along with the raising of hogwash fat refining techniques, a lot of simple physical and chemical indexs can only be differentiated and detect some mao of hogwash fat, rough hogwash fat, but cannot meet the requirement that refining hogwash fat is differentiated.So-called refining hogwash fat (RHOs) refers to that the outward appearance, part physical and chemical index and the common edible vegetable oil that after the operation processing such as further decoloration and deodorization are processed, obtain are without the compound lard of obviously distinguishing.This quasi-grease is because process through processing technologys such as meticulousr decolouring, deodorization, alkalization, the mensuration of conventional physical and chemical index, not clearly with normal edible vegetable oil difference, particularly part acid value index has approached normal plants oil, conductivity indices difference is not obvious, refrigeration test result shows also only have part can be solid-state, therefore, when carrying out the discriminating of such compound lard, difficulty is larger, method to the diagnostic test report of hogwash fat is numerous, but goes back up to now the very perfect and effective method of neither one.Therefore, the method that finds fast effective diagnostic test refining hogwash fat and edible vegetable oil to mix hogwash fat is current problem in the urgent need to address, this mixes pseudo-edible oil to being strictly on guard against, to guaranteeing the healthy of the people, the supervision and management dynamics that increases relevant department is also extremely important.
Summary of the invention
In view of this, the object of the present invention is to provide the method based on the normal edible vegetable oil of Fourier transform mid-infrared light spectrum discrimination and refining hogwash fat, the method is effective diagnostic test refining hogwash fat fast, and correct decision rate can reach 100%; Two of object of the present invention is to provide a kind of method based on Fourier transform mid-infrared light spectrum discrimination rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat, and the method is effective fast, and whole correct decision rate is 100%.
Method based on the normal edible oil of Fourier transform mid-infrared light spectrum discrimination and refining hogwash fat, comprises the step of carrying out as follows:
(1) establishment of standard
Get normal edible oil and refining hogwash fat is sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, sample is scanned, collected specimens spectrum spectrogram, described sample spectra spectrogram is carried out to spectrum pre-service and obtain pre-processed spectrum spectrogram, described pre-processed spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, sets up the PLS-DA analytical model standard of distinguishing normal edible oil and refining hogwash fat;
(2) evaluation to unknown oil sample
Get unknown oil sample sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, unknown oil sample sample is scanned, gather unknown oil sample sample spectra spectrogram, described unknown oil sample sample spectra spectrogram is carried out to spectrum pre-service and obtain unknown oil sample sample pretreatment spectrum spectrogram, described unknown oil sample sample pretreatment spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, with the analytical model Comparison of standards of setting up in step (1), determine that unknown oil sample sample is normal edible oil or refining hogwash fat.
Further, described method, spectrum pre-service described in described step (1) and step (2) all selects Savitzky-Golay convolution smoothing method to carry out spectrum pre-service, and the parameter that described Savitzky-Golay convolution smoothing method adopts is: second derivative, 5 are level and smooth, multinomial series is 2.
Further, described method, in described step (1) and step (2), in the analysis of described offset minimum binary diagnostic method, best main cause subnumber is 7.
Further, described method, in described step (1) and step (2), in the analysis of described offset minimum binary diagnostic method, selecting the first two main cause subnumber PC1 is that X-axis, PC2 are that Y-axis is set up linear relationship analysis.
Further, described method, keeps room temperature 25 ℃ ± 1 in described scanning process, controls indoor relative humidity at 20%-50%.
Further, described method, adopts coating method to analyze sample, and liquid oil sample is spread upon and on salt sheet, makes liquid film and analyze, and described salt sheet is KBr.
Method based on Fourier transform mid-infrared light spectrum discrimination rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat, comprises the step of carrying out as follows:
(1) establishment of standard
Getting rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat is sample, utilizes Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, sample is scanned, collected specimens spectrum spectrogram, described sample spectra spectrogram is carried out to spectrum pre-service and obtain pre-processed spectrum spectrogram, described pre-processed spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, sets up the PLS-DA analytical model standard of distinguishing rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat;
(2) evaluation to unknown oil sample
Get unknown oil sample sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, unknown oil sample sample is scanned, gather unknown oil sample sample spectra spectrogram, described unknown oil sample sample spectra spectrogram is carried out to spectrum pre-service and obtain unknown oil sample sample pretreatment spectrum spectrogram, described unknown oil sample sample pretreatment spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, with the analytical model Comparison of standards of setting up in step (1), determine the kind of unknown oil sample sample.
Beneficial effect of the present invention: 1) method of the present invention is selected full spectral range 4000～450cm
– 1interior middle infrared spectrum carries out modeling analysis, and does not need to select characteristic wave bands or characteristic peak, simple, convenient; 2) method of the present invention is effective fast, and whole correct decision rate reaches 100%, uses the method energy rapid screening to go out suspicious oil sample, is highly suitable for edible fat production, intermediate links enforcement hogwash fat field screening.
Accompanying drawing explanation
Fig. 1 is the FT-MIR spectrogram of 29 calibration set samples through Savitzky-Golay processing, and wherein r1-r29 represents sample sequence number, and r1-r7 represents different RHOs, r8-r14 rapeseed oil, r15-r19 soybean oil, r20-r24 peanut oil, r25-r29 corn oil.
Fig. 2 is the dispersion point diagram of the PC1/PC2 of 29 modeling samples, wherein refining hogwash fat sample (1-7) concentrates on the middle region of left, rapeseed oil sample (8-14) concentrates on a region, below, soybean oil sample (15-19) concentrates on a right region, peanut oil sample (20-24) concentrates on a region, top, and corn oil sample (25-29) concentrates on Yi Ge region, upper right side.
Fig. 3 is the cluster analysis result figure of different normal edible vegetable oil and refining hogwash fat.
Fig. 4 is the PLS-DA model prediction result figure of 22 unknown samples, and in figure, the value of giving 0,1,2,3,4 represents respectively RHOs, rapeseed oil, soybean oil, peanut oil and corn oil.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described.Illustrated embodiment is in order better content of the present invention to be described, but is not that content of the present invention only limits to illustrated embodiment.So those of ordinary skill in the art carry out nonessential improvement and adjustment according to foregoing invention content to embodiment, still belong to protection scope of the present invention.
The experimental technique of unreceipted actual conditions in illustrated embodiment of the present invention, carries out according to normal condition conventionally, or the condition of advising according to manufacturer.In illustrated embodiment of the present invention, agents useful for same and medicine all obtain by commercial sources.
Material and reagent:
Hogwash fat sample collection reclaims refinery's (produce with reference to hogwash fat extractive process Chongqing City grease company limited, Chongqing City biological liquid fuel company limited and by experiment chamber) in hogwash fat, and normal edible vegetable oil (rapeseed oil, soybean oil, peanut oil and corn oil) is all bought in domestic each large supermarket and market of farm produce; Potassium bromide (chromatographically pure) is purchased from Chengdu Ke Long chemical reagent factory.
Instrument and equipment:
Spectrum GX Fourier transform infrared spectrometer Perkin Elmer company; In infrared DTGS detecting device ALPAAI-4LSC U.S. Christ company.
Embodiment 1
1, sample preparation and processing
Hogwash fat sample all passes through refining treatment, and concrete grammar is with reference to hogwash fat method of refining (Pan Jianyu, Yin Pinghe, Yu Hanhao etc., the discriminating research of hogwash fat, frying oil and qualified edible vegetable oil, Food Science, 2003,24(8): 27-29), the judgment basis reference method (Zhu Shiping of refining hogwash fat sample, Liang Jing, slaughter David etc., the hogwash fat method for quick identification based near infrared spectrum and DPLS, Southwestern University's journal: natural science edition, 2012,34(5): 1-6).
Test is total to 51 of sample thiefs, 29 (RHOs7 kind, 7 kinds of rapeseed oils, 5 kinds of soybean oils, 5 kinds of peanut oil, 5 kinds of corn oils), for the foundation of PLS-DA model, all the other 22 blind samples (RHOs6 kind, 4 kinds of rapeseed oils, 4 kinds of soybean oils, 4 kinds of peanut oil, 4 kinds of corn oils) are for prediction (external certificate).
2, spectra collection
FT-MIR spectra collection adopts potassium bromide (KBr) pressed disc method: accurately get 0.3g potassium bromide (chromatographically pure) and carry out compressing tablet, pressure is 27MPa, and the compressing tablet time is 4-5min.Smearing skim oil sample (2 μ L) on smooth kbr tablet, puts into Fourier transform infrared spectrometer and scans, the spectrum of collected specimens.Use the kbr tablet of not smearing oil sample as blank.
The FT-MIR condition of scanning: use DTGS detecting device, spectral resolution 4cm
– 1, the sweep limit of infrared spectrometer is 4000-450cm
– 1, scanning times 16 times.During scanning, pass through background correction, thereby remove moisture content and CO
2to the interference of measuring.Before scanning, after pre-thermal instrument 1h, use.In scanning process, keep 25 ℃ of room temperatures, and strictly control indoor relative humidity 20-50%, keep the consistance of environment.The function software sampling and processing infared spectrum that adopts FT-MIR to carry.
The screening of embodiment 2 influence factors
1, disposal route
In order to remove the impacts such as high frequency random noise, baseline wander and sample be inhomogeneous, the Spectrum3.0 function software that adopts infrared spectrometer to carry carries out necessary pre-service to infared spectrum.According to functional groups such as the C-O existing in all band, C-H, C=O, C=C, all band is divided into 9 regions; Pretreatment mode comprises moving average level and smooth (MAS), normalization (Nor), polynary scatter correction (MSC), standard normal conversion (SNV), the level and smooth differentiate of Savitzky-Golay etc.Mean square deviation and the related coefficient of according to each, analyzing the PLS-DA model of wave band foundation, evaluate the impact of each factor on the discriminatory analysis model of setting up.The OPUS function software that adopts infrared spectrometer to carry carries out necessary pre-service near infrared collection of illustrative plates.Adopt Unscramber statistical analysis software to analyze, adopt offset minimum binary diagnostic method (PLS-DA) to set up normal edible vegetable oil (EOs) and refining hogwash fat (RHOs) discriminatory analysis model.To correct the related coefficient (R of collection and validation-cross collection
2) and mean square deviation standard deviation (RMSE, RMSECV) etc. be evaluation index, adopt leave one cross validation to determine the best number of principal components of modeling.
2, the impact of spectral band on discriminatory analysis model
Fourier transform middle infrared spectrum is widely used in the research of molecular structure and material chemical composition as " fingerprint of molecule ", its wavelength coverage is 4000-450cm
-1, due to the main several functions group in this region and different vibrating modes thereof, bands of a spectrum are wide, overlapping more serious, and a little less than absorption signal, information analysis is complicated.According to functional groups such as the C-O existing in all band, C-H, C=O, C=C, all band is divided into 9 regions, as shown in table 1.Mean square deviation and the related coefficient of the PLS-DA model of setting up according to each analysis wave band are as shown in table 1, and table 1 explanation all band spectrum can reflect sample message more fully, all sidedly, therefore determine employing all band establishment of spectrum PLS-DA discrimination model.
The related coefficient of PLS-DA model and mean square deviation under table 1 different-waveband
3, the impact of data preprocessing method on discriminatory analysis model
Original spectrum cannot with the naked eye directly be distinguished, can not be from original spectrogram the difference of Direct Analysis vegetable oil and hogwash fat, need to it, process by Chemical Measurement software.Simultaneously, in the original spectrum that infrared spectrometer gathers, not only mainly comprise the information relevant to material chemical constitution, also comprise that inhomogeneous, the light scattering of sample or the signal that instrument random noise produces disturb, the existence of these signals can affect accuracy and the repeatability of mensuration.By spectrogram pre-service, these non-information factors can be down to minimum, thereby improve accuracy and the reliability of model.In this research, adopt the methods such as moving average level and smooth (MAS), normalization (Nor), polynary scatter correction (MSC), standard normal conversion (SNV), the level and smooth differentiate of Savitzky-Golay to carry out pre-service to spectrum, the impact of the disturbing factors such as signal and scattering can effectively abate the noise, strengthen the spectral absorption information relevant to evaluation index, specifically in Table 2, utilize PLS-DA method to set up calibration model, by the R of each model relatively
2with indexs such as mean square deviations, select best preprocessing procedures, finally adopt Savitzky-Golay convolution smoothing processing.
The impact of table 2 data processing method on the related coefficient of PLS-DA model and standard deviation
4, PLS-DA main cause subnumber determines
While using PLS-DA method to set up quantitative correction model, the selection of main cause subnumber (claiming again number of principal components) is directly connected to the actual prediction ability of model.To linear system, can select front several main PLS-DA component number, all the other compositions are regarded as random noise or insignificant multiple conllinear factor, yet may also comprise very important information to these compositions of nonlinear system, select best main cause subnumber, make it can overcome the correlativity between independent variable, can reflect the nonlinear relationship between independent variable and dependent variable again, model has quite high fitting precision and obtains enough Accurate Prediction results like this.Application stays a cross verification to verify institute's established model, take and predicts that irregular quadratic sum (PRESS) is parameter, and major component residual analysis is as shown in table 3, determines that best main cause subnumber is 7.
The accumulative total variance contribution ratio of table 3 major component
The method of embodiment 3 based on Fourier transform mid-infrared light spectrum discrimination rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat
1, spectra collection and Savitzky-Golay process
Sample thief is 51 altogether, 29 (RHOs7 kind, 7 kinds of rapeseed oils, 5 kinds of soybean oils, 5 kinds of peanut oil, 5 kinds of corn oils), for the foundation of PLS-DA model, all the other 22 blind samples (RHOs6 kind, 4 kinds of rapeseed oils, 4 kinds of soybean oils, 4 kinds of peanut oil, 4 kinds of corn oils) are for prediction.According to the condition of embodiment 1 and method, carry out spectra collection.
In order to remove the impacts such as high frequency random noise, baseline wander and sample be inhomogeneous, the Spectrum3.0 function software that adopts infrared spectrometer to carry carries out necessary pre-service to infared spectrum.According to top condition definite in embodiment 2, initial full spectral band data are through second order differentiate (Savitzky-Golay, 5 points) convolution smoothing processing, the impact of the disturbing factors such as signal and scattering not only can effectively abate the noise, strengthen the spectral absorption information relevant to evaluation index, and after second order differentiate convolution smoothing processing, can improve the related coefficient of discriminatory analysis model.The FT-MIR spectrogram of 29 calibration set samples processing through Savitzky-Golay as shown in Figure 1.
2, the foundation of PLS-DA calibration model
Adopting Unscramber7.8(CAMO AS, Oslo, Norway) statistical analysis software carries out infrared data analysis, sets up PLS-DA analytical model.
By the oil sample of five kinds, the value of giving 0(RHOs respectively), 1(rapeseed oil), 2(soybean oil), 3(peanut oil), 4(corn oil), as the reference value of spectroscopic data, with related coefficient (r
2) and mean square deviation be evaluation index, adopt leave one cross validation to determine the best number of principal components of modeling.General r
2be worth greatlyr, mean square deviation is less, and model accuracy is higher.If predicted value in the value of giving (0,1,2,3 and 4) ± 0.5 scope, judges its prediction accurately.
Cluster analysis parameter: adopting Euclidean (euclidean) distance, agglomeration method is ward ' s sum of squares of deviations method.
The PLS-DA calibration model of setting up according to test, calculate the major component value (PCs) of 29 samples, linear system can be selected front several main PLS-DA component number, all the other compositions are regarded as random noise or insignificant multiple conllinear factor, in the present embodiment, with the first two, take respectively PC1 as X-axis, PC2 are as Y-axis, 29 sample spot are marked respectively in coordinate system, obtained the X-Y scheme of PCs score, as shown in Figure 2.(in figure, digital 1-29 represents respectively normal edible vegetable oil and refining hogwash fat sample) as seen from Figure 2, each point is the sample spot dimensionality reduction mapping by former hyperspace, has reflected the classification situation of 29 samples.Wherein refining hogwash fat sample (1-7) concentrates on the middle region of left, rapeseed oil sample (8-14) concentrates on a region, below, soybean oil sample (15-19) concentrates on a right region, peanut oil sample (20-24) concentrates on a region, top, and corn oil sample (25-29) concentrates on Yi Ge region, upper right side.
On this basis, adopt the way of cluster analysis (HDA), divide sample into Different groups and carry out similarity evaluation, cluster result as shown in Figure 3, analytic target accurately can be divided into 5 large classes, the first kind is refining hogwash fat, comprises sample 7,1,5,4,6,3,2; Equations of The Second Kind is rapeseed oil, comprises sample 13,11,12,8,14,10,9; The 3rd class is soybean oil, comprises sample 16,15,19,18,17; The 4th class is peanut oil, comprises sample 24,23,22,21,20; The 5th class is corn oil, comprises sample 27,26,25,28,29.As can be seen here, aspect discriminating normal edible vegetable oil (rapeseed oil, soybean oil, peanut oil and corn oil) and refining hogwash fat, the method for cluster analysis can be for selecting suitable raw material that effective guidance is provided.
3, the prediction of PLS-DA discriminatory analysis model
According to the PLS-DA model of above-mentioned foundation, other 22 samples (RHOs6 kind, 4 kinds of rapeseed oils, 4 kinds of soybean oils, 4 kinds of peanut oil, 4 kinds of corn oils) are predicted and found, neither one erroneous judgement, the whole correct decision rate of discrimination model is 100%, result as shown in Figure 4, differentiate respond wellly, illustrate that set up discrimination model is reliable.
Modeling parameters: spectral band 4000-450cm
-1, second order differentiate (Savitzky-Golay, 5 points) process, under this optimal conditions, the mid-infrared spectral PLS-DA discriminatory analysis of the Fourier transform model of setting up has been predicted 22 samples (RHOs6 kind, 4 kinds of rapeseed oils, 4 kinds of soybean oils, 4 kinds of peanut oil, 4 kinds of corn oils) accurately, PLS-DA Model checking accuracy is 100%, for the fast detecting of refining hogwash fat provides new thinking, also significant to improving China's edible oil quality level of control.This result shows that FT-MIR technology can be used as refining hogwash fat and a kind of effective technology means that Bu Tong normal edible vegetable oil is distinguished in conjunction with chemometrics method.
Finally explanation is, above embodiment is only unrestricted in order to technical scheme of the present invention to be described, although the present invention is had been described in detail with reference to preferred embodiment, those of ordinary skill in the art is to be understood that, can modify or be equal to replacement technical scheme of the present invention, and not departing from aim and the scope of technical solution of the present invention, it all should be encompassed in the middle of claim scope of the present invention.
Claims (7)
1. the method based on the normal edible oil of Fourier transform mid-infrared light spectrum discrimination and refining hogwash fat, is characterized in that, comprises the step of carrying out as follows:
(1) set up model
Get normal edible oil and refining hogwash fat is sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, sample is scanned, collected specimens spectrum spectrogram, described sample spectra spectrogram is carried out to spectrum pre-service and obtain pre-processed spectrum spectrogram, described pre-processed spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, sets up the PLS-DA analytical model standard of distinguishing normal edible oil and refining hogwash fat;
(2) evaluation to unknown oil sample
Get unknown oil sample sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, unknown oil sample sample is scanned, gather unknown oil sample sample spectra spectrogram, described unknown oil sample sample spectra spectrogram is carried out to spectrum pre-service and obtain unknown oil sample sample pretreatment spectrum spectrogram, described unknown oil sample sample pretreatment spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, with the analytical model Comparison of standards of setting up in step (1), determine that unknown oil sample sample is normal edible oil or refining hogwash fat.
2. method according to claim 1, it is characterized in that, spectrum pre-service described in described step (1) and step (2) all selects Savitzky-Golay convolution smoothing method to carry out spectrum pre-service, and the parameter that described Savitzky-Golay convolution smoothing method adopts is: second derivative, 5 are level and smooth, multinomial series is 2.
3. method according to claim 1, is characterized in that, in described step (1) and step (2), in the analysis of described offset minimum binary diagnostic method, best main cause subnumber is 7.
4. method according to claim 3, is characterized in that, in described step (1) and step (2), in the analysis of described offset minimum binary diagnostic method, selecting the first two main cause subnumber PC1 is that X-axis, PC2 are that Y-axis is set up linear relationship analysis.
5. method according to claim 1, is characterized in that, keeps room temperature 25 ℃ ± 1 in described scanning process, controls indoor relative humidity at 20%-50%.
6. method according to claim 1, is characterized in that, adopts coating method to analyze sample, and liquid oil sample is spread upon and on salt sheet, makes liquid film and analyze, and described salt sheet is KBr.
7. the method based on Fourier transform mid-infrared light spectrum discrimination rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat, comprises the step of carrying out as follows:
(1) set up model
Getting rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat is sample, utilizes Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, sample is scanned, collected specimens spectrum spectrogram, described sample spectra spectrogram is carried out to spectrum pre-service and obtain pre-processed spectrum spectrogram, described pre-processed spectrum spectrogram is analyzed in conjunction with offset minimum binary diagnostic method, sets up the PLS-DA analytical model standard of distinguishing rapeseed oil, soybean oil, peanut oil, corn oil and refining hogwash fat;
(2) evaluation to unknown oil sample
Get unknown oil sample sample, utilize Fourier transform mid-infrared light spectrometer, at 4000-450cm
– 1spectral wavelength within the scope of, unknown oil sample sample is scanned, gather unknown oil sample sample spectra spectrogram, described unknown oil sample sample spectra spectrogram is carried out to spectrum pre-service and obtain unknown oil sample sample pretreatment spectrum spectrogram, described unknown oil sample sample pretreatment spectrum spectrogram is analyzed through offset minimum binary diagnostic method, with the analytical model Comparison of standards of setting up in step (1), determine the kind of unknown oil sample sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310629904.6A CN103592256A (en) | 2013-11-29 | 2013-11-29 | Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310629904.6A CN103592256A (en) | 2013-11-29 | 2013-11-29 | Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103592256A true CN103592256A (en) | 2014-02-19 |
Family
ID=50082482
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310629904.6A Pending CN103592256A (en) | 2013-11-29 | 2013-11-29 | Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103592256A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105424636A (en) * | 2015-12-29 | 2016-03-23 | 南京财经大学 | Rapid detection method for grain fungal pollution cases and applications thereof |
| CN105675534A (en) * | 2016-03-25 | 2016-06-15 | 北京市农林科学院 | Method for quickly and nondestructively identifying polished grains |
| CN108287046A (en) * | 2018-02-05 | 2018-07-17 | 广东轻工职业技术学院 | A kind of vehicular engine liquid leakage detector device and method |
| CN109358022A (en) * | 2018-11-22 | 2019-02-19 | 陕西中烟工业有限责任公司 | A kind of method of the quick-fried pearl type of quick discrimination cigarette |
| CN111220563A (en) * | 2018-11-26 | 2020-06-02 | 吉林农业大学 | Method for detecting recovered oil by infrared spectrum |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100260903A1 (en) * | 2007-01-08 | 2010-10-14 | Wei Ai-Ping | Device for the qualification of cooking oils, and methods |
| CN102636454A (en) * | 2012-05-15 | 2012-08-15 | 武汉工业学院 | Method for quickly measuring content of low carbon number fatty acid in edible oil by near infrared spectrum |
-
2013
- 2013-11-29 CN CN201310629904.6A patent/CN103592256A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100260903A1 (en) * | 2007-01-08 | 2010-10-14 | Wei Ai-Ping | Device for the qualification of cooking oils, and methods |
| CN102636454A (en) * | 2012-05-15 | 2012-08-15 | 武汉工业学院 | Method for quickly measuring content of low carbon number fatty acid in edible oil by near infrared spectrum |
Non-Patent Citations (7)
| Title |
|---|
| BANU F 等: "Detection oh Hazelnut Oil Adulteration Using FT-IR Spectroscopy", 《AGRICULTURAL AND FOOD CHEMISTRY》, vol. 50, no. 14, 31 December 2002 (2002-12-31), pages 3898 - 3901, XP026892749, DOI: doi:10.1016/j.talanta.2009.11.024 * |
| GOZDE GURDENIZ 等: "Detection of adulteration of extra-virgin olive oil by chemometric analysis of mid-infrared spectral data", 《FOOD CHEMISTRY》, vol. 116, 31 December 2009 (2009-12-31), XP026098338, DOI: doi:10.1016/j.foodchem.2009.02.068 * |
| 严衍禄: "《近红外光谱分析基础与应用》", 31 January 2005, article "近红外光谱分析的信号处理", pages: 98-103 * |
| 孙通 等: "地沟油鉴别的研究现状和展望", 《食品工业科技》, vol. 33, no. 24, 31 December 2012 (2012-12-31), pages 418 - 422 * |
| 梁逸曾 等: "《复杂体系仪器分析-白、灰、黑分析体系及其多变量解析方法》", 31 March 2012, article "主成分分析及偏最小二乘线性判别法", pages: 524-525 * |
| 沈雄: "基于近红外光谱的餐饮废弃油脂快速鉴别模型及优化研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》, no. 06, 15 June 2013 (2013-06-15) * |
| 许洪勇 等: "傅里叶变换红外光谱法鉴别地沟油的研究", 《现代食品科技》, vol. 28, no. 6, 31 December 2012 (2012-12-31), pages 1 - 3 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105424636A (en) * | 2015-12-29 | 2016-03-23 | 南京财经大学 | Rapid detection method for grain fungal pollution cases and applications thereof |
| CN105424636B (en) * | 2015-12-29 | 2018-04-27 | 南京财经大学 | A kind of quick determination method of control pollution condition and its application |
| CN105675534A (en) * | 2016-03-25 | 2016-06-15 | 北京市农林科学院 | Method for quickly and nondestructively identifying polished grains |
| CN108287046A (en) * | 2018-02-05 | 2018-07-17 | 广东轻工职业技术学院 | A kind of vehicular engine liquid leakage detector device and method |
| CN108287046B (en) * | 2018-02-05 | 2024-04-05 | 广东轻工职业技术学院 | Vehicle engine liquid leakage detection instrument and method |
| CN109358022A (en) * | 2018-11-22 | 2019-02-19 | 陕西中烟工业有限责任公司 | A kind of method of the quick-fried pearl type of quick discrimination cigarette |
| CN111220563A (en) * | 2018-11-26 | 2020-06-02 | 吉林农业大学 | Method for detecting recovered oil by infrared spectrum |
| CN111220563B (en) * | 2018-11-26 | 2023-02-10 | 吉林农业大学 | Method for detecting recovered oil by infrared spectrum |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Evaluation of matcha tea quality index using portable NIR spectroscopy coupled with chemometric algorithms | |
| CN102636454B (en) | The method of low carbon number content of fatty acid near infrared spectrum quick test edible oil | |
| CN104807775B (en) | It is a kind of for identify frying oil quality NIR spectra analysis model and method | |
| CN102692388B (en) | Digital imaging system and method for rapid detection of organophosphorus pesticide residues in fruits and vegetables by using same | |
| CN103592256A (en) | Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform | |
| CN102778442B (en) | Method for rapidly identifying type of balsam material liquid for cigarette | |
| CN103543123A (en) | Infrared spectrum recognition method for adulterated milk | |
| CN102288572A (en) | Method for quickly detecting content of index ingredient of traditional Chinese medicinal material by utilizing near infrared spectrum technique | |
| CN103901094A (en) | Oil detection and identification method based on ion mobility spectrometer | |
| CN102053072B (en) | Method for identifying brewed sauce and prepared sauce based on partial least squares (PLS) method | |
| CN102706915B (en) | A kind of detection method of waste oil | |
| Li et al. | Identification of aged-rice adulteration based on near-infrared spectroscopy combined with partial least squares regression and characteristic wavelength variables | |
| CN103033486A (en) | Method for near infrared spectrum monitoring of quality of pericarpium citri reticulatae and citrus chachiensis hortorum medicinal materials | |
| Li et al. | Nondestructive detection of frying times for soybean oil by NIR-spectroscopy technology with Adaboost-SVM (RBF) | |
| CN112595692A (en) | Establishment method of fruit total sugar content prediction model and fruit total sugar content prediction method | |
| CN109374548A (en) | A method of quickly measuring nutritional ingredient in rice using near-infrared | |
| Chen et al. | Determination of total polar compounds in frying oils by PE‐film‐based FTIR and ATR‐FTIR spectroscopy | |
| Sheng et al. | Nondestructive determination of lignin content in Korla fragrant pear based on near-infrared spectroscopy | |
| CN110672578A (en) | Model universality and stability verification method for polar component detection of frying oil | |
| CN102175813B (en) | Method for half quantitatively and fast detecting contents of delta 9-THC (tetrahydrocannabinol) in cannabis plants | |
| Chen et al. | Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information | |
| CN103743697A (en) | Method for monitoring tea production in real time by adopting near infrared spectrum | |
| CN104764837A (en) | Hogwash oil identification method | |
| Dong et al. | Nondestructive method for analysis of the soybean quality | |
| CN104297205A (en) | Method for quickly and nondestructively identifying edible oil |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140219 |