CN105758981A - Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm - Google Patents
Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm Download PDFInfo
- Publication number
- CN105758981A CN105758981A CN201410797245.1A CN201410797245A CN105758981A CN 105758981 A CN105758981 A CN 105758981A CN 201410797245 A CN201410797245 A CN 201410797245A CN 105758981 A CN105758981 A CN 105758981A
- Authority
- CN
- China
- Prior art keywords
- pso
- vegetable oil
- svm
- technology
- fatty acid
- 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
Landscapes
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
Abstract
The invention provides a plant oil classifying method based on combination of a GC-MS technology and a PSO-SVM algorithm. The method comprises the following steps: using the GC-MS technology to obtain the qualitative and quantified results of fatty acids in six different kinds of plant oils and establish the fatty acid fingerprint database of the plant oils; adopting the POS optimal parameter-based SVM classification algorithm to classify samples and establish a PSO-SVM model; and substituting data to the PSO-SVM model, and determining the class of the samples according to a calculation result. The classifying method has the advantages of simple pretreatment operation, high sensitivity and high accuracy.
Description
Technical field
The invention relates to the sorting technique of vegetable oil.Specifically refer to the fatty acid profile data based on GC-MS technology obtains in conjunction with PSO-SVM model, different types of vegetable oil is classified, for differentiating the classification of edible oil.
Background technology
Because of fatty acid necessary to multiple human body in vegetable oil, its nutritive value is subject to people's attention higher than animal oil, now substantially becomes modal edible oil in people's daily life.Edible oil is the necessary that people live, and is to provide human body heat energy and the important foodstuffs of essential fatty acid, promotion fat soluble vitamin absorption, provides the most direct energy source for human body.Along with the raising of expanding economy and living standards of the people, people increasingly pay attention to the health problem of edible oil.Due to the complexity of edible oil constituent, edible oil can be divided into much different kinds and it becomes branch to change in the cooking with food processing.Therefore, the method setting up plant identification oil efficiently and accurately has great importance.
The instrument analysis technology method of current detection and identification edible oil mainly has ultraviolet visible spectrophotometry, near infrared spectroscopy, fluorescence spectrophotometry, gas chromatography, liquid chromatography, nuclear magnetic resonance method etc..
In pertinent literature, Galtier etc. adopts near infrared spectroscopy to analyze the composition of triacylglycerol and fatty acid in olive oil, then utilizes offset minimum binary-discriminant analysis to differentiate the place of production of olive oil, yields good result.Dupuy etc. apply principal component analysis and Fourier transform infrared derivative spectrum are analyzed, it is achieved that the differentiation to Oleum Arachidis hypogaeae semen and olive oil.Olivers etc. adopt head-space sampler and mass spectrum to be joined directly together, and have detected the volatile ingredient in different sources primary olive oil, and these data PCA, LDA olive oil are classified, accuracy respectively 94.3%, 90.5% and 80%.Shaw etc. utilize nuclear magnetic resonance technique to be classified different sources and different classes of olive oil identifying in conjunction with PCA, PLS, the predictive ability of PLS and PCA respectively 90% and 70%.In above-mentioned document, each algorithm is solving to have on linear problem good application power, but has significant limitation processing in Nonlinear Classification problem.
Summary of the invention
The problem that the invention to solve is the deficiency overcoming vegetable oil Sample Pretreatment Technique and sorting technique, it is proposed that a kind of method quickly, accurately distinguishing vegetable oil.Specific embodiments of the present invention are:
1) different sources, different types of vegetable oil sample are collected；
2) methyl esterification of fatty acid processes and GC-MS technical Analysis: vegetable oil sample is carried out esterification process, by the fatty acid finger printing of GC-MS technology herborization oil；
3) kind of vegetable oil distinguished by the SVM model that PSO optimizes: the vegetable oil sample of collection is randomized into three data sets, i.e. calibration set, checking collection and forecast set, each data set comprises different types of vegetable oil sample, calibration set and checking collection are for building the PSO SVM model optimized, it was predicted that collection is for proving the performance of disaggregated model.
Further, described method is being used in combination of the GC-MS fatty acid profile SVM model with PSO optimization.
Further, step 2) described in methyl esterification of fatty acid processing method be: take 0.1mL edible oil in centrifuge tube, add the mixed liquor of 2mL1:1 (v/v) petroleum ether and toluene, after jolting makes lipid solubilization, add the methanol solution of 2mL5mol/LKOH, 5min is at room temperature stood after mixing, add the hydrochloric acid of 2mL2mol/L, jolting, stand, add anhydrous sodium sulfate to dry, take the supernatant to be measured.
Further, step 3) described GC-MS analysis condition is: chromatographic column is capillary column, injector temperature is 250 DEG C, sample size 1 μ L, and injection port split ratio is 40:1, the heating schedule of furnace temperature is initial temperature 60 DEG C, rise to 215 DEG C with 15 DEG C/min, rise to 250 DEG C with 10 DEG C/min, rise to 260 DEG C with 2 DEG C/min, rise to 280 DEG C with 5 DEG C/min, keep 2min.Carrier gas is helium, and carrier gas flux is 1mL/min, and data acquisition modes is total ions chromatogram.Ion source is EI, and ion source temperature is 250 DEG C, and transmission line temperature is 270 DEG C.
The invention has the advantage that and has the benefit effect that
1) the edible oil oils and fats esterification method that the present invention takes need not heat, simple to operate, saves the time.
2) GC-MS technology had both had the ability of gas chromatogram high efficiency separation, had again the feature of mass spectrum precise Identification compound structure, can reach to carry out qualitative, quantitative requirement simultaneously, be used widely in chemical analysis field.
3) particle group optimizing (PSO) algorithm has good performance in different optimization problems, and has of a relatively high convergence efficiency, and can guarantee that again the global optimization of solution simultaneously.SVM based on Statistical Learning Theory has more advantage, has solid theoretical basis, good Generalization Ability, powerful Nonlinear Processing ability and high dimensional data disposal ability.Even if training sample is less, SVM still has good performance, it is ensured that Learning machine has good generalization ability.In addition, SVM is eventually converted into convex optimization problem, it is ensured that the global optimum of algorithm, it is to avoid the problem that algorithm is absorbed in local minimum.SVM can obtain the result of optimum in conjunction with the disaggregated model that PSO algorithm is set up.
Detailed description of the invention
Below in conjunction with embodiment, the present invention is further described, and following embodiment is illustrative, is not determinate, it is impossible to limits protection scope of the present invention with following embodiment.
Embodiment 1
Sample collection: 66 edible oil sample standard deviations are purchased and local supermarket, soybean oil 14 kinds, Oleum Arachidis hypogaeae semen 8 kinds, Oleum Brassicae campestris 14 kinds, Oleum Camelliae 8 kinds, Semen Maydis oil 8 kinds, Oleum sesami 10 kinds.
Plant Oil Fatty Acid Methyl Ester: take 0.1mL oil in centrifuge tube, add the mixed liquor of 2mL petroleum ether (30 DEG C-60 DEG C) and toluene (by volume 1:1), after jolting makes lipid solubilization, add the methanol solution of 20mL5mol/LKOH, at room temperature stand 5min after mixing, add the hydrochloric acid of 2mL2mol/L, jolting, stand, add anhydrous sodium sulfate and dry, take the supernatant to be measured.
The GC-MS analysis condition of fatty acid methyl ester: instrument: LecoPegasus4DGC-TOFMS (Leco Corporation of the U.S.).GC conditions: chromatographic column is capillary column, model is DB-5MS (30m*0.25mm*0.25 μm), injector temperature is 250 DEG C, sample size 1 μ L, and injection port split ratio is 40:1, the heating schedule of furnace temperature is initial temperature 60 DEG C, rise to 215 DEG C with 15 DEG C/min, rise to 250 DEG C with 10 DEG C/min, rise to 260 DEG C with 2 DEG C/min, rise to 280 DEG C with 5 DEG C/min, keep 2min.Carrier gas is the helium of purity 99.99%, and carrier gas flux is 1mL/min, and data acquisition modes is total ions chromatogram.Mass Spectrometry Conditions: needed for mass spectrum, total time is identical with gas chromatogram, solvent delay 10min, institute's mass metering ranges for 30-500amu, and ion source is EI, and its temperature is 250 DEG C, and transmission line temperature is 270 DEG C, and voltage is 1500V, and electron energy is-70eV.
The collection of illustrative plates of 66 Plant Oil Fatty Acid Methyl Esters is analyzed, calculates each content of fatty acid percentage ratio in each sample.Concrete outcome table 1
In table 1 vegetable oil, the GC-MS of fatty acid relative amount analyzes result
A: soybean oil B: Oleum Brassicae campestris C: Oleum Arachidis hypogaeae semen D: Semen Maydis oil E: Oleum sesami F: Oleum Camelliae SFA: satisfied fatty acid PUFA: polyunsaturated fatty acid MUFA: monounsaturated fatty acid
Although having difference between individual vegetable oil, still the kind either directly through total ions chromatogram differentiation vegetable oil is very difficult.The present invention adopts the PSO SVM model optimized to distinguish the kind of vegetable oil.
The foundation of SVM model that PSO optimizes: the sample matrix according to fatty acid relative amount data construct in table 1 vegetable oil.When setting up model, 66 vegetable oil samples of collection are randomized into three data sets, wherein 33 samples of calibration set, checking 17 samples of collection, it was predicted that collecting 16 samples, each data set comprises the sample of six kinds of different vegetable oil.Calibration set and checking collection are for building the PSO SVM model optimized, it was predicted that collection is for proving the performance of disaggregated model.Table 2 shows that the PSO SVM model prediction different pieces of information optimized concentrates the result of vegetable oil kind.
The SVM model prediction different pieces of information that table 2.2PSO optimizes concentrates the result of vegetable oil kind
It is above to the invention example to be described in detail, but described content is only the specific embodiment of the invention, it is not to be regarded as the practical range for limiting the invention, all equalizations done according to the invention application range change and improvement etc., all should still belong within the patent covering scope of the invention.
Claims (4)
1. based on the GC-MS technology vegetable oil sorting technique in conjunction with PSO-SVM algorithm, it is characterised in that comprise the steps:
1) different sources, different types of vegetable oil sample are collected；
2) methyl esterification of fatty acid processes and GC-MS technical Analysis: vegetable oil sample is carried out esterification process, by the fatty acid finger printing of GC-MS technology herborization oil；
3) kind of vegetable oil distinguished by the SVM model that PSO optimizes: the vegetable oil sample of collection is randomized into three data sets, i.e. calibration set, checking collection and forecast set, each data set comprises different types of vegetable oil sample, calibration set and checking collection are for building the PSO SVM model optimized, it was predicted that collection is for proving the performance of disaggregated model.
2. according to claim 1 based on the GC-MS technology vegetable oil sorting technique in conjunction with PSO-SVM algorithm, it is characterised in that: this method is GC-MS fatty acid profile and being used in combination of the PSO SVM model optimized.
3. according to claim 1 based on the GC-MS technology vegetable oil sorting technique in conjunction with PSO-SVM algorithm, it is characterized in that: described step 2) methyl esterification of fatty acid processing method is: take 0.1mL edible oil in centrifuge tube, add the mixed liquor of 2mL1:1 (v/v) petroleum ether and toluene, after jolting makes lipid solubilization, add the methanol solution of 2mL5mol/LKOH, 5min is at room temperature stood after mixing, add the hydrochloric acid of 2mL2mol/L, jolting, stand, add anhydrous sodium sulfate to dry, take the supernatant to be measured.
4. according to claim 1 based on the GC-MS technology vegetable oil sorting technique in conjunction with PSO-SVM algorithm, it is characterized in that: described step 3) in GC-MS analysis condition be: chromatographic column is capillary column, injector temperature is 250 DEG C, sample size 1 μ L, and injection port split ratio is 40:1, the heating schedule of furnace temperature is initial temperature 60 DEG C, rise to 215 DEG C with 15 DEG C/min, rise to 250 DEG C with 10 DEG C/min, rise to 260 DEG C with 2 DEG C/min, rise to 280 DEG C with 5 DEG C/min, keep 2min.Carrier gas is helium, and carrier gas flux is 1mL/min, and data acquisition modes is total ions chromatogram, and ion source is EI, and ion source temperature is 250 DEG C, and transmission line temperature is 270 DEG C.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410797245.1A CN105758981A (en) | 2014-12-19 | 2014-12-19 | Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410797245.1A CN105758981A (en) | 2014-12-19 | 2014-12-19 | Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105758981A true CN105758981A (en) | 2016-07-13 |
Family
ID=56339246
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410797245.1A Pending CN105758981A (en) | 2014-12-19 | 2014-12-19 | Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105758981A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106932510A (en) * | 2017-01-17 | 2017-07-07 | 江苏大学 | The sorting technique of one vegetable oil |
| CN110715985A (en) * | 2019-08-29 | 2020-01-21 | 北京工商大学 | Method for judging frying oil quality by combining gas phase fingerprint spectrum with partial least square method |
| CN111060633A (en) * | 2019-12-30 | 2020-04-24 | 北京工商大学 | Method for establishing grease waste judgment model in frying process based on characteristic flavor components and waste judgment method |
| CN113033066A (en) * | 2021-04-07 | 2021-06-25 | 温州大学 | Method for establishing near infrared spectrum identification model of sargassum fusiforme production area, strain and cultivation mode and identification method |
-
2014
- 2014-12-19 CN CN201410797245.1A patent/CN105758981A/en active Pending
Non-Patent Citations (5)
| Title |
|---|
| R.M.SHARKAWY ET AL: "Particle Swarm Optimization Feature Selection for the Classification of Conducting Particles in Transformer Oil", 《IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION》 * |
| 夏克文等: "基于改进PSO算法的LS-SVM油层识别模型", 《控制与决策》 * |
| 姜显光等: "功能性植物油中脂肪酸的分析", 《鞍山师范学院学报》 * |
| 李英等: "用气相色谱-质谱联用法测定植物油中脂肪酸的组成", 《山西食品工业》 * |
| 顾婷: "化学计量学结合GC-MS技术在食用油与癌细胞鉴别中的应用", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106932510A (en) * | 2017-01-17 | 2017-07-07 | 江苏大学 | The sorting technique of one vegetable oil |
| CN110715985A (en) * | 2019-08-29 | 2020-01-21 | 北京工商大学 | Method for judging frying oil quality by combining gas phase fingerprint spectrum with partial least square method |
| CN110715985B (en) * | 2019-08-29 | 2022-11-15 | 北京工商大学 | Method for judging frying oil quality by combining gas phase fingerprint spectrum with partial least square method |
| CN111060633A (en) * | 2019-12-30 | 2020-04-24 | 北京工商大学 | Method for establishing grease waste judgment model in frying process based on characteristic flavor components and waste judgment method |
| CN113033066A (en) * | 2021-04-07 | 2021-06-25 | 温州大学 | Method for establishing near infrared spectrum identification model of sargassum fusiforme production area, strain and cultivation mode and identification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Authentication of edible vegetable oils adulterated with used frying oil by Fourier Transform Infrared Spectroscopy | |
| Huang et al. | Identification of waste cooking oil and vegetable oil via Raman spectroscopy | |
| CN105758981A (en) | Plant oil classifying method based on combination of GC-MS technology and PSO-SVM algorithm | |
| Chu et al. | Accuracy and stability improvement for meat species identification using multiplicative scatter correction and laser-induced breakdown spectroscopy | |
| CN105574474B (en) | A kind of biometric image recognition methods based on Information in Mass Spectra | |
| CN101532991B (en) | Method for quickly detecting content of fatty acid in marine microalgae | |
| Mondello et al. | Rapid, micro-scale preparation and very fast gas chromatographic separation of cod liver oil fatty acid methyl esters | |
| Chen et al. | Detection of adulteration in canola oil by using GC‐IMS and chemometric analysis | |
| CN103217411B (en) | Raman-spectrum-based method for detecting content of oleic acid, linoleic acid and saturated fatty acid in edible vegetable oil | |
| CN102841171A (en) | Detection method for tea seed oil adulteration based on ratio of main fatty acids | |
| Xie et al. | Rapid detection and quantification by GC–MS of camellia seed oil adulterated with soybean oil | |
| Hu et al. | Profiling of triacylglycerols in plant oils by high-performance liquid chromatography–atmosphere pressure chemical ionization mass spectrometry using a novel mixed-mode column | |
| CN104237409A (en) | Construction method for GC-MS (Gas Chromatography-Mass Spectrometer) fingerprint of edible oil and application of GC-MS fingerprint | |
| Muthukumar et al. | Detection of adulteration in sunflower oil using paper-based microfluidic lab-on-a-chip devices | |
| Jiang et al. | Qualitative identification of the edible oil storage period using a homemade portable electronic nose combined with multivariate analysis | |
| CN104267123A (en) | Analysis method of free sterol components in edible oil sample and swill-cooked dirty oil identification method | |
| CN105092512A (en) | Fourier transform infrared spectroscopy technology-based method for detecting camellia oleosa seed oil | |
| Goding et al. | Comparison of GC stationary phases for the separation of fatty acid methyl esters in biodiesel fuels | |
| Han et al. | Benefit of the use of GCxGC/MS profiles for 1D GC/MS data treatment illustrated by the analysis of pyrolysis products from east Asian handmade papers | |
| Flood et al. | Analysis of biodiesel feedstock using GCMS and unsupervised chemometric methods | |
| Chen et al. | Comprehensive adulteration detection of sesame oil based on characteristic markers | |
| CN104007214B (en) | A kind of method detecting volatile ingredient in fragrant product based on PY-GC-MS | |
| CN108802209A (en) | A kind of cod-liver oil Adulteration identification method | |
| WANG et al. | Analysis of yield stability and test site representativeness of sugarcane trials using combined AMMI and HA-GGE biplot models | |
| CN103743849A (en) | Ion chromatography-high resolution mass spectrum hyphenation method for screening and authenticating multiple organic acids in dairy products synchronously and rapidly |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160713 |
|
| RJ01 | Rejection of invention patent application after publication |