CN101021495A - Method for distinguishing petroleum products variety - Google Patents
Method for distinguishing petroleum products variety Download PDFInfo
- Publication number
- CN101021495A CN101021495A CN 200710078320 CN200710078320A CN101021495A CN 101021495 A CN101021495 A CN 101021495A CN 200710078320 CN200710078320 CN 200710078320 CN 200710078320 A CN200710078320 A CN 200710078320A CN 101021495 A CN101021495 A CN 101021495A
- Authority
- CN
- China
- Prior art keywords
- petroleum products
- sample
- variety
- dielectric spectra
- discriminating
- 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 provides a method to identify oil products species. It is characterized in that: Achieve fingerprint information of dielectric property of known oil products by analyzing dielectric spectrum. Build mathematic model to identify oil products by chemical metrology. Analyze dielectric spectrum of unknown oil samples using method like above and compare with identified model of oil products species to identify. With its quick analyzing and identifying speed of this invention, detecting process of dielectric spectrum can be completed in about 2 minutes. The invention can identify hydrocarbon oil products including gasoline, coal oil, diesel oil, gas engine oil, gear oil and hydraulic oil etc., which has an extensive application.
Description
Technical field
The present invention relates to a kind of method of differentiating petroleum products variety, specifically a kind of method of using dielectric spectra analysis discriminating petroleum products variety.
Background technology
It is that links such as petroleum products production, circulation, management are needed that the kind of petroleum products is differentiated, because petroleum products is formed complicated, the base oil of the petroleum products of different manufacturers, different brands, adjuvant difference, the identification to petroleum products does not at present also have a kind of effective technical means.
The analysis of petroleum products adopts the desired analysis test method of petroleum products standard to carry out the analytical test of quality index or performance usually, identification to petroleum products can be adopted analytical approachs such as infrared spectrum, liquid chromatography, mass spectrum, or by people's sense organ petroleum products outward appearance, color, smell is carried out the petroleum products variety differentiation.But can not obtain the result of reliable petroleum products variety by the method for analyzing petroleum products quality or performance; Sense organ by the people carries out petroleum products variety to petroleum products outward appearance, color, smell, and to differentiate its reliability also very poor; Analytical approachs such as infrared spectrum, liquid chromatography, mass spectrum are differentiated petroleum products variety and are then needed long period and professional knowledge.Although the petroleum products administrative authority of many countries and petroleum products manufacturer are endeavouring to ban fake and forged petroleum products, their work is mainly still discerned the packing of fake and forged petroleum products.
Along with the development of analysis and testing technology, Chemical Measurement etc., these technology are organically combined, work out a kind of good reliability and can differentiate fast that the method for petroleum products variety just becomes instant problem.
Summary of the invention
Purpose of the present invention just is to provide a kind of method of quick discriminating petroleum products variety.
The object of the present invention is achieved like this: a kind of method of differentiating petroleum products variety, it is characterized in that: adopt the dielectric spectra analysis to obtain the finger print information of known petroleum products dielectric property, make up the mathematical model of petroleum products identification with chemometrics method, the sample of unknown petroleum products is carried out comparing with described petroleum products variety discriminating model after the dielectric spectra analysis identical with the sample of described various known petroleum products varieties, realize discriminating petroleum products.
Specifically, a kind of method of differentiating petroleum products variety is characterized in that: it is to carry out the dielectric spectra analysis by the sample to various known petroleum products varieties to set up the petroleum products variety discriminating model; Sample with unknown petroleum products carries out comparing with described petroleum products variety discriminating model after the dielectric spectra analysis identical with described various known petroleum products variety samples then; If the distance of a certain known petroleum products variety sample is less than this known petroleum products variety sample corresponding threshold in the sample of described unknown petroleum products and the described petroleum products variety discriminating model, so described unknown petroleum products sample just with described petroleum products variety discriminating model in the sample of this known petroleum products variety belong to the petroleum products of same kind, thereby differentiated the kind of described unknown petroleum products sample.
The method of above-mentioned discriminating petroleum products variety is to carry out as follows: 1), carry out the dielectric spectra analysis and set up the standard dielectric spectra collection of the sample of various known petroleum products varieties by the sample to various known petroleum products varieties; 2), the standard dielectric spectra collection of the sample of various known petroleum products varieties is proofreaied and correct and pre-service; 3), to the extraction of the dielectric spectra characteristic information of the sample of various known petroleum products varieties; 4), set up the discriminating model of the unknown petroleum products specimen types of identification with the dielectric spectra characteristic information of the sample of various known petroleum products varieties; 5), unknown petroleum products sample and known petroleum products sample being carried out kind relatively differentiates.
The above-mentioned standard dielectric spectra of setting up the sample of various known petroleum products varieties is meant and carries out the dielectric spectra analysis to the sample of various known petroleum products varieties is parallel, promptly measures the standard dielectric spectra of the sample of various known petroleum products varieties.
In order to set up reliable and stable standard dielectric spectra collection, the above-mentioned standard dielectric spectra of measuring various known petroleum products variety samples is meant the variety classes sample is carried out its dielectric spectra of twice duplicate measurements and with the standard dielectric spectra of its average dielectric spectra as this kind sample, thereby obtain the standard dielectric spectra spectrogram of various known petroleum products variety samples, the standard dielectric spectra collection of promptly above-mentioned various known petroleum products variety samples.
In order to make the standardization of above-mentioned standard dielectric spectra spectrogram, offset background interference and to improve the quality of dielectric spectra, the dielectric spectra collection of the sample of above-mentioned various known petroleum products varieties is proofreaied and correct and pre-service; Promptly adopt any one or any two kinds or any three kinds of dielectric spectra preprocess methods in level and smooth or centralization or differentiate or the normalization that the dielectric spectra collection of the sample of above-mentioned various known petroleum products varieties is proofreaied and correct; Adopt which kind of bearing calibration to select according to the quality of dielectric spectra and the situation of interference, pre-service also can amplify the original signal difference of hiding out, improves the resolution of dielectric spectra, and it is directly perceived more, reliable that petroleum products variety is differentiated.
In the present invention, the important evidence that petroleum products variety is differentiated is the similarity (or otherness) of dielectric spectra feature, and therefore correctly obtaining the dielectric spectra feature is key one ring of accurately differentiating.The characteristics that dielectric spectra is analyzed (technology) are that the dielectric spectra data volume is big, can react the composition of different petroleum productss, the characteristic information of structure, therefore, carry out qualitative discrimination in the more dependence dielectric spectras of the qualitative discrimination analysis full rate information of dielectric spectra analysis (technology) discriminating petroleum products variety; The dielectric spectra Feature Extraction of the sample of above-mentioned various known petroleum products varieties adopts principal component analysis (PCA) or partial least square method or wavelet analysis method, analyzes the reliability of identification result to improve dielectric spectra.
It is to adopt mode identification method that the dielectric spectra characteristic information of above-mentioned sample with various known petroleum products varieties is set up the petroleum products variety discriminating model, promptly with distance discrimination or cluster analysis or even linear multivariate analysis (SIMCA) method of statistics or partial least-square regression method.Carry out discriminatory analysis with pattern-recognition, the dielectric spectra characteristic information that is about to various known petroleum products variety samples is divided into training set and inspection set two parts; Wherein training set should be identical with kind of petroleum products kind in the inspection set; Then various petroleum productss are carried out initialize according to priori, set up the kind discriminating model of unknown petroleum products sample, come the performance of evaluation model at last with inspection set.
It is that unknown petroleum products sample is carried out the dielectric spectra analysis identical with known petroleum products variety sample that the kind of above-mentioned unknown petroleum products sample is differentiated, the extraction of dielectric spectra correction and pre-service and dielectric spectra characteristic information, compare with above-mentioned petroleum products variety discriminating model then, if the distance of a certain known petroleum products variety sample is less than this known petroleum products variety sample corresponding threshold in the dielectric spectra of above-mentioned unknown petroleum products sample and the above-mentioned discriminating model, so above-mentioned unknown petroleum products sample just with above-mentioned discriminating model in the sample of this known petroleum products variety belong to the petroleum products of same kind, thereby differentiated the kind of above-mentioned unknown petroleum products sample.
In order to guarantee the reliability of the inventive method, the above-mentioned extraction that unknown petroleum products sample is carried out dielectric spectra analysis, dielectric spectra correction and the pre-service identical with known petroleum products variety sample and dielectric spectra characteristic information is meant that each step and the parameters that wherein relate to are all identical.The measuring method that is adopted when promptly obtaining unknown petroleum products sample dielectric spectra, the measuring method that is adopted in the time of must be with the standard dielectric spectra that obtain in the discriminating model as the known petroleum products variety sample of reference sample is consistent, and that is to say that reference sample in the discriminating model and the source of variation between the unknown sample dielectric spectra should reduce as far as possible; The parameters such as actuation duration of the method for sampling, sensor temperature, excitation frequency, driving voltage, excitation waveform, each excitation frequency point should be consistent; Also should be consistent when the extraction of dielectric spectra correction and pre-service and dielectric spectra characteristic information the dielectric spectra of unknown petroleum products sample with the reference sample in the discriminating model.
Dielectric spectra analysis among the present invention, its driving voltage amplitude range is 1V～15V, frequency range is 50Hz～16MHz, excitation waveform be triangular wave or/and sine wave or/and square wave.
Petroleum products variety among the present invention is to be the various petroleum hydrocarbon fuel and the lubricating oil of principal ingredient with the petroleum hydrocarbon, as gasoline, kerosene, diesel oil, I. C. engine oil, gear oil, hydraulic pressure wet goods.
What dielectric spectra was measured is the dielectric properties of material, and it is from the say so function of frequency of dynamic excitation signal of mathematics.Dielectric spectra generally characterizes with specific inductive capacity based on the dipole moment and the external electromagnetic field interactions of sample.
The principle of dielectric spectra: the sign mechanism of material dielectric properties can be according to the medium that is studied to the difference of powering up magnetic responsiveness and difference.Various sign mechanism all are to interrelate with its characteristic frequency that adds electromagnetic field, and characteristic frequency and its characteristic time are to be mutually related.Therefore, from the general extent, the sign mechanism of dielectric properties can be divided into two kinds of relaxation and resonance process.Petroleum products is the dielectric material of a quasi-representative, and under the excitation of different excitation frequencies, driving voltage and excitation waveform, the different compounds of forming structure produce the relaxation motion of different characteristic in the petroleum products, show different dielectric characterization information.Therefore, can obtain the fingerprint characteristic of different petroleum products dielectric properties by the dielectric spectra analysis, these finger print informations have uniqueness, and this differentiates for petroleum products variety and lays a good foundation.Dielectric spectra technology, Chemical Measurement and computer technology are combined, and the fingerprint characteristic of the dielectric property of the different petroleum productss that obtain by the dielectric spectra analysis can be differentiated different petroleum productss.The present invention uses the finger print information that dielectric spectra is analyzed resulting petroleum products dielectric property, makes up the mathematical model that petroleum products is discerned with chemometrics method, has realized harmless, discriminating fast and effectively to petroleum products.
The present invention is a kind of petroleum products variety discrimination method that is based upon variety classes petroleum products sample and fastens from pass between the dielectric spectra of sample, and it comprises two stages of evaluation of setting up petroleum products variety discriminating model and unknown petroleum products sample.Setting up the petroleum products variety discriminating model stage: must collect the petroleum products of various known kinds, and obtain the dielectric spectra of these petroleum productss; The dielectric spectra of the petroleum products sample by collecting known kind is set up the petroleum products variety discriminating model; In fact to the checkout procedure of discriminating model is exactly to confirm that each petroleum products in the model all can only be by only evaluation, and can not obscure mutually with other kind petroleum productss.Validation phase at unknown petroleum products sample: the dielectric spectra of unknown petroleum products sample obtains the back and just can differentiate it; The dielectric spectra of unknown petroleum products sample can compare with the standard dielectric spectra of each petroleum products sample in the discriminating model; If in unknown petroleum products sample and the discriminating model certain-distance of petroleum products sample is less than this standard dielectric spectra corresponding threshold, so this unknown petroleum products sample just with discriminating model in this known petroleum products belong to the petroleum products of same kind.
The beneficial effect that the present invention compared with prior art has is:
1, utilize the present invention to differentiate petroleum products variety, it is fast that it analyzes discriminating speed.The mensuration process of dielectric spectra generally can be finished in 2 minutes;
2, do not use any chemical reagent, reduced the detection cost, also free from environmental pollution;
3, compare with chemical method, systematic error and personal error reduce, and have improved measuring accuracy;
4, compare with spectral technique, instrumentation is easy, and instrument is cheap;
5, can handle the great amount of samples analysis, save time;
6, can can't harm discriminating to analyzing samples;
Kind discriminating model among 7 the present invention can be carried out kind as gasoline, kerosene, diesel oil, I. C. engine oil, gear oil, hydraulic pressure wet goods and differentiate that it is applied widely the petroleum products of petroleum hydrocarbon.
Description of drawings
Fig. 1 is the dielectric spectra curve map of variety classes gasoline sample among the embodiment 1;
Fig. 2 is the major component 1 of variety classes gasoline sample correspondence among the embodiment 1 and the shot chart of major component 2.
In Fig. 1,1 for not adding the gasoline sample of pure and mild ether oxygenated additive; 2 for adding the gasoline sample of ethers oxygenated adjuvant; 3 for adding the gasoline sample of alcohols oxygenated additive; 4 for adding the gasoline sample of pure and mild ethers oxygenated adjuvant.
Embodiment
Further specify the present invention below by drawings and Examples, but the present invention is not limited only to these embodiment.
Gasoline product is added with oxygenatedchemicals usually to improve its octane value at present, and oxygenatedchemicals is mainly ether compound and alcohol compound.
The gasoline classification that is added with different oxygenatedchemicalss with discriminating in the present embodiment is that example illustrates the present invention.
Have 94 gasoline samples, 68 gasoline composition of sample training sets (as table 1) wherein, 26 gasoline composition of sample test sets.
The brief introduction of four kind gasoline of table 1. training set sample
| Sample number into spectrum | The classification numbering | The classification explanation |
| 1-46 | 1 | Do not add oxygenated additive |
| 47-61 | 2 | Add ethers oxygenated adjuvant |
| 63-67 | 3 | Add the alcohols oxygenated |
| 68 | 4 | Add pure and mild ethers oxygenated adjuvant |
1, the foundation of standard dielectric spectra collection
Gasoline sample is carried out the dielectric spectra analysis, obtain the dielectric spectra spectrogram as shown in Figure 1.
Analysis condition:
Driving voltage: 5V; Excitation waveform: triangular wave;
Excitation frequency scope: 50Hz～16MHz; Adopt an interval: 100Hz
Sample cell temperature: 25 ℃
2, dielectric spectra Feature Extraction
Use principal component analysis (PCA) and extract above-mentioned dielectric spectra principal character, as seeing the dendrogram of variety classes petroleum products on the principal component scores figure among Fig. 2.
3, the foundation of discriminating model:
Set up discriminating model with the SIMCA clustering algorithm.Table 2 is a SIMCA method modeling result.There is not the management cluster analysis after at first 68 samples being adopted the major component compression.Adopt the major component compression to set up class models respectively in the sample of each classification according to classification results then.At last 26 gasoline samples of unknown sample collection are carried out classification analysis.The cluster effect of major component 1 and 2 pairs of four kinds of petroleum productss of major component as shown in Figure 2.
Table 2 SIMCA method modeling result
| Sample number into spectrum | The classification numbering | The classification explanation | Identification classification numbering |
| 1-46 | 1 | Do not add pure and mild ethers oxygenated |
1 |
| 47-61 | 2 | Add ethers oxygenated |
2 |
| 63-67 | 3 | Add the alcohols oxygenated |
3 |
| 68 | 4 | Add pure and mild ethers oxygenated |
4 |
4, the unknown sample kind is differentiated:
Set up training set with 68 gasoline samples, set up the standard database of training set, adopt the mahalanobis distance method, 26 gasoline samples of the unknown are differentiated that identification result is as shown in table 3 with major component 1 and major component 2.
The identification result of 26 samples of table 3 pair the unknown
Remarks: numbering 1 is not for there being oxygenated additive; Numbering 2 is for adding ether compound; Numbering 3 is for adding alcohol compound.
This discriminating model is 100% to the kind resolution of unknown 26 gasoline samples, illustrates that it is accurately feasible utilizing the harmless petroleum products variety of differentiating of the present invention.
Embodiment 2:
In the present embodiment to differentiate that the variety classes petroleum products is that example illustrates the present invention.
39 lubricating oil samples are arranged, and are respectively the I. C. engine oils of hydraulic oil, hydrodynamic drive liquid, gear oil and the different quality grade and the viscosity grade of different manufacturers.
Wherein:
28 lubricating oil are set up training set such as table 4, and 11 lubricating oil samples are as test set.
28 lubricating oil samples of table 4 situation
| Sequence number | The training set |
| 4 in hydraulic oil and hydrodynamic drive liquid sample | L_HM100 hydraulic oil; The 10# aircraft fluid; HM_32_ hydraulic oil; 8# hydrodynamic drive liquid |
| 3 in gear oil sample | Yanshan Mountain GL_5_85W/90; Blue refining GL_5_85W/90; GL_5_80W/90 |
| 21 in I. C. engine oil sample | SJ 15W/50；CH4/SJ 15W/40；SF 15W/40；SG/CD 15W/40；SJ/CF 15W/50；SF/CC 50；SD/CC 10W/30； SD/CC 15W/40；SF/CC 15W/40；SD/CC 10W/30；CD 40； CD 30；CH4/SJ 15W/40；CC 50；CF_4 15W/40；CD 40； CC 40；CD 30；SH/CD 10W/40；SJ/CF 10W/40；SJ 15W/50 |
The lubricating oil sample dielectric spectra is analyzed:
Analysis condition: driving voltage: 5V; Excitation waveform: sine wave;
Excitation frequency scope: 50Hz～16MHz; Adopt an interval: 100Hz
Sample cell temperature: 25 ℃
The foundation of discriminating model:
Respectively 28 lubricating oil samples are carried out mensuration twice, with the first time measurement result set up the discriminating model storehouse as training set, discern as inspection set with predicting the outcome for the second time, obtain identification error, identification error be multiply by 2, obtain the threshold value of discriminating model, its discriminating model storehouse the results are shown in Table 5.
The assay of table 5 pair 28 lubricating oil samples
| Master sample | Identification error | Threshold value |
| L_HM100 hydraulic oil | 0.0887 | 0.1774 |
| The 10# aircraft fluid | 0.0898 | 0.1796 |
| 8# hydrodynamic drive liquid | 0.1290 | 0.258 |
| HM_32_ hydraulic oil | 0.1166 | 0.2332 |
| I. C. engine oil _ Mobil_Super SJ 15W/50 | 0.4318 | 0.8636 |
| I. C. engine oil _ Mobil_CH4/SJ 15W/40 | 0.1145 | 0.229 |
| I. C. engine oil _ Shell_Helix_SF 15W/40 | 0.1167 | 0.2334 |
| I. C. engine oil _ Shell_Helix_SG/CD 15W/40 | 0.3197 | 0.6394 |
| I. C. engine oil _ Shell_Helix_SJ/CF 15W/50 | 0.1075 | 0.215 |
| I. C. engine oil _ RotellaDX40_SF/CC 50 | 0.2206 | 0.4412 |
| I. C. engine oil _ SD/CC 10W/30 | 0.0948 | 0.1896 |
| I. C. engine oil _ SD/CC 15W/40 | 0.1091 | 0.2182 |
| I. C. engine oil _ SF/CC 15W/40 | 0.0979 | 0.1958 |
| I. C. engine oil _ SD/CC 10W/30 | 0.1056 | 0.2112 |
| I. C. engine oil _ orchid refining CD 40 | 0.1197 | 0.2394 |
| I. C. engine oil _ CD 30 | 0.0620 | 0.124 |
| I. C. engine oil _ CH4/SJ 15W/40 | 0.0907 | 0.1814 |
| I. C. engine oil _ CC 50 | 0.1065 | 0.213 |
| I. C. engine oil _ CF_4 15W/40 | 0.0919 | 0.1838 |
| I. C. engine oil _ CD 40 | 0.1119 | 0.2238 |
| I. C. engine oil _ CC 40 | 0.1017 | 0.2034 |
| I. C. engine oil _ perseverance fortune _ CD 30 | 0.0994 | 0.1988 |
| I. C. engine oil _ Esso_UNF_SH/CD 10W/40 | 0.1199 | 0.2398 |
| Gear oil _ Yanshan Mountain _ GL 5 85W/90 | 0.1075 | 0.215 |
| Gear oil _ orchid refining _ GL_5 85W/90 | 0.0958 | 0.1916 |
| Gear oil _ GL_5_80W/90 | 0.0706 | 0.1412 |
| I. C. engine oil _ Esso_Ultraflo_SJ/CF 10W/40 | 0.0952 | 0.1904 |
| I. C. engine oil _ |
0.0909 | 0.1818 |
11 lubricating oil samples of test set are carried out kind identification, and its assay sees Table 6:
The assay of table 6 pair 11 lubricating oil samples
| Master sample | Identification result | Error | Remarks |
| L_HM100 hydraulic oil | Hydraulic oil | 0.1181 | Correctly |
| The 10# aircraft fluid | Hydraulic oil | 0.1227 | Correctly |
| 8# hydrodynamic drive liquid | Hydrodynamic drive liquid | 0.1785 | Correctly |
| HM_32_ hydraulic oil | Hydraulic oil | 0.1614 | Correctly |
| I. C. engine oil _ Mobil_Super SJ 15W/50 | I. C. engine oil SJ 15W/50 | 0.4317 | Correctly |
| I. C. engine oil _ Mobil_CH4/SJ 15W/40 | I. C. engine oil CH4 15W/40 | 0.1517 | Correctly |
| I. C. engine oil _ Shell_Helix_SF 15W/40 | I. C. engine oil SF 15W/40 | 0.1526 | Correctly |
| I. C. engine oil _ Shell_Helix_SG/CD 15W/40 | I. C. engine oil _ SG/CD 15W/40 | 0.4646 | Correctly |
| Gear oil _ Yanshan Mountain _ GL_5 85W/90 | Gear oil _ GL_5 85W/90 | 0.1508 | Correctly |
| Gear oil _ orchid refining _ GL_5 85W/90 | Gear oil GL_5 85W/90 | 0.1422 | Correctly |
| Gear oil _ GL_5_80W/90 | Gear oil _ GL_5_80W/90 | 0.1044 | Correctly |
All the other steps are with embodiment 1.
Claims (8)
1, a kind of method of differentiating petroleum products variety, it is characterized in that: adopt the dielectric spectra analysis to obtain the finger print information of known petroleum products dielectric property, make up the mathematical model of petroleum products identification with chemometrics method, the sample of unknown petroleum products is carried out comparing with described petroleum products variety discriminating model after the dielectric spectra analysis identical with the sample of described various known petroleum products varieties, realize discriminating petroleum products.
2, the method for discriminating petroleum products variety as claimed in claim 1 is characterized in that: it is to carry out the dielectric spectra analysis and set up the petroleum products variety discriminating model by the sample to various known petroleum products varieties; Sample with unknown petroleum products carries out comparing with described petroleum products variety discriminating model after the dielectric spectra analysis identical with the sample of described various known petroleum products varieties then; If the standard dielectric spectra of the sample of a certain known petroleum products variety distance is less than the standard dielectric spectra corresponding threshold of the sample of this known petroleum products variety in the dielectric spectra of the sample of described unknown petroleum products and the described petroleum products variety discriminating model, the sample of so described unknown petroleum products just with described petroleum products variety discriminating model in the sample of this known petroleum products variety belong to the petroleum products of same kind, thereby differentiated the kind of described unknown petroleum products sample.
3, the method for discriminating petroleum products variety as claimed in claim 2 is characterized in that: it is to carry out as follows: 1), carry out the dielectric spectra analysis and set up the standard dielectric spectra collection of the sample of various known petroleum products varieties by the sample to various known petroleum products varieties; 2), the standard dielectric spectra collection of the sample of various known petroleum products varieties is proofreaied and correct and pre-service; 3), to the extraction of the dielectric spectra characteristic information of the sample of various known petroleum products varieties; 4), set up the discriminating model of the unknown petroleum products specimen types of identification with the dielectric spectra characteristic information of the sample of various known petroleum products varieties; 5), unknown petroleum products sample and known petroleum products sample being carried out kind relatively differentiates.
4, the method for discriminating petroleum products variety as claimed in claim 3, it is characterized in that: the described dielectric spectra collection of setting up various known petroleum products variety samples is meant that the sample to various known petroleum products varieties carries out the dielectric spectra analysis, promptly measures the standard dielectric spectra of various known petroleum products variety samples; The standard dielectric spectra of described various known petroleum products variety samples is meant to its dielectric spectra of twice duplicate measurements of different types of sample and with the standard dielectric spectra of its average dielectric spectra as this kind sample, thereby obtains the dielectric spectra spectrogram of various known petroleum products variety samples; It is to adopt any one or any two kinds or any three kinds of dielectric spectra preprocess methods in level and smooth or centralization or differentiate or the normalization that the dielectric spectra collection of the sample of described various known petroleum products varieties is proofreaied and correct that the dielectric spectra collection of described sample to various known petroleum products varieties is proofreaied and correct with pre-service; Principal component analysis (PCA) or partial least square method or wavelet analysis are adopted in the extraction of the dielectric spectra characteristic information of the sample of described various known petroleum products varieties; It is to adopt mode identification method that described dielectric spectra feature with various known petroleum products variety samples is set up the petroleum products variety discriminating model, promptly with distance discrimination analysis or cluster analysis or even linear multivariate analysis method of statistics or partial least-square regression method.
5, the method for discriminating petroleum products variety as claimed in claim 4, it is characterized in that: describedly carry out discriminatory analysis with pattern-recognition, the dielectric spectra that is about to the sample of various known petroleum products varieties is divided into training set and inspection set two parts, and wherein training set should be identical with kind of petroleum products kind in the inspection set; Then various petroleum productss are set up the petroleum products variety discriminating model according to dielectric spectra information; Estimate the performance of discriminating model at last with inspection set.
6, the method of discriminating petroleum products variety as claimed in claim 3, it is characterized in that: it is that unknown petroleum products sample is carried out the dielectric spectra analysis identical with known petroleum products variety sample that the kind of described unknown petroleum products sample is differentiated, the extraction of dielectric spectra correction and pre-service and dielectric spectra characteristic information, compare with described petroleum products variety discriminating model then, if the distance of a certain known petroleum products variety sample is less than this known petroleum products variety sample corresponding threshold in described unknown petroleum products sample and the described discriminating model, so described unknown petroleum products sample just with described discriminating model in the sample of this known petroleum products variety belong to the petroleum products of same kind, thereby differentiated the kind of described unknown petroleum products sample.
7, as the method for the described discriminating petroleum products variety of the arbitrary claim of claim 1～6, it is characterized in that: during described dielectric spectra is analyzed, the driving voltage amplitude range is 1V～15V, frequency range is 50Hz～16MHz, excitation waveform be triangular wave or/and sine wave or/and square wave is adopted an interval: 10Hz～100Hz.
8, as the method for the described discriminating petroleum products variety of the arbitrary claim of claim 1～6, it is characterized in that: described petroleum products variety is to be the various petroleum hydrocarbon fuel and the lubricating oil of principal ingredient with the petroleum hydrocarbon.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710078320 CN101021495A (en) | 2007-03-23 | 2007-03-23 | Method for distinguishing petroleum products variety |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710078320 CN101021495A (en) | 2007-03-23 | 2007-03-23 | Method for distinguishing petroleum products variety |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101021495A true CN101021495A (en) | 2007-08-22 |
Family
ID=38709329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710078320 Pending CN101021495A (en) | 2007-03-23 | 2007-03-23 | Method for distinguishing petroleum products variety |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101021495A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101949875A (en) * | 2010-08-12 | 2011-01-19 | 张爱民 | Method and device for detecting concentration of organic solution |
| CN102288573A (en) * | 2011-05-27 | 2011-12-21 | 中国人民解放军总后勤部油料研究所 | Method for fast recognizing fuel type and designation of engine by use of mid-infrared spectrum technique |
| CN101339150B (en) * | 2007-11-19 | 2012-11-21 | 冯新泸 | Method for determining octane number based on dielectric spectra technology |
| CN103440425A (en) * | 2013-09-05 | 2013-12-11 | 中国石油化工股份有限公司 | Establishment method of octane value regression model |
| CN103499611A (en) * | 2013-10-16 | 2014-01-08 | 管亮 | Diesel cetane number measuring method based on dielectric spectrum technology |
| CN105181634A (en) * | 2015-08-25 | 2015-12-23 | 江南大学 | Petroleum fractionate tower product analysis method based on near infrared technology |
-
2007
- 2007-03-23 CN CN 200710078320 patent/CN101021495A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339150B (en) * | 2007-11-19 | 2012-11-21 | 冯新泸 | Method for determining octane number based on dielectric spectra technology |
| CN101949875A (en) * | 2010-08-12 | 2011-01-19 | 张爱民 | Method and device for detecting concentration of organic solution |
| CN102288573A (en) * | 2011-05-27 | 2011-12-21 | 中国人民解放军总后勤部油料研究所 | Method for fast recognizing fuel type and designation of engine by use of mid-infrared spectrum technique |
| CN103440425A (en) * | 2013-09-05 | 2013-12-11 | 中国石油化工股份有限公司 | Establishment method of octane value regression model |
| CN103440425B (en) * | 2013-09-05 | 2016-07-06 | 中国石油化工股份有限公司 | A kind of method for building up of octane value regression model |
| CN103499611A (en) * | 2013-10-16 | 2014-01-08 | 管亮 | Diesel cetane number measuring method based on dielectric spectrum technology |
| CN105181634A (en) * | 2015-08-25 | 2015-12-23 | 江南大学 | Petroleum fractionate tower product analysis method based on near infrared technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103278609B (en) | A kind of Examination of Meat Fresh Level merged based on multi-source perception information | |
| CN101995389B (en) | Method for fast recognition of crude oil variety through near infrared spectrum | |
| CN101021495A (en) | Method for distinguishing petroleum products variety | |
| CN102778442B (en) | Method for rapidly identifying type of balsam material liquid for cigarette | |
| CN101782511B (en) | Method for rapidly identifying type and grade of lubricating oil | |
| CN1831516A (en) | Method for nondistructive discriminating variety and true and false of cigarette using visible light and near-infrared spectrum technology | |
| CN102269720A (en) | Method for identifying edible oil quality through combination of nuclear magnetic resonance and pattern recognition method | |
| CN102841069B (en) | Method for rapidly identifying types of crude oil by using mid-infrared spectrum | |
| CN101339150B (en) | Method for determining octane number based on dielectric spectra technology | |
| CN105223240A (en) | A kind of method utilizing detection by electronic nose crab freshness | |
| CN108845045A (en) | A kind of method that gas-phase fingerprint pattern combination principal component analytical method differentiates frying oil quality | |
| CN102338743B (en) | Mid-infrared spectrum method for identifying engine fuel type and brand | |
| CN101900672A (en) | Method for quickly identifying class and viscosity grade of lubricating oil | |
| CN101799411A (en) | Quick detection method of oil quality spectrum for lubricating oil | |
| CN105973861B (en) | The method of marine oil overflow type is differentiated based on oil product fluorescent characteristic Fisher diagnostic methods | |
| CN112362609A (en) | Method for identifying oil stain smoke pollution source based on infrared spectrum technology | |
| CN111272686A (en) | Hyperspectral detection method for iron grade of iron ore mineral powder | |
| CN110398472A (en) | A kind of vegetable oil identification classification method based on two-dimensional correlation spectra quantization | |
| CN105954228A (en) | Method for measuring content of sodium metal in oil sand based on near infrared spectrum | |
| CN109374565A (en) | A kind of methanol gasoline ethanol petrol differentiates and content assaying method | |
| CN108827925A (en) | Edible vegetable oil true and false rapid detection method and detection device based on optical fiber type fluorescence spectroscopy technique | |
| CN102330553B (en) | Method for identifying dynamic spectrum fluid for MDI (Manual Data Input) test | |
| CN102221534B (en) | Mid-infrared spectrum method for quickly identifying engine fuel type | |
| CN109598245B (en) | Edible oil transverse relaxation attenuation curve signal feature extraction method based on 1D-CNN | |
| CN105372224A (en) | Method for identifying different species of feed grease based on Fourier Ramman spectrum |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20070822 |