CN107389645A - The method that the Fisher models of wavelet transform parsing oil product fluorescent characteristic differentiate marine oil overflow - Google Patents
The method that the Fisher models of wavelet transform parsing oil product fluorescent characteristic differentiate marine oil overflow Download PDFInfo
- Publication number
- CN107389645A CN107389645A CN201710692532.XA CN201710692532A CN107389645A CN 107389645 A CN107389645 A CN 107389645A CN 201710692532 A CN201710692532 A CN 201710692532A CN 107389645 A CN107389645 A CN 107389645A
- Authority
- CN
- China
- Prior art keywords
- oil
- sample
- fisher
- wavelet
- marine
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000002189 fluorescence spectrum Methods 0.000 claims abstract description 16
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims abstract description 5
- 239000003921 oil Substances 0.000 claims description 52
- 239000000295 fuel oil Substances 0.000 claims description 24
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000002405 diagnostic procedure Methods 0.000 abstract description 7
- 239000000284 extract Substances 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 2
- 238000003912 environmental pollution Methods 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000012827 research and development Methods 0.000 abstract description 2
- 239000010779 crude oil Substances 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 8
- 239000000047 product Substances 0.000 description 6
- 238000000605 extraction Methods 0.000 description 4
- 150000001875 compounds Chemical group 0.000 description 3
- 239000000090 biomarker Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000003153 chemical reaction reagent Substances 0.000 description 2
- 238000001506 fluorescence spectroscopy Methods 0.000 description 2
- VLKZOEOYAKHREP-UHFFFAOYSA-N hexane Substances CCCCCC VLKZOEOYAKHREP-UHFFFAOYSA-N 0.000 description 2
- 238000002203 pretreatment Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 239000004215 Carbon black (E152) Substances 0.000 description 1
- WEEGYLXZBRQIMU-UHFFFAOYSA-N Eucalyptol Chemical class C1CC2CCC1(C)OC2(C)C WEEGYLXZBRQIMU-UHFFFAOYSA-N 0.000 description 1
- 150000001335 aliphatic alkanes Chemical class 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009514 concussion Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 238000010813 internal standard method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000003305 oil spill Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 125000005575 polycyclic aromatic hydrocarbon group Chemical group 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000002798 spectrophotometry method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 150000003431 steroids Chemical class 0.000 description 1
- 239000006228 supernatant Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6402—Atomic fluorescence; Laser induced fluorescence
Landscapes
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Optics & Photonics (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The present invention relates to the method that the Fisher models of wavelet transform parsing oil product fluorescent characteristic differentiate marine oil overflow, belong to marine environmental pollution monitoring and improvement field.The present invention carries out 6 layers of decomposition to oil sample fluorescence spectra by db7 wavelet basis functions, extracts d3 detail coefficients features, the quantitative formula for differentiating marine oil overflow is established with reference to Fisher diagnostic methods, and verified by actual sample.The result that this method quantifies differentiates that accuracy is higher for oil sample, the defects of overcoming current Qualitive test marine oil overflow, and for research and development are real-time, portable instrument of quick discriminating marine oil overflow provides a foundation.
Description
Technical field
The method that the Fisher models discriminating marine oil overflow of oil product fluorescent characteristic is parsed the present invention relates to wavelet transform,
Oil sample fluorogram is decomposed using wavelet transform more particularly to one kind, with reference to Fisher diagnostic methods quick discriminating sea
The fuel oil and crude oil of upper oil spilling and the method for further discriminating between middle-eastern crude, belong to marine environmental pollution monitoring and led with administering
Domain.
Background technology
With the development of marine transportation industry, oil spill accident happens occasionally, and not only threatens marine ecosystems, also the mankind are good for
Kang Zaocheng is seriously endangered.So promptly and accurately differentiating oil spilling source, taking urgent measure to protect the marine environment seems particularly heavy
Will[1]。
Oil be by different compound groups into complex mixture, domestic and foreign scholars by the n-alkane among oil sample,
The multiple compounds of polycyclic aromatic hydrocarbon and biomarker etc. 100 carry out discriminatory analysis to oil sample[2-5].Wang Xin equalitys[6]Pass through gas phase color
Internal standard method establishes n-alkane in crude oil to spectrum, the analysis method of biomarker (steroid, terpane class) differentiates to crude oil sample；
In view of contain abundant compound fragrant hydrocarbon in oil[7], and fluorescence spectrophotometry have high-resolution, high sensitivity,
The features such as sample pre-treatments are simple[8], the research of oil product fluorescent characteristic is of great interest.Wang etc.[9]With based on difference
Synchronous fluorimetric method under concentration, oil sample is entered respectively using principal component analysis, Partial Least Squares Method, Gobar conversion
Row feature extraction, contrasts the spectral signature between oil spilling sample and doubtful oil spilling source, total accuracy is respectively 77%, 79%,
92%, but Gobar conversion can not extract some abrupt informations and unstable information, information extraction is imperfect.It is and discrete small
Ripple (DWT) is by specific flexible and shift factor and selects suitable wavelet basis function to handle primary signal, can produce
The approximation coefficient of raw reflection primary signal large scale information and the detail coefficients of smaller scale information[10]。
Bibliography：
[1] Zhao Yunying, the progress marine environment science of development in oil spills identification by fluorescence spectroscopy, 1997,2 (16)
29-35
[2]Merv Fingas The basics of spill cleanup 2nd[M].New York:Lewis
Publishers,2001
[3]Wang Zhendi,Fingas M,Page D S.Oil spill identification.Journal of
Chromatography A,1999,843(1/2):369-411
[4]Ebrahimi D,Lia J,Hibbert D B.Classification of weathered petroleum
oils by multi-way of gas chromatography-mass spectrometry data using PARAFAC2
parallel factor analysis,Journal of Chromatography A,2007,1166(1/2):163-170
[5] Liu Xiaoxing, Sun Huiqing, Wang Yahui, high Yao distribution of normal alkanes characteristic differentiations sea mixed crude [J] rings are accorded with
Border chemistry, 2016,35 (2):305-310.
[6] Wang Xinping, Sun Peiyan, Zhou Qing, Li Mei, Cao Lixin, Zhao Yuhui, the internal standard method analysis of crude oil saturated hydrocarbons fingerprint
[J] analytical chemistry 2007,35 (8):1121-1126.
[7]Weng N,Wan S,Wang H,et al.Journal of Chromatography A,2015,1398:94
~107.
[8]Greene L V,Elzey B,Franklin M,et al.Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy,2017,174:316~325.
[9]Wang C,Shi X,Li W,et al.Marine Pollution Bulletin,2016,104(1-2):
322~328.
[10]Ha D,Park D,Koo J,et al.Computers&Chemical Engineering,2016,94:
362~369.
The content of the invention
The present invention establishes a kind of quick discriminating fuel oil and crude oil simultaneously using wavelet transform combination Fisher diagnostic methods
The method for further discriminating between middle-eastern crude, 6 layers of decomposition are carried out to oil sample fluorescence spectra by db7 wavelet basis functions, extract d3
Detail coefficients feature, the quantitative formula for differentiating marine oil overflow is established with reference to Fisher diagnostic methods, and tested by actual sample
Card.The result that this method quantifies differentiates that accuracy is higher for oil sample, the defects of overcoming current Qualitive test marine oil overflow, and
A foundation is provided to research and develop real-time, quick discriminating marine oil overflow portable instrument.
The method that the Fisher models of wavelet transform parsing oil product fluorescent characteristic differentiate marine oil overflow, including following steps
Suddenly：
1. carrying out 6 layers of decomposition to oil sample fluorescence spectra using db7 wavelet basis functions, d3 detail coefficients features are extracted, are obtained
Testing sample 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm, five characteristic wave strong points small echo
Coefficient X1~X5；
2. the wavelet coefficient of 1. five characteristic wave strong points that step is obtained brings following Fisher discrimination formulas Y into1And Y2
In,
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5
Wherein, X1~X5Represent d3 at 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm respectively
Wavelet coefficient.
3. 2. sample (Y that calculation procedure obtains1, Y2) Euclidean distance between value and each group barycenter judges to belong to, judge to advise
It is then：Smaller with certain group centroid distance, sample then belongs to this group of classification,
Wherein, the group barycenter of fuel oil is O1(Y1=3.293, Y2=0.211)；The group barycenter of middle-eastern crude is O2(Y1=-
2.114 Y2=0.769)；The group barycenter of non-middle-eastern crude is O3(Y1=-0.825, Y2=-0.505).
" d3 detail coefficients " of the present invention are that 6 layers of decomposition are carried out to oil sample fluorogram by db7 wavelet basis functions, will
Original spectrogram resolves into 6 layers of approximation coefficient and 6 layers of detail coefficients, and the present invention chooses the 3rd layer of 6 layers of detail coefficients, due to details
It is " d " that coefficient, which is write a Chinese character in simplified form, and the 3rd layer of detail coefficients herein write d3.
Preferred steps of the present invention 1. in, the fluorescence spectra of the sample utilizes permanent wavelength method measure.
Preferred steps of the present invention 1. in, the fluorescence spectra of the sample is measured by molecular fluorescence spectrophotometer.
Step of the present invention 2. described in discrimination formula utilize Fisher diagnostic methods obtain.Fisher diagnostic methods are by variance
The method that the thought of analysis constructs a linear discriminant function, this method to distinguish maximum between all kinds of packets, and makes packet
Internal deviation is minimum, and the purpose of classification is reached with this.Oil sample is divided into three classes by the present invention：Fuel oil, middle-eastern crude, the non-Middle East
Crude oil.
Beneficial effects of the present invention are：The present invention carries out details using wavelet transform to the fluorescence spectrum information of oil sample
Coefficient is extracted, and the Fisher discrimination models for differentiating oil product are established with this, and the model, can while fuel oil and crude oil is distinguished
Middle-eastern crude is further discriminated between.The model established has higher discriminating accuracy, such as discriminating to non-modeling oil sample just
True rate is 95.7%, and it is also applied for the discriminating of weathering oil product.The present invention discriminating formula can quickly, quantitatively differentiate sea
The fuel oil and crude oil of oil spilling simultaneously further discriminate between middle-eastern crude, to realize that online, discriminating marine oil overflow is portable in real time from now on
The research and development of formula instrument provide a theoretical foundation.
Brief description of the drawings
Fig. 1 is that embodiment 1 addresses non-No. 16 floor exploded view of weathering light-weight fuel oil；
Fig. 2 is that embodiment 1 addresses No. 16 floor exploded view of light-weight fuel oil after weathering；
Fig. 3 is that embodiment 2 addresses 6 layers of exploded view of non-weathering United Arab Emirates crude oil；
Fig. 4 is that embodiment 2 addresses 6 layers of exploded view of United Arab Emirates' crude oil after weathering；
Fig. 5 is that embodiment 3 addresses non-6 layers of exploded view of weathering Daqing crude oil；
Fig. 6 is that embodiment 3 addresses 6 layers of exploded view of Daqing crude oil after weathering.
Embodiment
Following non-limiting examples can make one of ordinary skill in the art be more fully understood the present invention, but not with
Any mode limits the present invention.
Test method described in following embodiments, it is conventional method unless otherwise specified；The reagent and material, such as
Without specified otherwise, commercially obtain.
Discrimination formula of the present invention is established according to the wavelet coefficient of oil sample, specific as follows：
1st, optimization extraction modeling parameters
6 layers of decomposition are carried out to oil sample fluorescence spectra using db7 wavelet basis functions, extract d3 detail coefficients features, extraction
255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, the wavelet coefficient at 354 ± 2nm are modeled；
2nd, fluorescence spectra determines
Using molecular fluorescence spectrometry, using the above-mentioned required so-called fluorescence spectra of decomposition of permanent wavelength method measure.
3rd, the foundation of discrimination model
Using 8 kinds of fuel oil, 14 kinds of non-middle-eastern crudes and 7 kinds of middle-eastern crudes as analysis object, all analysis objects are determined
Fluorescence spectra, carry out wavelet transform to it, obtain 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354
Wavelet coefficient combination Fisher diagnostic methods at ± 2nm establish model.
Following is the implementation example that a model is established：
Using 8 kinds of fuel oil, 14 kinds of non-middle-eastern crudes and 7 kinds of middle-eastern crudes as analysis object, all analysis objects are determined
Fluorescence spectra.The fluorescence intensity of oil sample fluorescence spectra is imported into Matlab R2012b, in Matlab R2012b
Command Window input " wavemenu " recall wavelet toolbox, " Wavelet 1-D " enter window, in the window for selection
Selection File → Load → Signal recalls the fluorescence intensity imported, and Wavelet selections db7, Level select 6 layers, with to spectrum
Figure carries out wavelet decomposition.Extract 255 ± 2nm of d3 layers, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm wavelet systems
Number is used as modeling parameters, carries out Fisher discriminant analyses.Utilize (the Statistical Analysis System of SAS 9.1
(SAS) application software) Fisher differentiations are carried out, the group of fuel oil is set to 0, and middle-eastern crude group is set to 1, non-middle-eastern crude group
It is not set to 2.
Program is as follows：
Data dataset names
255 ± 2nm of Input, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm wavelet coefficient grouping；
Datalines；
Input data ...
；
The above-mentioned dataset name out=output canonical variable data names (present invention Y of Proc Candisc data=
Represent)；
Class grouping；
255 ± 2nm of Var, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm wavelet coefficient；
Run；
The equation coefficients drawn by Candisc processes；
Y1Factor alpha1For (- 0.396 0.384-0.142-0.167 0.151)
Y2Factor alpha2For (0.207-0.103 of 0.803-0.314 0.114)；
Participate in 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, the 354 ± 2nm of 29 kinds of oil sample d3 layers of modeling
The average value β of wavelet coefficient be (11.882 28.426 8.060 9.707 7.929).Its formula constant term C=- αi*βT, i
=1,2；Being computed two formula is：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5
Wherein Y1、Y2For Fisher discrimination formulas, X1~X5Represent d3 layers 255 ± 2nm, 280 ± 2nm, 302 ± 2nm,
Wavelet coefficient at 332 ± 2nm, 354 ± 2nm.
4th, the accuracy of model is verified
In order to verify that the oil sample that the present invention establishes differentiates that model is former to the fuel oil of marine oil overflow, middle-eastern crude and the non-Middle East
The accuracy that oil differentiates, verified that total accuracy is 96.6% with 29 kinds of modeling oil samples after short-term weathering 30 days.In addition,
Also verified with 23 kinds of non-modeling oil samples, up to 95.7%, its accuracy rate differentiated is higher than is returned total accuracy using binary linearity
Return the model of establishing equation.
Oil sample fluorescence spectra is determined using molecular fluorescence spectrometry in following embodiments, and selected instrument and method are such as
Under：
Instrument and reagent：Analytical instrument is Cary Eclipse types Fluorescence spectrophotometer (Varian companies of the U.S.)；
Solvent：N-hexane (chromatographically pure, German Merck)；
Sample pre-treatments：With beaker weigh (0.05 ± 0.0002) g samples oil, by acquired oil sample be dissolved in 10mL just oneself
Alkane, concussion is to being completely dissolved, then static 5min, pipettes 40uL supernatants in test tube, adds 10mL n-hexanes, continues to employ to be measured.
Embodiment 1：With No. 1 checking of light-weight fuel oil before and after weathering
Table 1：Wavelet coefficient at non-No. 1 d3 floor five of weathering light-weight fuel oil
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 19.485 | 42.393 | 8.420 | 4.086 | 3.221 |
Wavelet systems numerical tape in table 1 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=2.531
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=1.087
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
1.161、4.656、3.715.It can thus be appreciated that and O1Closest, this oil sample is fuel oil, is met with known case.
Table 2：Wavelet coefficient at No. 1 d3 floor five of light-weight fuel oil after weathering
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 16.789 | 35.860 | 8.120 | 6.602 | 5.040 |
Wavelet systems numerical tape in table 2 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=0.987
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=1.272
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
2.538、3.142、2.982.It can thus be appreciated that and O1Closest, this oil sample is fuel oil, is met with known case.
Embodiment 2：Verified with United Arab Emirates' crude oil before and after weathering
Table 3：Wavelet coefficient at non-weathering United Arab Emirates crude oil d3 layers five
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 12.628 | 24.524 | 8.163 | 10.717 | 7.379 |
Wavelet systems numerical tape in table 3 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=-2.058
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=2.107
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
5.677、1.340、2.890.It can thus be appreciated that and O2Closest, this oil sample is middle-eastern crude, is met with known case.
Table 4：Wavelet coefficient at non-weathering United Arab Emirates crude oil d3 layers five
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 11.498 | 21.365 | 5.063 | 7.466 | 5.488 |
Wavelet systems numerical tape in table 4 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=-2.126
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=1.360
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
5.540、0.591、2.273.It can thus be appreciated that and O2Closest, this oil sample is middle-eastern crude, is met with known case.
Embodiment 3：Verified with the Daqing crude oil before and after weathering
Table 5：Wavelet coefficient at non-weathering Daqing crude oil d3 layers five
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 7.551 | 16.640 | 4.820 | 6.090 | 6.393 |
Wavelet systems numerical tape in table 5 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=-1.977
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=-0.732
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
5.353、1.501、1.174.It can thus be appreciated that and O3Closest, this oil sample is non-middle-eastern crude, is met with known case.
Table 6：Wavelet coefficient at Daqing crude oil d3 layers five after weathering
| λ/nm | 255±2 | 280±2 | 302±2 | 332±2 | 354±2 |
| Wavelet coefficient | 7.379 | 16.669 | 4.756 | 5.031 | 5.618 |
Wavelet systems numerical tape in table 6 is entered into following calculation formula：
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5=-1.828
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5=-1.026
(Y1, Y2) with the group barycenter O of fuel oil1(Y1=3.293, Y2=0.211), the group barycenter O of middle-eastern crude2(Y1=-
2.114 Y2=0.769), the group barycenter O of non-middle-eastern crude3(Y1=-0.825, Y2=-0.505) Euclidean distance is respectively
5.268、1.817、1.131.It can thus be appreciated that and O3Closest, this oil sample is non-middle-eastern crude, is met with known case.
Claims (4)
1. the method that the Fisher models of wavelet transform parsing oil product fluorescent characteristic differentiate marine oil overflow, it is characterised in that：
Comprise the steps：
1. carrying out 6 layers of decomposition to oil sample fluorescence spectra using db7 wavelet basis functions, d3 detail coefficients features are extracted, are treated
Test sample product 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm, five characteristic wave strong points wavelet coefficient
X1~X5；
2. the wavelet coefficient of 1. five characteristic wave strong points that step is obtained brings following Fisher discrimination formulas Y into1And Y2In,
Y1=-4.642-0.396*X1+0.384*X2-0.142*X3-0.167*X4+0.151*X5
Y2=-2.727+0.803*X1-0.314*X2+0.114*X3+0.207*X4-0.103*X5
Wherein, X1~X5Represent that d3 is small at 255 ± 2nm, 280 ± 2nm, 302 ± 2nm, 332 ± 2nm, 354 ± 2nm respectively
Wave system number；
3. 2. sample (Y that calculation procedure obtains1, Y2) Euclidean distance between value and each group barycenter judges to belong to, judgment rule
For：Smaller with certain group centroid distance, sample then belongs to this group of classification,
Wherein, the group barycenter of fuel oil is O1(Y1=3.293, Y2=0.211)；The group barycenter of middle-eastern crude is O2(Y1=-
2.114 Y2=0.769)；The group barycenter of non-middle-eastern crude is O3(Y1=-0.825, Y2=-0.505).
2. according to the method for claim 1, it is characterised in that：Step 1. in, the fluorescence spectra of the sample is using permanent
Wavelength method determines.
3. according to the method for claim 1, it is characterised in that：Step 1. in, the fluorescence spectra of the sample is to pass through
Molecular fluorescence spectrophotometer measures.
4. according to the method for claim 1, it is characterised in that：Step 2. in, the discrimination formula utilize Fisher differentiate
Method obtains.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710692532.XA CN107389645B (en) | 2017-08-14 | 2017-08-14 | The method that the Fisher model that wavelet transform parses oil product fluorescent characteristic identifies marine oil overflow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710692532.XA CN107389645B (en) | 2017-08-14 | 2017-08-14 | The method that the Fisher model that wavelet transform parses oil product fluorescent characteristic identifies marine oil overflow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107389645A true CN107389645A (en) | 2017-11-24 |
| CN107389645B CN107389645B (en) | 2019-08-27 |
Family
ID=60354810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710692532.XA Expired - Fee Related CN107389645B (en) | 2017-08-14 | 2017-08-14 | The method that the Fisher model that wavelet transform parses oil product fluorescent characteristic identifies marine oil overflow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107389645B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110907625A (en) * | 2019-12-06 | 2020-03-24 | 大连海事大学 | Method for distinguishing marine oil spill type based on multidimensional chemical fingerprint quantification model |
| CN114492213A (en) * | 2022-04-18 | 2022-05-13 | 中国石油大学(华东) | Wavelet neural operator network model-based residual oil saturation and pressure prediction method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339131A (en) * | 2008-08-13 | 2009-01-07 | 中国石油天然气集团公司 | Rock core microscopic various light spectrum image-forming information comprehensive processing method |
| CN101561395A (en) * | 2009-03-20 | 2009-10-21 | 中国海洋大学 | Phytoplankton composition quick determination method |
| US20110184654A1 (en) * | 2008-09-17 | 2011-07-28 | Opticul Diagnostics Ltd. | Means and Methods for Detecting Bacteria in an Aerosol Sample |
| CN102608085A (en) * | 2012-01-09 | 2012-07-25 | 暨南大学 | Method for detecting activity of alga hematoxin and application of method |
| CN104316505A (en) * | 2014-11-14 | 2015-01-28 | 深圳市朗诚实业有限公司 | Structure and application of three-dimensional fluorescence standard spectrum library used for recognizing toxic-to-fish algae |
-
2017
- 2017-08-14 CN CN201710692532.XA patent/CN107389645B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339131A (en) * | 2008-08-13 | 2009-01-07 | 中国石油天然气集团公司 | Rock core microscopic various light spectrum image-forming information comprehensive processing method |
| US20110184654A1 (en) * | 2008-09-17 | 2011-07-28 | Opticul Diagnostics Ltd. | Means and Methods for Detecting Bacteria in an Aerosol Sample |
| CN101561395A (en) * | 2009-03-20 | 2009-10-21 | 中国海洋大学 | Phytoplankton composition quick determination method |
| CN102608085A (en) * | 2012-01-09 | 2012-07-25 | 暨南大学 | Method for detecting activity of alga hematoxin and application of method |
| CN104316505A (en) * | 2014-11-14 | 2015-01-28 | 深圳市朗诚实业有限公司 | Structure and application of three-dimensional fluorescence standard spectrum library used for recognizing toxic-to-fish algae |
Non-Patent Citations (1)
| Title |
|---|
| 尹晓楠: "基于三维荧光光谱和小波分析的油品种类识别技术研究", 《中国博士学位论文全文数据库 工程科技I辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110907625A (en) * | 2019-12-06 | 2020-03-24 | 大连海事大学 | Method for distinguishing marine oil spill type based on multidimensional chemical fingerprint quantification model |
| CN110907625B (en) * | 2019-12-06 | 2022-02-22 | 大连海事大学 | Method for distinguishing marine oil spill type based on multidimensional chemical fingerprint quantification model |
| CN114492213A (en) * | 2022-04-18 | 2022-05-13 | 中国石油大学(华东) | Wavelet neural operator network model-based residual oil saturation and pressure prediction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107389645B (en) | 2019-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Stout et al. | A strategy and methodology for defensibly correlating spilled oil to source candidates | |
| CN101458213B (en) | Oil species identification method by sea oil spill concentration auxiliary parameter fluorescence spectrum | |
| CN109557071B (en) | Raman spectrum qualitative and quantitative identification method for dangerous liquid mixture | |
| CN107709983A (en) | The method for carrying out detailed batch classification analysis to complex sample using vacuum ultraviolet spectrometry and gas-chromatography | |
| CN113567401B (en) | Rapid detection method and application of landfill leachate polluted underground water condition | |
| CN108398416A (en) | A kind of mix ingredients assay method based on laser Raman spectroscopy | |
| CN105184000A (en) | Nonnegative-constrain-factor pollution source apportionment method based on naive Bayesian source identification | |
| Sun et al. | An efficient classification method for fuel and crude oil types based on m/z 256 mass chromatography by COW-PCA-LDA | |
| CN105973861B (en) | The method of marine oil overflow type is differentiated based on oil product fluorescent characteristic Fisher diagnostic methods | |
| CN107389645B (en) | The method that the Fisher model that wavelet transform parses oil product fluorescent characteristic identifies marine oil overflow | |
| CN111241458A (en) | Method for tracing odor in vehicle through multi-factor coupling analysis | |
| CN108548888B (en) | Method for accurately monitoring and evaluating petroleum hydrocarbon in organic pollution site | |
| Chua et al. | Tiered approach to long-term weathered lubricating oil analysis: GC/FID, GC/MS diagnostic ratios, and multivariate statistics | |
| Cui et al. | Excitation emission matrix fluorescence spectroscopy and parallel factor framework-clustering analysis for oil pollutants identification | |
| CN111551644A (en) | Method for tracing origin of imported fragrant rice based on ion mobility spectrometry technology | |
| Christensen et al. | A multivariate approach to oil hydrocarbon fingerprinting and spill source identification | |
| Ferreira et al. | 13C NMR spectroscopy of monoterpenoids | |
| Roman‐Hubers et al. | A comparative analysis of analytical techniques for rapid oil spill identification | |
| Ferreiro-González et al. | Characterization of petroleum-based products in water samples by HS-MS | |
| Wang et al. | Chemometric techniques in oil spill identification: A case study in Dalian 7.16 oil spill accident of China | |
| CN102221534A (en) | Mid-infrared spectrum method for quickly identifying engine fuel type | |
| Hur et al. | Petroinformatics | |
| RU2714517C1 (en) | Method of determining sources of hydrocarbon contamination in open water areas of seas in areas of development of oil and gas deposits | |
| CN105021747B (en) | The method being made up of proton nmr spectra prediction diesel oil race | |
| CN102980876A (en) | Multi-dimensional chemical fingerprint and method for identifying offshore weathered crude and bunker fuel oil |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190827 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |