Oil source comparison method and equipment based on multivariate statistical analysis

Abstract

The invention provides an oil source comparison method and equipment based on multivariate statistical analysis, and relates to the technical field of petroleum geological prospecting, and the method comprises the steps: carrying out the secondary change analysis of an oil source type set, and obtaining a rejected oil source type set; selecting a biomarker compound parameter set according to the removed oil source type set; obtaining a clustering pedigree diagram through the biomarker compound parameter set; obtaining a principal component load diagram through the biomarker compound parameter set; a key parameter set is obtained through the principal component load graph, factor scores are obtained through calculation of the key parameter set, and a factor score cross plot is constructed through the factor scores; and constructing a discriminant analysis chart through the clustering pedigree chart and the factor score cross plot. According to the method, the relation among the indexes in the main components is fully analyzed, the oil source relation is analyzed in a multi-dimensional mode, effective oil source comparison can be carried out under the complex background of development of multiple sets of hydrocarbon source rocks, and the method has the advantages of being high in comparison analysis efficiency, convenient to use and high in precision.

Description

Technical Field

[0001] The present invention relates to the technical field of petroleum geological exploration, and in particular to an oil source comparison method and equipment based on multivariate statistical analysis. Background Art

[0002] At present, oil-source correlation refers to the application of the basic principles of organic geochemistry, based on the fact that there is a genetic relationship between kerogen and asphalt in the source rock and the oil and gas from the stratum, and there must be a certain degree of similarity in chemical composition to reasonably select correlation parameters to study the relationship between oil, gas and source.

[0003] The development of oil and gas basins has gone through a long geological history. The formation time of oil and gas in different regions and layers is long, and their maturity and hydrocarbon generation cycles are also different. In addition, they may undergo a series of changes in the process of migration and accumulation. Different oil and gas reservoirs are not isolated from each other. Secondary faults and faults provide good channels for oil and gas transportation and migration, making the mixing of crude oil from different sources very common, greatly increasing the multi-solution and complexity of oil source comparison. However, the existing oil source comparison method cannot perform high-precision comparative analysis on mixed oil sources, resulting in inaccurate analysis results. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide an oil source comparison method and equipment based on multivariate statistical analysis, which is used to solve the problem that the existing oil source comparison method cannot perform high-precision comparison analysis on mixed oil sources.

[0005] The present invention provides an oil source comparison method based on multivariate statistical analysis, comprising the steps of:

[0006] S1: Obtain an oil source type set, perform secondary change analysis on the oil source type set, remove crude oil types affected by secondary changes, and obtain an oil source type set after removal;

[0007] S2: Select a biomarker compound parameter set based on the source rock type and crude oil type in the eliminated oil source type set;

[0008] S3: Perform hierarchical cluster analysis on the eliminated oil source type set through the biomarker compound parameter set to obtain a cluster pedigree diagram;

[0009] S4: principal component analysis is performed on the eliminated oil source type set through the biomarker compound parameter set to obtain a principal component loading diagram;

[0010] S5: Obtain the key parameter set through the principal component load diagram, calculate the factor scores of each principal component in the eliminated oil source type set through the key parameter set, and construct the factor score intersection diagram through the factor scores of each principal component;

[0011] S6: Construct a discriminant analysis graph through the clustering phylogenetic tree and the factor score intersection graph, and obtain the oil source correlation result through the discriminant analysis graph.

[0012] Preferably, step S1 is specifically as follows:

[0013] S11: The secondary change analysis includes: thermal maturity analysis, deasphalting analysis, biodegradation analysis, water washing analysis, oxidation analysis, and sulfidation analysis;

[0014] S12: Conduct secondary change analysis on each crude oil type in the oil source type set, and calculate to obtain the secondary change parameter set of each crude oil type. The secondary change parameter set includes: thermal evolution degree, degradation level, and water washing intensity;

[0015] S13: Take the crude oil types with secondary change parameters not within the preset range as the crude oil types affected by secondary changes;

[0016] S14: Eliminate all the crude oil types affected by secondary changes in the oil source type set to obtain the oil source type set after elimination.

[0017] Preferably, step S2 is specifically as follows:

[0018] Based on the color - mass spectrometry diagrams of biomarker compounds of source rocks and crude oils, select the biomarker compound parameter set in the oil source type set after elimination. Each biomarker compound parameter needs to satisfy that the sampling adequacy measure is greater than 0.6, and the significant value of the sphericity test is 0.

[0019] Preferably, step S3 is specifically as follows:

[0020] S31: Standardize the data of each biomarker compound parameter to obtain a standardized parameter set;

[0021] S32: Calculate the Euclidean distance of each standardized parameter, and construct a clustering phylogenetic tree through each Euclidean distance.

[0022] Preferably, step S4 is specifically as follows:

[0023] S41: Standardize the data of each biomarker compound parameter to obtain a standardized parameter set;

[0024] S42: Calculate the cumulative contribution rate of variance of each standardized parameter, and take the standardized parameters with the cumulative contribution rate of variance above the preset value as the principal components;

[0025] S43: Calculate the load of each index in each principal component, and construct a principal component load graph through each load.

[0026] Preferably, step S5 is specifically as follows:

[0027] S51: In the principal component loading diagram, the index with the largest absolute value of the loading in each principal component is used as the key parameter;

[0028] S52: Calculate the factor scores of each principal component through the key parameters, project each factor score into a factor score point in a three-dimensional coordinate system, and construct a factor score intersection diagram.

[0029] Preferably:

[0030] The calculation formula for the factor score of the principal component is:

[0031] F 1 = a ij1 X ij1 +...+...+ a ijn X ijn

[0032] F 2 = a ij2 X ij2 +...+...+ a ijm X ijm ...

[0034] F n = a ij3 X ij3 +...+...+ a ijp X ijp

[0035] where i represents the number of the principal component, j represents the number of the index, n is the total number of principal components, F n represents the factor score of the nth principal component, n, m, and p represent the nth, mth, and pth key variables, a ijn , a ijm , and a ijp represent the loadings of the principal component i index j corresponding to the nth, mth, and pth key variables, X ijn , X ijm , and X ijp represent the standardized parameters of the principal component i index j corresponding to the nth, mth, and pth key variables.

[0036] Preferably, step S6 is specifically as follows:

[0037] S61: Obtain the set of factor score points corresponding to each principal component through the factor score intersection diagram;

[0038] S62: Construct an initial two-dimensional diagram, calculate the first discriminant function value and the second discriminant function value of each factor score point, and use the first discriminant function value and the second discriminant function value as the coordinates of the factor score point on the initial two-dimensional diagram;

[0039] S63: Obtain the coordinates of the group centroids in the set of factor score points corresponding to each principal component on the initial two-dimensional graph through the clustering dendrogram;

[0040] S64: Construct a discriminant analysis graph through the initial two-dimensional graph, the coordinates of each factor score point, and the coordinates of each group centroid;

[0041] S65: Conduct initial verification and cross-verification on the discriminant analysis graph, and calculate to obtain a series of values and the current oil-source correlation result;

[0042] S66: Repeat steps S61 - S65 until the series of values converges and is greater than a preset value, and output the final oil-source correlation result.

[0043] A storage medium, which stores instructions and data for implementing the above-mentioned oil-source correlation method based on multivariate statistical analysis.

[0044] An oil-source correlation device based on multivariate statistical analysis, comprising: a processor and a storage medium; the processor loads and executes the instructions and data in the storage medium for implementing the above-mentioned oil-source correlation method based on multivariate statistical analysis.

[0045] The present invention has the following beneficial effects:

[0046] 1. By comprehensively considering biomarker compound parameters selected from source rocks and crude oils for hierarchical clustering analysis, it overcomes the inaccuracy in oil-source correlation when using biomarker compound fingerprint maps and intersection maps of biomarker compound parameters such as dual-endmember and triple-endmember in areas with complex crude oil sources, and realizes effective analysis of oil-source correlation;

[0047] 2. Conduct principal component analysis on the set of oil-source types, fully analyze the relationships between various indicators in each principal component, analyze oil-source relationships in multiple dimensions, and can conduct effective oil-source correlation under the complex background of the development of multiple sets of source rocks, with the characteristics of high efficiency, convenience, and high precision in comparative analysis. Description of the Drawings

[0048] Figure 1 It is a flowchart of the method in the embodiment of the present invention;

[0049] Figure 2 It is a clustering dendrogram;

[0050] Figure 3 It is a principal component loading diagram;

[0051] Figure 4 It is a factor score intersection diagram;

[0052] Figure 5 It is a discriminant analysis graph;

[0053] Figure 6 Structural diagram of the device according to an embodiment of the present invention;

[0054] The realization, functional characteristics and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0055] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0056] Referring to Figure 1 , the present invention provides an oil source comparison method based on multivariate statistical analysis, including the steps of:

[0057] S1: Obtain a set of oil source types, perform secondary change analysis on the set of oil source types, eliminate the crude oil types affected by secondary changes, and obtain the set of oil source types after elimination;

[0058] As an embodiment, step S1 is specifically:

[0059] S11: The secondary change analysis includes: thermal maturation analysis, deasphalting analysis, biodegradation analysis, water washing analysis, oxidation analysis and sulfidation analysis;

[0060] S12: Perform secondary change analysis on each crude oil type in the set of oil source types, and calculate the set of secondary change parameters for each crude oil type. The set of secondary change parameters includes: thermal evolution degree, degradation level and water washing intensity;

[0061] Specifically, (1) Based on the influence of thermal maturation on the hydrocarbon group components and biomarker compounds of crude oil, determine the thermal evolution degree of the crude oil;

[0062] Sterane C 29 20S / (20S + 20R), C 29 αββ / (ααα + αββ), C 31 22S / (22S + 22R), Ts / (Ts + Tm), rearranged sterane / (regular sterane + rearranged sterane), etc. These indicators can indicate the maturity information of source rocks and crude oil.

[0063] Steranes have relatively good thermal stability. C 29 ααα 20S / (20S + 20R) and C 29 Sterane αββ / (αββ + ααα) is also an important indicator for judging maturity. Generally, it is considered that C 29 ααα 20S / (20S + 20R) is 0 - 0.35, C 29The sterane αββ / (αββ+ααα) ranges from 0 to 0.25, belonging to the mature stage; when the former is 0.35-0.48 and 0.25-0.48, it reaches the low-mature stage; when the former > 0.48 and > 0.48, it reaches the mature stage.

[0064] (2) Based on the effects of biodegradation on the physical properties, group components, and biomarker compounds of crude oil, determine the degradation level;

[0065] With the increase of the biodegradation level, the content of saturated hydrocarbon components in crude oil gradually decreases, and the proportion of gum and asphaltene gradually increases, thereby leading to the increase of the density and viscosity of crude oil. It is also possible to determine the degradation level by observing the missing content of these biomarker compounds from the chromatogram-mass spectrum of biomarker compounds based on the different anti-degradation abilities of different biomarker compound series in crude oil.

[0066] The degradation level can also be obtained based on the content change of compounds with strong anti-degradation ability. Aromatic compounds have strong anti-degradation ability, and the anti-degradation abilities of dimethylphenanthrene (DMP), 9-methylphenanthrene (9-MP), 3-methylphenanthrene (3-MP), 1-methylphenanthrene (1-MP), and phenanthrene (P) decrease in turn. Therefore, it can be determined whether there is biodegradation in crude oil based on the decrease of the 9-MP / 1-MP value and the increase of the DMP / (P+MP) value. Among adamantane compounds, the increase of the methyladamantane / adamantane (MA / A) value can also be used to judge the degree of biodegradation.

[0067] (3) Based on the effects of water washing on the aromatic compounds of crude oil, determine the intensity of water washing;

[0068] The loss of aromatic compounds such as toluene and dibenzothiophene is common in the water washing process. The decrease of dibenzothiophene / phenanthrene (DBT / P) can be used to judge the intensity of water washing.

[0069] S13: Regard the crude oil type with secondary change parameters not within the preset range as the crude oil type affected by secondary changes;

[0070] S14: Remove all the crude oil types affected by secondary changes from the oil source type set to obtain the oil source type set after removal.

[0071] S2: Select the biomarker compound parameter set according to the source rock type and crude oil type in the oil source type set after removal;

[0072] As an embodiment, step S2 is specifically:

[0073] Based on the chromatogram-mass spectrometry diagrams of biomarker compounds in source rocks and crude oils, a set of biomarker compound parameters is selected from the set of oil source types after elimination. Each biomarker compound parameter needs to satisfy that the Kaiser-Meyer-Olkin measure of sampling adequacy is greater than 0.6, and the significance value of the sphericity test is 0.

[0074] Specifically, in the isoprenoid series, parameters such as pristane / phytane (Pr / Ph), Pr / nC 17 , Ph / nC 18 etc. can represent the sedimentary redox environment;

[0075] In the terpane series, the trinorhopane parameter (Ts / Tm) can be used as a maturity index, and gammacerane / hopane (Ga / (C 31 H(22R + 22S) / 2)), Ga / (2*C29Hop), Ga / C 30 Hop can indicate the water environment;

[0076] For the sterane series, the C 27 -C 28 -C 29 regular steranes can be used as an indicator of the parent material source.

[0077] In addition to these biomarker compound ratios, parameters such as tricyclic terpanes TriC19 / C21, TriC19 / C23, TriC19-20 / C23, TriC16-22 / C23-26, TriC30 / C30Hop, TriC26 / C25, Tri / C30Hop, Tri / Hop can all indicate the biogenic properties and environmental characteristics.

[0078] Based on the parameters needing to pass the Kaiser-Meyer-Olkin measure of sampling adequacy (KMO > 0.6) and Bartlett's test of sphericity (significance value = 0), parameters with a certain correlation are selected.

[0079] S3: Perform hierarchical cluster analysis on the set of oil source types after elimination using the set of biomarker compound parameters to obtain a cluster dendrogram;

[0080] As an example, step S3 is specifically as follows:

[0081] S31: Standardize the data of each biomarker compound parameter to obtain a set of standardized parameters;

[0082] Specifically, the purpose of data standardization is to reduce the error caused by different dimensions;

[0083] S32: Calculate the Euclidean distance of each standardized parameter, and construct a cluster dendrogram through each Euclidean distance.

[0084] Specifically, based on the standardized parameters, through the cyclic calculation of the Euclidean distance, classification is carried out according to the calculation results, and a clustering dendrogram is made according to the classification results to preliminarily determine the conclusion of oil-source correlation. The clustering dendrogram is as shown in Figure 2 shown, Figure 2 where different colors in it represent different clustering clusters, and different shapes represent different oil sources.

[0085] S4: Perform principal component analysis on the set of oil-source types after elimination through the set of biomarker parameters to obtain a principal component loading plot;

[0086] As an embodiment, step S4 is specifically as follows:

[0087] S41: Standardize the data of each biomarker parameter to obtain a set of standardized parameters;

[0088] Specifically, data standardization is carried out to reduce the error caused by different dimensions. The standardized parameters need to meet the Kaiser-Meyer-Olkin sampling adequacy measure (KMO≥0.6) and Bartlett's sphericity test (significance value<0.05);

[0089] S42: Calculate the cumulative contribution rate of variances of each standardized parameter, and take the standardized parameters with the cumulative contribution rate of variances above the preset value as the principal components;

[0090] Specifically, principal component analysis is carried out based on the standardized parameters to obtain new variables that are uncorrelated with each other. Taking the cumulative contribution rate of variances above 60% as the standard, the required number of principal components is extracted;

[0091] S43: Calculate the load of each index in each principal component, and construct a principal component loading plot through each load.

[0092] Specifically, analyze the parameters with high correlation with the principal component with the largest contribution rate through the principal component loading plot to determine the main factors controlling the oil-source correlation result. The principal component loading plot is as shown in Figure 3 shown, where the abscissa is the index of the principal component and the ordinate is the load.

[0093] S5: Obtain a set of key parameters through the principal component loading plot, calculate the factor scores of each principal component in the set of oil-source types after elimination through the set of key parameters, and construct a factor score intersection plot through the factor scores of each principal component;

[0094] As an embodiment, step S5 is specifically as follows:

[0095] S51: In the principal component loading plot, take the index with the largest absolute value of the load in each principal component as the key parameter;

[0096] S52: Calculate the factor scores of each principal component through various key parameters, project each factor score as a factor score point in a three-dimensional coordinate system, and construct a factor score intersection diagram.

[0097] Specifically, the factor score intersection diagram is as Figure 4 shown, and the factor score points corresponding to different types of oil sources are distributed in the three-dimensional coordinate system in the form of clustering clusters.

[0098] As an example, multiply the obtained principal component load value by the standardized parameter to obtain the factor score of the principal component;

[0099] The calculation formula for the factor score of the principal component is:

[0100] F 1 = a ij1 X ij1 +...+...+ a ijn X ijn

[0101] F 2 = a ij2 X ij2 +...+...+ a ijm X ijm ...

[0103] F n = a ij3 X ij3 +...+...+ a ijp X ijp

[0104] Among them, i represents the number of the principal component, j represents the number of the index, n is the total number of principal components, F n represents the factor score of the nth principal component, n, m, and p represent the nth, mth, and pth key variables, a ijn , a ijm , and a ijp represent the load of the principal component i index j corresponding to the nth, mth, and pth key variables, X ijn , X ijm , and X ijp represent the standardized parameters of the principal component i index j corresponding to the nth, mth, and pth key variables.

[0105] S6: Construct a discriminant analysis diagram through the clustering dendrogram and the factor score intersection diagram, and obtain the oil source comparison result through the discriminant analysis diagram.

[0106] As an example, step S6 is specifically:

[0107] S61: Obtain the set of factor score points corresponding to each principal component through the factor score intersection diagram;

[0108] S62: Construct an initial two-dimensional graph, calculate and obtain the first discriminant function value and the second discriminant function value of each factor score point, and use the first discriminant function value and the second discriminant function value as the coordinates of the factor score point on the initial two-dimensional graph;

[0109] S63: Obtain the coordinates of the group centroid in the set of factor score points corresponding to each principal component on the initial two-dimensional graph through the clustering dendrogram;

[0110] S64: Construct a discriminant analysis graph through the initial two-dimensional graph, the coordinates of each factor score point, and the coordinates of each group centroid;

[0111] Specifically, the discriminant analysis graph is as Figure 5 shown. The factor score points are distributed in the discriminant analysis graph in the form of clustering clusters, and the group centroid of each clustering cluster is marked;

[0112] S65: Conduct initial verification and cross-verification on the discriminant analysis graph, calculate and obtain a series of values and the current oil source comparison result;

[0113] Specifically, as shown in Table 1 for the verification results, if both verification results are greater than 75%, it indicates that the oil source comparison result has high reliability;

[0114] Table 1 Verification results of initial verification and cross-verification

[0115]

[0116] S66: Repeat steps S61 - S65 until the series of values converges and is greater than a preset value, and output the final oil source comparison result.

[0117] Please refer to Figure 6 , Figure 6 which is the schematic diagram of the working of the hardware device in the embodiment of the present invention. The hardware device specifically includes: an oil source comparison device 401 based on multivariate statistical analysis, a processor 402, and a storage medium 403.

[0118] An oil source comparison device 401 based on multivariate statistical analysis: The oil source comparison device 401 based on multivariate statistical analysis implements the oil source comparison method based on multivariate statistical analysis.

[0119] Processor 402: The processor 402 loads and executes the instructions and data in the storage medium 403 to implement the oil source comparison method based on multivariate statistical analysis.

[0120] Storage medium 403: The storage medium 403 stores instructions and data; the storage medium 403 is used to implement the oil source comparison method based on multivariate statistical analysis.

[0121] It should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or system including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or system including such element.

[0122] The serial numbers of the embodiments of the present invention above are only for description and do not represent the superiority or inferiority of the embodiments. In the unit claims listing several devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not denote any order and these words may be interpreted as identifiers.

[0123] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structural or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present invention.

Claims

1. An oil source comparison method based on multivariate statistical analysis, characterized in that: Includes steps: S1: Obtain an oil source type set, perform secondary change analysis on the oil source type set, remove crude oil types affected by secondary changes, and obtain an oil source type set after removal; S2: Selecting a biomarker compound parameter set based on the source rock type and crude oil type in the eliminated oil source type set; S3: Perform hierarchical cluster analysis on the eliminated oil source type set through the biomarker compound parameter set to obtain a cluster pedigree diagram; S4: principal component analysis is performed on the eliminated oil source type set through the biomarker compound parameter set to obtain a principal component loading diagram; S5: Obtain a key parameter set through the principal component load diagram, calculate the factor scores of each principal component in the eliminated oil source type set through the key parameter set, and construct a factor score intersection diagram through the factor scores of each principal component; S6: The discriminant analysis diagram is constructed through the cluster pedigree diagram and the factor score intersection diagram, and the oil source comparison results are obtained through the discriminant analysis diagram.

2. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S1 is specifically as follows: S11: Secondary change analysis includes: thermal maturation analysis, deasphalting analysis, biodegradation analysis, water washing analysis, oxidation analysis and sulfidation analysis; S12: performing secondary change analysis on each crude oil type in the oil source type set, and calculating and obtaining a secondary change parameter set of each crude oil type, wherein the secondary change parameter set includes: thermal evolution degree, degradation grade, and water washing intensity; S13: The crude oil type having the secondary change parameter not within the preset range is regarded as the crude oil type affected by the secondary change; S14: Eliminate all crude oil types affected by secondary changes in the oil source type set to obtain the eliminated oil source type set.

3. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S2 is specifically as follows: Based on the chromatographic-mass spectra of biomarker compounds of source rocks and crude oil, a set of biomarker compound parameters was selected from the eliminated oil source type set. Each biomarker compound parameter was required to satisfy the sampling suitability value greater than 0.6 and the significance value of the sphericity test was 0.

4. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S3 is specifically as follows: S31: performing data standardization on parameters of each biomarker compound to obtain a standardized parameter set; S32: Calculate the Euclidean distance of each standardized parameter, and construct a cluster pedigree diagram through each Euclidean distance.

5. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S4 is specifically as follows: S41: performing data standardization on parameters of each biomarker compound to obtain a standardized parameter set; S42: Calculate the cumulative variance contribution rate of each standardized parameter, and take the standardized parameter whose cumulative variance contribution rate is above a preset value as the principal component; S43: Calculate and obtain the load of each indicator in each principal component, and construct a principal component load diagram through each load.

6. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S5 is specifically as follows: S51: In the principal component load diagram, the index with the largest absolute value of the load in each principal component is taken as the key parameter; S52: Obtain the factor score of each principal component through calculation of each key parameter, project each factor score into a factor score point in a three-dimensional coordinate system, and construct a factor score intersection diagram.

7. The oil source comparison method based on multivariate statistical analysis according to claim 6, characterized in that: The calculation formula for the factor score of the principal component is: F1=a ij1 X ij1 +...+...+a ijn X ijn F2=a ij2 X ij2 +...+...+a ijm X ijm ... F n =a ij3 X ij3 +...+...+a ijp X ijp Among them, i represents the number of the principal component, j represents the number of the indicator, n represents the total number of principal components, and F n represents the factor score of the nth principal component, n, m, and p represent the nth, mth, and pth key variables, a ijn 、a ijm and a ijp represents the load of the principal component i index j corresponding to the nth, mth and pth key variables, X ijn , X ijm and X ijp Represents the standardized parameters of the principal component i index j corresponding to the nth, mth and pth key variables.

8. The oil source comparison method based on multivariate statistical analysis according to claim 1, characterized in that: Step S6 is specifically as follows: S61: Obtain the factor score point set corresponding to each principal component through the factor score intersection diagram; S62: construct an initial two-dimensional graph, calculate and obtain the first discriminant function value and the second discriminant function value of each factor score point, and use the first discriminant function value and the second discriminant function value as the coordinates of the factor score point on the initial two-dimensional graph; S63: Obtaining the coordinates of the center of gravity of the group in the factor score point set corresponding to each principal component on the initial two-dimensional graph through the cluster pedigree graph; S64: construct a discriminant analysis diagram using the initial two-dimensional diagram, the coordinates of each factor score point, and the coordinates of the center of gravity of each group; S65: Perform initial verification and cross-validation on the discriminant analysis diagram, and calculate the series values ​​and current oil source comparison results; S66: Repeat steps S61-S65 until the series values ​​converge and are greater than a preset value, and output the final oil source comparison result.

9. A storage medium, characterized in that: The storage medium stores instructions and data for implementing the oil source comparison method based on multivariate statistical analysis as described in any one of claims 1 to 8.

10. An oil source comparison device based on multivariate statistical analysis, characterized in that: include: A processor and a storage medium; the processor loads and executes instructions and data in the storage medium to implement the oil source comparison method based on multivariate statistical analysis as described in any one of claims 1 to 8.

Images