A method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells

Abstract

The present invention discloses a method for characterizing well-to-well connectivity based on quantitative analysis of endogenous compounds between wells, comprising the steps of: collecting the internal solution of each injection and production well to prepare a sample solution; processing and analyzing the sample solution to obtain chromatographic data and mass spectrometry data, and inferring the compound component data of each sample solution; using the sample solution number as a sample and the compound component data as a variable, and performing orthogonal partial least squares discriminant analysis on it to obtain a score graph and an S-plot graph. The present invention adopts a method for analyzing endogenous compounds in each injection and production well, by collecting the internal solution of each injection and production well as a sample and obtaining the endogenous compounds and content data of each injection and production well solution through liquid chromatography and mass spectrometry processing and analysis, and quantitatively analyzing each injection and production well and its endogenous compounds and content data by means of orthogonal partial least squares discriminant analysis to separate the relevant and irrelevant components of each injection and production well, thereby characterizing the connectivity relationship between each injection and production well.

Description

Technical Field

[0001] The invention relates to the technical field of oil reservoir management, and in particular to a method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells. Background Art

[0002] Reservoir connectivity is an important part of reservoir evaluation and the basis of oilfield development and management plans. Effective identification and analysis of reservoir well connectivity and determination of the main flow channels of underground fluids can provide guidance for optimizing water plugging, profile adjustment, acid fracturing and other measures, which is of great significance for the rational and effective development of water-flooded reservoirs in the later stage.

[0003] Reservoir well connectivity can be divided into static connectivity and dynamic connectivity based on influencing factors. Static connectivity is a description of the original connectivity of the reservoir before development, that is, a description of the continuity of the reservoir sandstone space. Although the research method of static connectivity is simple and applicable, it cannot reflect the changes in the reservoir in the later stage. Dynamic connectivity is described by various methods in the development stage, such as the inter-well tracer analysis method, which injects tracers into the injection wells and monitors the output time, output concentration, output peak size and number of tracers in the production wells. It can monitor the flow direction of the tracer and the water drive speed, and then obtain the dynamic connectivity between the oil and water wells. However, this method requires a large amount of tracers and captures and monitors the tracers. Its economic cost and operation time cost are high, and it is not suitable for large-scale use in oil fields.

[0004] Therefore, there is an urgent need to provide a method for judging inter-well connectivity, so as to conveniently and quickly quantitatively analyze the inter-well connectivity of reservoirs, provide more accurate data support and decision-making basis for oilfield production, and thus improve the efficiency and economic benefits of oilfield development. Summary of the invention

[0005] The purpose of the present invention is to provide a method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells, so as to solve the technical problem that the conventional dynamic connectivity judgment method using tracer analysis in the prior art requires the investment of a large amount of tracers and the capture and monitoring of the tracers, which has high economic cost and operation time cost and is not suitable for large-scale use in oil fields.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] A method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells comprises the following steps:

[0008] Step 100, collecting the internal solution of each injection and production well and dissolving it in equal amounts of solvents in sequence and numbering them as sample solutions of each injection and production well;

[0009] Step 200: subjecting the sample solutions of each injection and production well to liquid chromatography and mass spectrometry analysis in turn to obtain chromatography data and mass spectrometry data, and inferring the compound component data of the internal solution of each injection and production well in each sample solution based on the chromatography data and mass spectrometry data of each sample solution;

[0010] Step 300: Taking the sample solution number of each injection and production well as the sample and the compound component data of the internal solution of each injection and production well as the variable, an orthogonal partial least squares discriminant analysis is performed to obtain a score graph reflecting the relationship between groups and the relationship within groups and an S-plot graph reflecting the projection value of the variable importance, so as to determine the connectivity relationship of each injection and production well and its differential marker compounds based on the score graph and the S-plot graph.

[0011] As a preferred solution of the present invention, in step 200, the compound component data includes the compound molecular formula and the content of each compound in the solution composition of each injection and production well.

[0012] As a preferred solution of the present invention, in step 300, the specific method of the orthogonal partial least squares discriminant analysis is:

[0013] Establish an orthogonal partial least squares mathematical analysis model and give the data input format based on this model;

[0014] Inferring the number of compound molecular formula types of all sample solutions based on the chromatographic data and mass spectrometry data of each sample solution;

[0015] The sample solution numbers of each injection and production well are binary classified to divide the sample solution numbers into two groups to form a comparison;

[0016] Form a data table matching the data input format with all the compound molecular formula types and their corresponding compound contents in each injection and production well, sample solution numbers and classification quantities;

[0017] Import the data table into the orthogonal partial least squares mathematical analysis model to obtain the score graph and S-plot graph;

[0018] The relationship between the injection and production wells was determined based on the aggregation in the score graph, and the differential marker compounds between the injection and production wells were determined based on the dispersion in the S-plot graph.

[0019] As a preferred embodiment of the present invention, the data table includes a horizontal table with compound molecular formula as a variable, a vertical table with sample solution number as a sample, and a table with compound content as an element between compound molecular formula and sample solution number, wherein the sample solution number is classified into two groups.

[0020] As a preferred embodiment of the present invention, the expression of the orthogonal partial least squares mathematical analysis model is:

[0021]

[0022] in, is a The matrix of is the number of sample solution numbers, is the total number of compound types in the solution inside each injection and production well; yes The matrix of is the number of sample solution numbers; is with Related score matrix; is with The associated loading matrix; is with Orthogonal score matrix; is with Orthogonal loading matrix; is the residual matrix;

[0023] in, ,in yes indivual The matrix composed of It is The first The compound content data of the column.

[0024] As a preferred solution of the present invention, in step 200, the specific method of inferring the compound component data of the internal solution of each injection and production well in each sample solution is:

[0025] Sorting the mass spectrometry data and chromatographic data of the sample solution of each injection and production well, and distinguishing the chromatographic data and mass spectrometry data of each compound in the solution inside each injection and production well;

[0026] Compare the molecular ion peaks, fragment ion peaks and isotope ion peaks of the mass spectrum data of each compound with the data in the mass spectrum database to determine the molecular formula of each compound;

[0027] The chromatographic peak area of ​​the chromatographic data of each compound is calculated to obtain the content of each compound in the sample solution.

[0028] As a preferred solution of the present invention, in step 200, the specific method of obtaining chromatographic data and mass spectrometric data after sequentially subjecting the sample solutions of each injection and production well to liquid chromatography and mass spectrometry analysis is as follows:

[0029] The sample solutions of each injection and production well are sequentially introduced into the liquid chromatography-mass spectrometry equipment;

[0030] The compounds constituting each sample solution are layered in the chromatographic column and flow out of the chromatographic column in an elution order, so as to separate the compounds constituting each sample solution and obtain chromatographic data of each compound constituting the internal solution of each injection and production well;

[0031] Each compound flowing out of the chromatographic column is ionized in turn and ion separation and detection are performed to obtain mass spectrum data of each compound.

[0032] As a preferred embodiment of the present invention, in step 100, the number of internal solutions collected from each injection and production well is at least three, and the internal solutions of each injection and production well collected in each batch are treated as the same group for orthogonal partial least squares discriminant analysis, and the analysis structure of the internal solutions of each injection and production well collected in each batch is used as a control.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention adopts a method for analyzing endogenous compounds in each injection and production well. The internal solution of each injection and production well is collected as a sample and the endogenous compounds and content data of the solution of each injection and production well are obtained through liquid chromatography and mass spectrometry processing and analysis. The orthogonal partial least squares discriminant analysis method is used to quantitatively analyze each injection and production well and its endogenous compounds and content data to separate the relevant and irrelevant components of each injection and production well, thereby characterizing the connectivity relationship between the injection and production wells. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0036] Figure 1 A schematic flow chart of a method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells provided in an embodiment of the present invention;

[0037] Figure 2 An OPLS-DA model score diagram of a method for characterizing well-to-well connectivity based on quantitative analysis of endogenous compounds between wells provided in an embodiment of the present invention;

[0038] Figure 3 A permutation test result diagram of the OPLS-DA model of the method for characterizing well-to-well connectivity based on quantitative analysis of endogenous compounds between wells provided in an embodiment of the present invention;

[0039] Figure 4 An S-plot diagram of an OPLS-DA model of a method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells provided in an embodiment of the present invention;

[0040] Figure 5 A table of differential compounds obtained by screening the OPLS-DA model of the method for characterizing well-to-well connectivity based on quantitative analysis of endogenous compounds between wells provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the present invention provides a method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells, comprising the following steps:

[0043] Step 100, collecting the internal solution of each injection and production well and dissolving it in equal amounts of solvents in sequence and numbering them as sample solutions of each injection and production well;

[0044] Step 200: subjecting the sample solutions of each injection and production well to liquid chromatography and mass spectrometry analysis in turn to obtain chromatography data and mass spectrometry data, and inferring the compound component data of the internal solution of each injection and production well in each sample solution based on the chromatography data and mass spectrometry data of each sample solution;

[0045] Step 300: Taking the sample solution number of each injection and production well as the sample and the compound component data of the internal solution of each injection and production well as the variable, an orthogonal partial least squares discriminant analysis is performed to obtain a score graph reflecting the relationship between groups and the relationship within groups and an S-plot graph reflecting the projection value of the variable importance, so as to determine the connectivity relationship of each injection and production well and its differential marker compounds based on the score graph and the S-plot graph.

[0046] The method of judging the connectivity between wells in this embodiment is mainly based on petroleum omics to establish the connection between well materials from a molecular perspective, and to analyze the compound components of the well materials by liquid chromatography-mass spectrometry. In combination with the orthogonal partial least squares discriminant analysis method, the differential compounds between each sample (between wells) are screened, thereby quantitatively characterizing the connectivity between wells according to the quantitative distribution of the differential compounds.

[0047] Compared with the existing numerical simulation methods, the method for judging the connectivity between wells in this embodiment is mainly to directly collect the injected water of the water injection well and the produced fluid of the oil and water well as samples without affecting the oil wells, and analyze them from a molecular perspective to obtain the compound component information of the samples. According to the compound component information of the samples, the orthogonal partial least squares discriminant analysis method is used to screen the differential compounds between the samples (between wells), so as to quantitatively characterize the connectivity between wells according to the quantitative distribution of the differential compounds.

[0048] This implementation method is to combine petroleum omics and machine learning technology, use the OPLS-DA model (orthogonal partial least squares discriminant analysis method) for supervised pattern recognition, quickly and accurately screen and analyze the compounds in petroleum samples, reveal the connectivity between different wells, and screen out marker compounds that distinguish the connectivity between wells, providing important theoretical support for optimizing production of oil and gas fields.

[0049] The sample is dissolved in an organic solvent for testing to filter out inorganic matter (i.e., the part not related to petroleum) and reduce interference with the test.

[0050] Wherein, in step 200, the compound component data includes the compound molecular formula and the content of each compound of the internal solution composition of each injection and production well.

[0051] Since the solutes in the sample solution come from the internal solution of each injection and production well, and the solvent is an organic solvent with a known molecular formula and content, the constituent compounds of the sample solution are analyzed to accurately obtain the endogenous compounds of each injection and production well, that is, the types (molecular formulas) and contents of the constituent compounds of the internal solution.

[0052] Wherein, in step 300, the specific method of the orthogonal partial least squares discriminant analysis is:

[0053] Establish an orthogonal partial least squares mathematical analysis model and give the data input format based on this model;

[0054] Inferring the number of compound molecular formula types of all sample solutions based on the chromatographic data and mass spectrometry data of each sample solution;

[0055] The sample solution numbers of each injection and production well are binary classified to divide the sample solution numbers into two groups to form a comparison;

[0056] Form a data table matching the data input format with all the compound molecular formula types and their corresponding compound contents in each injection and production well, sample solution numbers and classification quantities;

[0057] Import the data table into the orthogonal partial least squares mathematical analysis model to obtain the score graph and S-plot graph;

[0058] The relationship between the injection and production wells was determined based on the aggregation in the score graph, and the differential marker compounds between the injection and production wells were determined based on the dispersion in the S-plot graph.

[0059] Chromatographic data include chromatographic peaks, retention time and peak area. The number of chromatographic peaks with different retention times indicates the number of compound components, and the size of the peak area indicates the relative content of each compound component.

[0060] The mass spectrometry data includes total ion current intensity and mass-to-charge ratio, the total ion current intensity includes molecular ion intensity, fragment ion intensity, metastable ion intensity and isotope ion intensity, and the molecular ion intensity, fragment ion intensity, metastable ion intensity and isotope ion intensity form molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak with mass-to-charge ratio as horizontal reference axis, and the molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak are multi-level mass spectrometry information of the mass spectrometry data;

[0061] The chromatographic data and mass spectrometry data constitute a three-dimensional spectrum of each compound component used for analysis. Ion characteristic information can be obtained based on the three-dimensional spectrum, including mass-to-charge ratio, retention time, molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak, and the ion characteristic information is obtained by preprocessing the chromatographic data and mass spectrometry data in the three-dimensional spectrum.

[0062] The data table includes a horizontal table with compound molecular formula as a variable, a vertical table with sample solution number as a sample, and a table with compound content as an element between compound molecular formula and sample solution number, wherein the sample solution number is classified into two groups.

[0063] As a preferred embodiment of the present invention, the expression of the orthogonal partial least squares mathematical analysis model is: ; ;

[0064] in, is a The matrix of is the number of sample solution numbers, is the total number of compound types in the solution inside each injection and production well; yes The matrix of is the number of sample solution numbers; is with Related score matrix; is with The associated loading matrix; is with Orthogonal score matrix; is with Orthogonal loading matrix; is the residual matrix; where ,in yes indivual The matrix composed of It is The first The compound content data of the column.

[0065] In this expression, ; ;in, is a The matrix of It is Individual and The associated score vector; is a The matrix of It is Individual and The associated load vector; is a The matrix of It is Individual and Orthogonal score vectors; is a The matrix of It is Individual and Orthogonal load vectors.

[0066] In step 200, the specific method of inferring the compound component data of the internal solution of each injection and production well in each sample solution is as follows:

[0067] Sorting the mass spectrometry data and chromatographic data of the sample solution of each injection and production well, and distinguishing the chromatographic data and mass spectrometry data of each compound in the solution inside each injection and production well;

[0068] Compare the molecular ion peaks, fragment ion peaks and isotope ion peaks of the mass spectrum data of each compound with the data in the mass spectrum database to determine the molecular formula of each compound;

[0069] The chromatographic peak area of ​​the chromatographic data of each compound is calculated to obtain the content of each compound in the sample solution.

[0070] The formation of a three-dimensional spectrum in a computer is currently a commonly used data integration management method for LC-MS. The three-dimensional spectrum contains all the mass spectral data of each compound, that is, it includes multi-level mass spectral information.

[0071] In order to identify the compound components in the sample, it is necessary to extract the ion characteristic information of each compound component, so as to deduce the compound components based on the ion characteristic information. Before extracting the ion characteristic information, the mass spectrometry data needs to be preprocessed. Specifically:

[0072] After mass spectrometry data preprocessing, the specific method for obtaining the ion characteristic information of each compound component is as follows:

[0073] Outputting the chromatogram and mass spectrum of each compound component of each sample in sequence from the chromatographic data and mass spectrum data acquired by the computer;

[0074] A baseline curve is established for noise filtering to remove peaks below the noise level and retain the true peak signal;

[0075] The mass-to-charge ratio, retention time, molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak of each compound component of each sample are obtained according to the real peak signal, that is, the ion characteristic information of each compound component of each sample is obtained.

[0076] During this process, a chromatogram and a mass spectrum are generated for each sample from the chromatographic data (liquid chromatography LC) and mass spectrometric data (MS) acquired by the computer. The chromatogram shows the signal intensity over time, while the mass spectrum shows the intensity of ions at different mass-to-charge ratios (m / z).

[0077] A baseline curve is established based on the mass spectrum. The baseline curve is a threshold set based on known signal intensity and noise level. Any signal below this threshold is considered noise and will be filtered out to improve the signal-to-noise ratio (SNR) and ensure that only significant signals are retained for subsequent analysis.

[0078] True peak signals in a mass spectrum are identified using peak detection algorithms (conventional technology), which typically look for continuous signals above the noise level to determine the presence of a compound.

[0079] The above information is integrated to create a list or database entry containing mass-to-charge ratio, retention time, molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak for each compound component, thus obtaining ion characteristic information.

[0080] Data preprocessing also includes quality control steps, such as correcting drift time, normalizing peak intensity, and removing background signals to ensure data consistency and accuracy.

[0081] After these steps are completed, the obtained ion characteristic information can be used for subsequent chemometric analysis such as compound identification and quantitative analysis.

[0082] Wherein, in step 200, the specific method of obtaining chromatographic data and mass spectrometric data after subjecting the sample solutions of each injection and production well to liquid chromatography and mass spectrometry analysis in sequence is as follows:

[0083] The sample solutions of each injection and production well are sequentially introduced into the liquid chromatography-mass spectrometry equipment;

[0084] The compounds constituting each sample solution are layered in the chromatographic column and flow out of the chromatographic column in an elution order, so as to separate the compounds constituting each sample solution and obtain chromatographic data of each compound constituting the internal solution of each injection and production well;

[0085] Each compound flowing out of the chromatographic column is ionized in turn and ion separation and detection are performed to obtain mass spectrum data of each compound.

[0086] After the mass spectrometry data is preprocessed to extract the ion characteristic information, the molecular formula of each compound component can be inferred based on the ion characteristic information of each compound component. Specifically:

[0087] In step 200, the specific method of identifying the compound components of each sample according to the ion characteristic information is as follows:

[0088] Determine molecular ion peaks, fragment ion peaks, metastable ion peaks and isotope ion peaks;

[0089] Compare the molecular ion peak, fragment ion peak, metastable ion peak and isotope ion peak with the data in the mass spectrum database to confirm the mass spectrum information of the compound components, including molecular mass, relative intensity and mass number parity, missing group and cleavage mode;

[0090] Based on the mass spectrum information of the compound components obtained through comparative analysis, the molecular formula of the compound is inferred, and reverse detection is performed based on the molecular formula of the compound to confirm the correct molecular formula of the compound component.

[0091] Among them, molecular ion peak (MI): determines the original ion peak of the compound molecule, usually corresponding to the peak with the highest m / z value. Fragment ion peak (FI): analyzes the various fragment ions produced by further fragmentation of molecular ions, which provide important clues about the structure of the compound.

[0092] Metastable ion peak (MI*): An ion peak produced by fragmentation in the flight path, whose energy state is different from the fragment ions produced by fragmentation in the source, providing additional structural information.

[0093] Isotope peaks: Ion peaks composed of natural isotopes of elements in a compound are analyzed to verify the correctness of the molecular formula.

[0094] These ion features are compared with the data in standard mass spectrometry databases, such as NIST, Wiley, HMDB, etc. The databases contain mass spectra and related feature information of a large number of known compounds.

[0095] Comparative analysis focuses on key parameters such as molecular mass, relative intensity, parity of mass number, possible missing groups and fragmentation mode to find the reference compound that best matches the sample data.

[0096] The molecular formula of the compound is preliminarily inferred based on the mass and isotope pattern of the molecular ion peak. The isotope pattern can reveal the type and quantity of specific elements in the compound, such as carbon (C), hydrogen (H), nitrogen (N), oxygen (O), etc. The fragment ion peaks are analyzed to infer possible fragmentation pathways and further verify the inferred molecular formula.

[0097] Using the inferred molecular formula, the theoretical molecular ion peaks and isotope distributions are calculated and then compared with the experimental data.

[0098] If the theoretical predictions are consistent with experimental observations, the inferred molecular formula can be confirmed to be correct; if not, the molecular formula needs to be re-evaluated or the data quality needs to be checked.

[0099] Combining the molecular formula and fragmentation pattern of the fragment ions, using chemical rules and experience, possible compound structures are constructed.

[0100] You can use software tools to assist in structure analysis, such as ChemDraw, MolView, etc., or you can refer to literature and professional knowledge.

[0101] Through the above steps, the compound components in the sample can be identified based on the ion characteristic information in the mass spectrometry data.

[0102] Wherein, in step 100, the number of internal solutions collected from each injection and production well is at least three, and each internal solution collected from each injection and production well is treated as the same group for orthogonal partial least squares discriminant analysis, and the analysis structure of each internal solution collected from each injection and production well is used as a control.

[0103] The accuracy of the analysis can be verified by using a control, and in order to ensure the accuracy of orthogonal partial least squares discriminant analysis, a permutation test is usually required, specifically:

[0104] The grouping labels (Y variables) of the samples were randomly shuffled, and the corresponding OPLS-DA models were established 500 times to obtain the R2Y and Q2 values ​​of the random model. The Q2 value represents the proportion of data variance that can be predicted by the current model, that is, the prediction rate. The larger the value, the better the prediction ability of the model. The R2Y value is the cumulative variance value of the model, which refers to the proportion of data variance or variation that can be explained by the current model, that is, the explanation rate. The larger the value, the stronger the explanatory ability of the model.

[0105] Based on the above method, a specific embodiment is provided below:

[0106] Block H of XX oilfield has 16 production wells, reservoir temperature of 53℃, reservoir pressure of 11MPa, burial depth of 1760m~1950m, porosity of 52.7%, recoverable reserves of 3.4×104t, crude oil viscosity of 1549mPa.s, and formation water salinity of 13648mg / L; the injection-production relationship between the wells in this block is unclear, so in order to optimize the well group in the next step, it is necessary to understand the inter-well connectivity of the 16 wells in this block. The inter-well connectivity is calculated by the method established by the present invention, and the specific process is as follows:

[0107] Step 1: On-site sampling: Use sterile bags to collect production fluid samples from different wells. When collecting, first release the liquid in the pipeline. Take about 100 ml of each sample. Seal the samples and store them in a dark place at 4°C.

[0108] Step 2: Sample pretreatment: The injected water or output fluid (supernatant) is dissolved in an organic solvent, filtered, and analyzed by liquid chromatography-orbitrap mass spectrometry (LC-Orbitrap MS).

[0109] Step 1: Take 10 mg of sample, dissolve it in 1 ml of toluene, and filter it with a 0.22 μm filter;

[0110] Step 2: The filtered sample was injected into the liquid chromatography-Orbitrap mass spectrometry system. The sample first entered the C18 chromatographic column and was eluted by the mobile phase (water and acetonitrile). The gradient elution program was: 0-0.5min, A=90%; 0.5-7min, A: decreasing to 90%~0%; 7-8.5min, A=0%; 8.6min, A=90%; 8.6-10min, A=90%; injection volume: 1μL, flow rate: 0.3mL / min.

[0111] Step 3: The compound is ionized in the mass spectrometer to generate ions, which are then screened and detected by the mass spectrometer. The ionization modes of the mass spectrometer are set to positive and negative ion modes. Positive ion mode: spray voltage: 3.8kv; sheath gas: 40; auxiliary gas: 10; ion transfer tube temperature: 350°C resolution: 17500; microscan: 1; AGCtarget: 2e5; normalized collision energy: 50. Negative ion mode: spray voltage: 2.9kv; sheath gas* 40; auxiliary gas: 0; ion transfer tube temperature: 350°C- resolution: 17500; microscan: 1; AGC target: 2e5; normalized collision energy: 50.

[0112] Step 3: Mass spectrometry data preprocessing: The collected data were processed using the MZmine v3 application with the following parameter settings: the mass range of the mass spectrometry was 150–1800 Da and the mass tolerance was 0.01 Da.

[0113] Step 1: Data import: Import the raw data (such as .raw format) obtained from the mass spectrometer into the MZmine software.

[0114] Step 2: Pretreatment: initial retention time 1 min, final retention time 10 min, mass range 100-1500 Da, mass tolerance 0.01 Da, no specific masses or adducts were excluded.

[0115] Step 3: Feature extraction: Extract the features in the sample (such as mass-to-charge ratio, retention time, etc.) and related peak area information.

[0116] Step 4: Identification of compounds: Sirius v5.6.3 was used to infer the structure of the compound's mass spectrum information and obtain the inferred structure within the corresponding number and error range.

[0117] Step 1: Use the mzmine2 software to export the secondary mass spectrum information of the substance in the mgf format from the peak alignment feature list. The secondary mass spectrum information includes the molecular mass of the compound in the mass spectrometer, the fragmentation spectrum of the molecular ion, and other information.

[0118] Step 2: Sirius v5.6.3 processes the molecular weight and secondary mass spectrometry fragment data and assigns molecular formulas by setting the ion mode and calculating the elements contained as C, H, O, N, and S with an error range of 2 ppm.

[0119] Step 3: Export the identification information of the compounds and merge it with the peak area information obtained by Mzmine processing for cluster analysis.

[0120] Step 5: OPLS-DA model analysis: Use the OPLS-DA algorithm to perform multivariate data analysis on the compound components of the sample, with the compound components as variables and the well names as samples. Step 1: Mathematical expression of the OPLS-DA model: Assume that there is a data matrix and the response variable matrix The basic expression of the OPLS-DA model is: .in, is a The matrix of is the number of sample solution numbers, is the total number of compound types in the solution inside each injection and production well; yes The matrix of is the number of sample solution numbers; is with Related score matrix; is with The associated loading matrix; is with Orthogonal score matrix; is with Orthogonal loading matrix; is the residual matrix.

[0121] Step 2: OPLS-DA analysis: The number of each injection and production well is used as a sample group, and a group contains at least three samples. The total amount of compound types in each injection and production well and the content relative to the number of each injection and production well are used as variables and imported into the OPLS-DA model to obtain the score scatter plot ( Figure 2 ) and S-plot score graph ( Figure 4 ). The horizontal axis of the S-plot graph represents the co-correlation coefficient between the principal component and the compound, and the vertical axis represents the correlation coefficient between the principal component and the compound. The closer the compound is to the upper right corner and the lower left corner, the more significant its difference is. The red points indicate that the VIP value of these compounds is greater than or equal to 1, and the green points indicate that the VIP value of these compounds is less than 1.

[0122] Step 2: Permutation test: The permutation test model randomly shuffles the grouping labels (Y variables) of the sample groups, and builds the corresponding OPLS-DA model 500 times to obtain the R2Y and Q2 values ​​of the random model. The Q2 value indicates the proportion of data variance that the current model can predict, that is, the prediction rate. The larger the value, the better the prediction ability of the model. The R2Y value is the cumulative variance value of the model, which refers to the proportion of data variance or variation that the current model can explain, that is, the explanation rate. The larger the value, the stronger the model's explanatory ability. Figure 3 .

[0123] The judgment results are as follows: wells 1, 11, and 16 are grouped together as group A, wells 5, 12, 13, and 17 are grouped together as group B, wells 2, 6, 9, and 15 are grouped together as group C, and wells 3, 4, 7, 8, and 14 are grouped together as group D, indicating that the connectivity of the wells in the same group is good. The permutation test results are good and the model accuracy is high. The differential compounds can be obtained through the S-plot score diagram, see Figure 5 The table in .

[0124] It is worth mentioning that the use of liquid chromatography-mass spectrometry to analyze the molecular formula and compound content of sample constituent compounds is a commonly used analysis method for sample solution components, and OPLS-DA analysis (orthogonal partial least squares discriminant analysis) is a commonly used statistical modeling method for classification and prediction. The liquid chromatography-mass spectrometry analysis method can determine the endogenous compounds of each injection and production well. Based on the endogenous compounds of each injection and production well, OPLS-DA analysis can effectively separate components related to and unrelated to the classification information from the data, that is, through the quantitative analysis of endogenous compounds, according to the correlation and difference of the analysis results, the connectivity relationship of each injection and production well, as well as the differential marker compounds between each injection and production well are characterized, which is convenient for relevant personnel to manage oil well production according to the connectivity relationship and differential marker compounds of each injection and production well.

[0125] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells, characterized in that: The steps include: Step 100, collecting the internal solution of each injection and production well and dissolving it in equal amounts of solvents in sequence and numbering them as sample solutions of each injection and production well; Step 200: subjecting the sample solutions of each injection and production well to liquid chromatography and mass spectrometry analysis in turn to obtain chromatography data and mass spectrometry data, and inferring the compound component data of the internal solution of each injection and production well in each sample solution based on the chromatography data and mass spectrometry data of each sample solution; Step 300: taking the sample solution number of each injection and production well as the sample and the compound component data of the solution inside each injection and production well as the variable, performing orthogonal partial least squares discriminant analysis on them to obtain a score graph reflecting the relationship between groups and the relationship within groups and an S-plot graph reflecting the projection value of variable importance, thereby determining the connectivity relationship of each injection and production well and its differential marker compound according to the score graph and the S-plot graph; In step 200, the compound component data includes the compound molecular formula and the content of each compound in the solution of each injection and production well; In step 300, the specific method of the orthogonal partial least squares discriminant analysis is: Establish an orthogonal partial least squares mathematical analysis model and give the data input format based on this model; Inferring the number of compound molecular formula types of all sample solutions based on the chromatographic data and mass spectrometry data of each sample solution; The sample solution numbers of each injection and production well are binary classified to divide the sample solution numbers into two groups to form a comparison; Form a data table matching the data input format with all the compound molecular formula types and their corresponding compound contents in each injection and production well, sample solution numbers and classification quantities; Import the data table into the orthogonal partial least squares mathematical analysis model to obtain the score graph and S-plot graph; The relationship between the injection and production wells is determined based on the aggregation in the score graph, and the differential marker compounds between the injection and production wells are determined based on the dispersion in the S-plot graph; In step 200, the specific method of inferring the compound component data of the internal solution of each injection and production well in each sample solution is as follows: Sorting the mass spectrometry data and chromatographic data of the sample solution of each injection and production well, and distinguishing the chromatographic data and mass spectrometry data of each compound in the solution inside each injection and production well; Compare the molecular ion peaks, fragment ion peaks and isotope ion peaks of the mass spectrum data of each compound with the data in the mass spectrum database to determine the molecular formula of each compound; The chromatographic peak area of ​​the chromatographic data of each compound is calculated to obtain the content of each compound in the sample solution.

2. A method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells according to claim 1, characterized in that: The data table includes a horizontal table with compound molecular formula as a variable, a vertical table with sample solution numbers as samples, and a table with compound content as elements between compound molecular formula and sample solution numbers, wherein the sample solution numbers are classified into two groups.

3. A method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells according to claim 2, characterized in that: The expression of the orthogonal partial least squares mathematical analysis model is: ; ;in, is a The matrix of is the number of sample solution numbers, is the total number of compound types in the solution inside each injection and production well; yes The matrix of is the number of sample solution numbers; is with Related score matrix; is with The associated loading matrix; is with Orthogonal score matrix; is with Orthogonal loading matrix; is the residual matrix; where ,in yes indivual The matrix composed of It is The first The compound content data of the column.

4. The method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells according to claim 1, characterized in that: In step 200, the sample solution of each injection and production well is subjected to liquid chromatography and mass spectrometry analysis in sequence to obtain chromatographic data and mass spectrometry data in the following specific manner: The sample solutions of each injection and production well are sequentially introduced into the liquid chromatography-mass spectrometry equipment; The compounds constituting each sample solution are layered in the chromatographic column and flow out of the chromatographic column in an elution order, so as to separate the compounds constituting each sample solution and obtain chromatographic data of each compound constituting the internal solution of each injection and production well; Each compound flowing out of the chromatographic column is ionized in turn and ion separation and detection are performed to obtain mass spectrum data of each compound.

5. The method for characterizing inter-well connectivity based on quantitative analysis of endogenous compounds between wells according to claim 1, characterized in that: In step 100, the number of internal solutions collected from each injection and production well is at least three, and each internal solution collected from each injection and production well is treated as a group for orthogonal partial least squares discriminant analysis, and the analysis structure of each internal solution collected from each injection and production well is used as a control.

Images