CN117541082B - Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration - Google Patents
Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration Download PDFInfo
- Publication number
- CN117541082B CN117541082B CN202410016502.7A CN202410016502A CN117541082B CN 117541082 B CN117541082 B CN 117541082B CN 202410016502 A CN202410016502 A CN 202410016502A CN 117541082 B CN117541082 B CN 117541082B
- Authority
- CN
- China
- Prior art keywords
- comprehensive evaluation
- evaluation
- production data
- data
- model
- 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.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 269
- 230000010354 integration Effects 0.000 title claims abstract description 18
- 238000004519 manufacturing process Methods 0.000 claims abstract description 260
- 238000013210 evaluation model Methods 0.000 claims abstract description 157
- 238000000034 method Methods 0.000 claims abstract description 148
- 239000011159 matrix material Substances 0.000 claims abstract description 60
- 238000011160 research Methods 0.000 claims abstract description 40
- 239000003129 oil well Substances 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 24
- 238000012795 verification Methods 0.000 claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims description 42
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 16
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 8
- 230000008878 coupling Effects 0.000 claims description 7
- 238000010168 coupling process Methods 0.000 claims description 7
- 238000005859 coupling reaction Methods 0.000 claims description 7
- 238000010219 correlation analysis Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000005484 gravity Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 abstract description 2
- 238000005457 optimization Methods 0.000 abstract description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 20
- 230000008859 change Effects 0.000 description 16
- 239000007788 liquid Substances 0.000 description 12
- 239000012530 fluid Substances 0.000 description 8
- 230000035699 permeability Effects 0.000 description 8
- 239000004568 cement Substances 0.000 description 7
- 239000004927 clay Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 239000004576 sand Substances 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008685 targeting Effects 0.000 description 2
- 238000012351 Integrated analysis Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000013329 compounding Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration, and particularly relates to the technical field of oil and gas field development. The method comprises the steps of obtaining production data pretreatment of all wells in a research area to obtain an effective production data set, sorting and screening comprehensive evaluation indexes according to comprehensive evaluation index screening principles and feature importance, constructing a comprehensive evaluation judgment matrix, determining weights of all comprehensive evaluation indexes by using a hierarchical analysis method-entropy weight method, respectively calculating scoring results of all wells in the research area by using a plurality of evaluation models, optimizing the scoring results to obtain an optimal comprehensive evaluation model and an optimal comprehensive score of an oil well, and verifying accuracy by using verification indexes. The invention comprehensively considers the influence of oil reservoirs, shafts and equipment on the oil field production, realizes multidirectional accurate evaluation of the oil field production state, and provides directional information for oil field optimization.
Description
Technical Field
The invention relates to the technical field of oil and gas field development, in particular to a comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration.
Background
The oil well production condition evaluation is used as directional information for guiding on-site operation, a foundation is laid for follow-up optimized production, problems existing in the production process can be identified by periodically evaluating the oil well production condition, and measures are taken to improve the production efficiency, so that oil reservoir resources are utilized to the greatest extent, reasonable exploitation and distribution of the resources are ensured, market demands are met to the greatest extent, and waste is reduced. Therefore, well production condition assessment is critical to ensure the efficiency, sustainability and safety of oil and gas field production.
The traditional oil well production evaluation method is mostly used for on-site individual combat, has the problems of data sharing and analysis and fracture, mainly relies on unilateral analysis and evaluation of professional data, and often has the condition of incomplete analysis. Meanwhile, the oil reservoir shaft equipment is an organic whole which is mutually related and influenced, when the problem of oil reservoir shaft compounding occurs, the oil well production condition evaluation is relatively single-sided only by relying on an oil reservoir or a shaft or equipment single system, the pertinence of treatment measures is not strong, and the analysis is weak particularly for the common nodes among the oil reservoir, the shaft and the equipment.
The existing integrated analysis method for petroleum production is deficient, and an oil reservoir-shaft coupling analysis method, an oil reservoir-shaft-pipe network coupling analysis method, an integrated productivity analysis method and the like are mostly adopted, however, node fluid models are needed in the analysis process of the analysis methods, and the analysis methods are poor in universality and low in evaluation accuracy under the assumption that conditions are too many and experience is relied on. Therefore, there is a need to propose a comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration, comprehensively considering oil reservoir, shaft and lifting equipment, and comprehensively evaluating the oil well production state in multiple targets.
Disclosure of Invention
The invention aims to improve the accuracy of oil well production condition evaluation, provides a comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration, solves the problem of single existing oil field evaluation mode, comprehensively considers the influence of oil reservoir, shaft and lifting equipment on the oil field production state, and improves the accuracy of oil well production evaluation results.
In order to achieve the above purpose, the invention adopts the following technical scheme:
an oil reservoir-shaft-equipment evaluation index integration-based comprehensive evaluation method comprises the following steps:
step 1, acquiring production data of all wells in a research area based on oil reservoir data, shaft data and lifting equipment data, constructing a production data set, and preprocessing the production data of each well in the production data set to obtain an effective production data set;
Step 2, based on the effective production data set, screening comprehensive evaluation indexes for evaluating oil reservoirs, shaft and equipment according to a comprehensive evaluation index screening principle and feature importance ranking pairs, and constructing a comprehensive evaluation discrimination matrix;
step 3, determining the weight of each comprehensive evaluation index based on an analytic hierarchy process-entropy weight process;
step 4, based on the comprehensive evaluation discrimination matrix, calculating the scoring results of each well in the research area respectively by using a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy comprehensive evaluation model and a rank sum ratio comprehensive evaluation model, wherein the scoring results comprise oil reservoir evaluation scores, shaft evaluation scores, equipment evaluation scores and comprehensive evaluation scores;
step 5, optimizing scoring results based on the TOPSIS comprehensive evaluation model, the gray correlation evaluation model, the entropy method comprehensive evaluation model and the rank sum ratio method comprehensive evaluation model to obtain an optimal comprehensive evaluation model and an optimal comprehensive score of the oil well;
and 6, verifying the accuracy of the optimal comprehensive evaluation model by using the verification index.
Preferably, in the step 1, the method includes the following steps:
step 1.1, acquiring production data of all wells in a research area, constructing a production data set, determining that each production data is digital characteristic data or character characteristic data according to the type of each production data in the production data set, performing data cleaning on the production data in the production data set, removing abnormal oil well production data in the oil well production data set, and performing blank value filling on the production data missing in the production data set;
Step 1.2, based on the production data in the production data set after data processing, the oil reservoir data, the shaft data and the lifting equipment data in the production data set are fused, and the production data of the same well are combined to form a sub-data set according to the well number;
and 1.3, performing tag coding on the text type characteristic data in the production data set, performing standardized processing on various production data in the production data set after converting the text type characteristic data assignment into digital type characteristic data, and obtaining an effective production data set.
Preferably, after searching abnormal production data in the production data set to determine data abnormal wells, deleting the production data of all the data abnormal wells in the production data set for data cleaning;
searching missing production data in the production data set after data cleaning, determining a well with a blank value of the production data as a data missing well, acquiring a well number of the data missing well, and judging the missing production data as digital characteristic data or character characteristic data;
when the missing production data is digital characteristic data, filling blank values of the data missing wells according to the well numbers, if the same type of production data exists in the wells positioned in the same row with the data missing wells, filling the production data of the data missing wells by using the average value of the same type of production data of all the wells positioned in the same row, and if the same type of production data does not exist in the wells positioned in the same row with the data missing wells, filling the blank values of the data missing wells by using the average value of the same type of production data of all the wells in the research area;
And when the missing production data is character feature data, filling the blank value of the production data of the data missing well according to the well number, if the same type of production data exists in the wells positioned in the same row with the data missing well, filling the production data of the data missing well by using the production data corresponding to the mode of the same type of production data in all the wells in the same row, and if the same type of production data does not exist in the wells positioned in the same row with the data missing well, filling the blank value of the data missing well by using the production data corresponding to the mode of the same type of production data of all the wells in the research area.
Preferably, in the step 2, the method includes the following steps:
step 2.1, based on a comprehensive evaluation index screening principle, primarily screening evaluation indexes for evaluating oil reservoirs, shaft and equipment according to production indexes corresponding to various production data in an effective production data set;
step 2.2, carrying out correlation analysis on the evaluation indexes, and calculating the spearman correlation coefficient of each evaluation index to be used as an importance score;
step 2.3, constructing an XGBoost model, training the XGBoost model by using an effective production data set, verifying, and determining the importance of each evaluation index by using the verified XGBoost model;
Step 2.4, constructing a comprehensive evaluation index matrix by using the production data corresponding to each comprehensive evaluation index, and normalizing the comprehensive evaluation index matrix to obtain a comprehensive evaluation discrimination matrix。
Preferably, in the step 2.3, the method specifically includes the following steps:
step 2.3.1, dividing production data in an effective production data set into an evaluation index and a target variable, setting an accuracy value of an XGBoost model, and dividing the effective production data set into a test set and a training set;
step 2.3.2, training an XGBoost model by using production data in a training set, acquiring production data corresponding to each evaluation index by using the XGBoost model, calculating importance scores of each evaluation index, outputting according to descending order of the importance scores, acquiring the calculation precision of the current XGBoost model in each training process, comparing with a preset precision value, entering the step 2.3.3 if the calculation precision of the XGBoost model reaches the preset precision value, otherwise, repeating the step 2.3.2 to continuously train the XGBoost model;
2.3.3, verifying that the calculation accuracy of the XGBoost model reaches a preset accuracy value by using a test set to obtain a verified XGBoost model;
step 2.3.4, inputting the production data in the effective production data set into the validated XGBoost model, calculating the importance scores of all the evaluation indexes by using the XGBoost model, outputting the importance scores according to descending order, obtaining the importance ranking of all the evaluation indexes, and selecting the importance ranking before selecting The name evaluation index is used as a comprehensive evaluation index.
Preferably, in the step 3, the method includes the following steps:
step 3.1, constructing a hierarchical structure model, wherein the hierarchical structure model comprises a target layer, a criterion layer and a scheme layer, wherein the target layer is internally provided with ranking of oil well production conditions in a research area, the criterion layer is internally provided with the comprehensive evaluation indexes obtained by screening in the step 2, and the scheme layer is internally provided with a well number;
step 3.2, constructing a discrimination matrix by adopting an entropy weight methodCalculating the entropy weight of each comprehensive evaluation index;
and 3.3, constructing a discrimination matrix based on an analytic hierarchy process, coupling an entropy weight method with the analytic hierarchy process, and replacing the entropy weight method weight of each comprehensive evaluation index through a scale to obtain the weight of each comprehensive evaluation index.
Preferably, the discriminant matrix constructed based on entropy weight methodThe method comprises the following steps:
（1）
in the method, in the process of the invention,in the number of rows of the drawing,and is also provided withIs an integer of the number of the times,as a total number of samples,is of the number of columns andis an integer of the number of the times,the total number of the comprehensive evaluation indexes is calculated;to distinguish the first in the matrixLine 1Column sample data corresponding to the first valid production data setLine 1Production data of the columns;
normalized processing discrimination matrix based on linear proportional conversion methodObtaining a normalization matrix;
if the comprehensive evaluation index in the judgment matrix is a positive index, a positive index transformation formula is utilized to transform to obtain a processed sample data value, and if the comprehensive evaluation index in the judgment matrix is a negative index, a negative index transformation formula is utilized to transform to obtain a processed sample data value;
The transformation formula of the forward index is as follows:
（2）
in the method, in the process of the invention,to distinguish the first in the matrixLine 1The processed sample data value of the column sample data,as a function of the minimum value of the function,is a maximum function;
the transformation formula of the negative index is as follows:
（3）
and calculating the entropy weight of each comprehensive evaluation index, as shown in a formula (4):
（4）
wherein,
（5）
in the method, in the process of the invention,is the firstThe entropy weight of the comprehensive evaluation index,is the firstThe information entropy of the comprehensive evaluation index,is the firstThe first of the comprehensive evaluation indexesSpecific gravity of the individual sample data;
constructing a discrimination matrix based on an analytic hierarchy process, and coupling an entropy weight method with the analytic hierarchy process to obtain the discrimination matrix after scale replacementThe method comprises the following steps:
（6）。
preferably, in the step 4, the calculation process of the TOPSIS comprehensive evaluation model is as follows: determining positive ideal solutions and negative ideal solutions by taking each well in a research area as an evaluation object, and then calculating the distances between each evaluation object and the positive ideal solutions and the negative ideal solutions to obtain the relative proximity of each evaluation object to the ideal solutions, so as to determine the TOPSIS comprehensive evaluation model scores of each evaluation object;
the gray correlation evaluation model calculation process comprises the following steps: taking each well in a research area as an evaluation object, firstly selecting optimal sample data of each comprehensive evaluation index, constructing an optimal index set, and respectively calculating gray correlation coefficients corresponding to each comprehensive evaluation index in each evaluation object to obtain gray correlation of each evaluation object to an ideal solution so as to determine a gray correlation evaluation model score of each evaluation object;
The calculation process of the entropy method comprehensive evaluation model is as follows: taking each well in the research area as an evaluation object, and calculating the target score of each evaluation object based on the entropy weight method weight of each comprehensive evaluation index calculated in the step 3.2, thereby determining the entropy method comprehensive evaluation model score of each evaluation object;
the calculation process of the rank sum ratio method comprehensive evaluation model is as follows: using each well in the research area as an evaluation object, calculating the rank of each comprehensive evaluation index in the evaluation object based on the comprehensive evaluation discrimination matrix to obtain a rank matrixAnd then, calculating the rank and the ratio of each evaluation object, thereby determining the rank and the ratio method comprehensive evaluation model score of each evaluation object.
Preferably, in the step 5, the method includes the following steps:
step 5.1, carrying out correlation analysis on a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy method comprehensive evaluation model and a rank sum ratio method comprehensive evaluation model, calculating a spearman correlation coefficient of each evaluation model, and determining a correlation value among the evaluation models;
step 5.2, constructing a correlation matrix by taking the sum of correlation values between the selected evaluation model and other evaluation models as a preferred sampleAs shown in formula (7):
（7）
In the method, in the process of the invention,for TOPSIS comprehensive evaluation model and other three evaluationThe sum of the correlation coefficients between the models,，；for the sum of the correlation coefficients between the grey correlation evaluation model and the other three evaluation models,，；the sum of correlation coefficients between the comprehensive evaluation model and the other three evaluation models by an entropy method,，；the sum of correlation coefficients between the rank sum ratio method comprehensive evaluation model and the other three evaluation models,，；、are all numbers of the evaluation model, whenOr (b)When the TOPSIS comprehensive evaluation model is expressed, whenOr (b)When representing the grey correlation evaluation model, whenOr (b)3, representing an entropy method comprehensive evaluation model whenOr (b)4, representing a rank sum ratio method comprehensive evaluation model;
and 5.3, selecting an evaluation model corresponding to the maximum value of the preferred sample as an optimal comprehensive evaluation model according to the preferred sample of each evaluation model, and obtaining the optimal comprehensive score of the oil well of each current well by using the optimal comprehensive evaluation model.
Preferably, in the step 6, a verification index of the well in the research area is obtained in advance, and the oil reservoir evaluation score, the well shaft evaluation score, the equipment evaluation score and the correlation coefficient between the comprehensive evaluation score and the verification index calculated by the optimal comprehensive evaluation model are calculated based on Kendall correlation coefficient analysis, so that the consistency between the optimal comprehensive evaluation model and the verification index is checked;
The Kendall correlation coefficient calculation formula is as follows:
（8）
in the method, in the process of the invention,for the Kendall correlation coefficient,in order to be consistent for the number of pairs,in order to divide the number of pairs into the two,score sequences calculated for optimal comprehensive evaluation modelsThe number of parallel rows in the row is equal,is an authentication indexIs arranged in parallel.
The beneficial technical effects brought by the invention are as follows:
according to the method, the oil reservoir evaluation index, the shaft evaluation index and the equipment evaluation index are combined, so that the omnibearing and multi-target accurate evaluation of the oil well production state is realized, and compared with the conventional oil well production evaluation method which adopts a single index for evaluation, the oil reservoir evaluation, the shaft evaluation and the equipment evaluation are mutually independent and mutually split.
Drawings
FIG. 1 is a flow chart of a comprehensive evaluation method based on oil reservoir-well bore-equipment evaluation index integration.
FIG. 2 is a schematic diagram of the oil reservoir evaluation scores for each well in the study area.
FIG. 3 is a schematic diagram of wellbore evaluation scores for wells in a study area.
FIG. 4 is a schematic diagram of equipment evaluation scores for wells in a study area.
FIG. 5 is a graph of the overall evaluation index evaluation scores for each well in the study area.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Example 1
The embodiment discloses a comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration, which specifically comprises the following steps as shown in fig. 1:
step 1, based on oil reservoir data, shaft data and lifting equipment data, obtaining production data of all wells in a research area, constructing a production data set, and preprocessing the production data of all wells in the production data set to obtain an effective production data set, wherein the method comprises the following steps:
step 1.1, obtaining production data of all wells in the research area, and constructing a production data set.
In this embodiment, the production data includes oil reservoir data, wellbore data, and lifting device data, specifically geological parameter data, oil reservoir energy production data, oil reservoir dynamic change data, wellbore design data, well track design data, casing design data, perforation data, lifting device design data, lifting device power efficiency data, and lifting device energy consumption economic data.
And determining the data type of each production data according to the type of each production data in the production data set, wherein the data type comprises digital characteristic data or text characteristic data, the digital characteristic parameters comprise stroke, stroke frequency, oil pressure, casing pressure, back pressure, pump efficiency, production days, design well depth, well completion depth, cement return depth, perforation thickness, perforation density, pump depth, casing wall thickness, production thickness, porosity, permeability, oil saturation and clay content, and the text characteristic parameters comprise well type, well completion method, well cementation quality and casing test time.
And carrying out data cleaning on the production data in the production data set, searching abnormal production data in the production data set, determining the well with the abnormal production data as a data abnormal well, and eliminating the production data of all the data abnormal wells in the production data set to finish data cleaning on the production data in the production data set.
And filling blank values of the production data missing in the production data set, searching the missing production data in the production data set after data cleaning, determining a well with the blank values of the production data as a data missing well, acquiring the well number of the data missing well, and judging the missing production data as digital characteristic data or character characteristic data.
And when the missing production data is digital characteristic data, filling the production data blank value of the data missing well according to the well number, filling the production data of the data missing well by using the average value of the same type of production data of all the wells in the same row if the same type of production data exists in the wells in the same row with the data missing well, and filling the production data blank value of the data missing well by using the average value of the same type of production data of all the wells in the research area if the same type of production data does not exist in the wells in the same row with the data missing well.
And when the missing production data is character feature data, filling the production data blank value of the data missing well according to the well number, if the same type of production data exists in the wells positioned in the same row with the data missing well, filling the production data of the data missing well by using the production data corresponding to the mode of the same type of production data in all the wells in the same row, and if the same type of production data does not exist in the wells positioned in the same row with the data missing well, filling the production data blank value of the data missing well by using the production data corresponding to the mode of the same type of production data of all the wells in the research area.
And 1.2, based on the production data in the production data set after data processing, fusing the oil reservoir data, the shaft data and the lifting equipment data in the production data set, and combining the production data of the same well according to the well number to form a sub-data set.
In this embodiment, for single data, the production data of the same well are combined to form a sub-data set according to the well number, and for the small-layer data of the oil reservoir including the porosity, the permeability, the oil saturation and the argillaceous content, the weighted average value of the oil reservoir thickness is used for replacing, and the thickness weighted calculation is performed on the measurement data of each layer to obtain the small-layer data of the oil reservoir formed by fusing the measurement data.
And 1.3, performing tag coding on the text type characteristic data in the production data set, converting the text type characteristic data (such as well cementation quality, whether sleeve damage exists or not, whether dog leg deflection influences pump down or pump failure type or not) assignment into digital type characteristic data, and performing standardized processing on various production data in the production data set to obtain an effective production data set.
Step 2, based on the effective production data set, screening comprehensive evaluation indexes for evaluating oil reservoirs, shaft and equipment according to a comprehensive evaluation index screening principle and feature importance ranking pairs, and constructing a comprehensive evaluation discrimination matrix, wherein the method comprises the following steps:
and 2.1, primarily screening the evaluation index for evaluating the oil reservoir-shaft-equipment according to the production index corresponding to various production data in the effective production data set based on the comprehensive evaluation index screening principle.
In this embodiment, the comprehensive evaluation index screening principle includes a comprehensive principle, a targeting principle, an operability principle, an independence principle and a participation principle, where the comprehensive principle is used to determine key features such as objectivity, scientificity, feasibility, comparability and stability of an evaluation object; the targeting principle is used for determining the target and task of the evaluation object, so that the contradiction or deviation between the evaluation index and the target of the evaluation object can be avoided; operability rules are used to ensure that the selected evaluation index is operable; the independence principle is used for ensuring that no repetition or conflict exists between the selected evaluation indexes; the participation principle is used for considering the comments and suggestions of the evaluation object and the evaluation stakeholder, improving the participation degree and representativeness of the evaluation and ensuring the fairness and credibility of the evaluation result.
And 2.2, carrying out correlation analysis on the evaluation indexes, and calculating the spearman correlation coefficient of each evaluation index to be used as an importance score.
2.3, constructing an XGBoost model, training the XGBoost model by using an effective production data set, verifying, and determining the importance of each evaluation index by using the verified XGBoost model, wherein the method comprises the following steps of:
And 2.3.1, dividing the production data in the effective production data set into an evaluation index and a target variable, wherein the target variable is set as daily oil production, setting the precision value of the XGBoost model, and dividing the effective production data set into a test set and a training set, wherein the quantity ratio of the production data in the test set to the production data in the training set is 1:2.
And 2.3.2, training the XGBoost model by using production data in a training set, acquiring production data corresponding to each evaluation index by using the XGBoost model, calculating importance scores of each evaluation index, outputting according to descending order of the importance scores, acquiring the calculation precision of the current XGBoost model in each training process, comparing with a preset precision value, entering the step 2.3.3 if the calculation precision of the XGBoost model reaches the preset precision value, otherwise, repeating the step 2.3.2 to continuously train the XGBoost model.
And 2.3.3, verifying that the calculation accuracy of the XGBoost model reaches a preset accuracy value by using a test set, and obtaining the verified XGBoost model.
Step 2.3.4, inputting the production data in the effective production data set into the validated XGBoost model, calculating the importance scores of all the evaluation indexes by using the XGBoost model, outputting the importance scores according to descending order, obtaining the importance ranking of all the evaluation indexes, and selecting the importance ranking before selecting The name evaluation index is used as a comprehensive evaluation index.
Step 2.4, constructing a comprehensive evaluation index matrix by using the production data corresponding to each comprehensive evaluation index, and normalizing the comprehensive evaluation index matrix to obtain a comprehensive evaluation discrimination matrix。
Step 3, determining the weight of each comprehensive evaluation index based on an analytic hierarchy process-entropy weight process, wherein the method comprises the following steps:
and 3.1, constructing a hierarchical structure model, wherein the hierarchical structure model comprises a target layer, a criterion layer and a scheme layer, wherein the target layer is internally provided with ranking of oil well production conditions in a research area, the criterion layer is internally provided with the comprehensive evaluation indexes obtained by screening in the step 2, and the scheme layer is internally provided with a well number.
Step 3.2, constructing a discrimination matrix by adopting an entropy weight methodAnd calculating the entropy weight of each comprehensive evaluation index.
In this embodiment, the decision matrix is constructed based on entropy weight methodThe method comprises the following steps of:
（1）
in the method, in the process of the invention,in the number of rows of the drawing,and is also provided withIs an integer of the number of the times,as a total number of samples,is of the number of columns andis an integer of the number of the times,the total number of the comprehensive evaluation indexes is calculated;to distinguish the first in the matrixLine 1Column sample data corresponding to the first valid production data setLine 1Production data of the columns.
Discriminant matrix based on linear proportional transformation method And (5) carrying out standardization processing to obtain a standardized matrix. And if the comprehensive evaluation index in the judgment matrix is a positive index, the positive index conversion formula is utilized to convert the comprehensive evaluation index into a processed sample data value, and if the comprehensive evaluation index in the judgment matrix is a negative index, the negative index conversion formula is utilized to convert the comprehensive evaluation index into a processed sample data value.
The transformation formula of the forward index is as follows:
（2）
in the method, in the process of the invention,to distinguish the first in the matrixLine 1The processed sample data value of the column sample data,as a function of the minimum value of the function,as a function of the maximum value.
The transformation formula of the negative index is as follows:
（3）
and calculating to obtain the entropy weight method weight of each comprehensive evaluation index, wherein the entropy weight method weight is shown in a formula (4):
（4）
wherein,
（5）
in the method, in the process of the invention,is the firstThe entropy weight of the comprehensive evaluation index,is the firstThe information entropy of the comprehensive evaluation index,is the firstThe first of the comprehensive evaluation indexesSpecific gravity of the individual sample data.
And 3.3, constructing a discrimination matrix based on an analytic hierarchy process, coupling an entropy weight method with the analytic hierarchy process, and replacing the entropy weight method weight of each comprehensive evaluation index by a scale to obtain the weight of each comprehensive evaluation index.
The discrimination matrix after the scale replacement is as follows:
（6）
In the method, in the process of the invention,is the discrimination matrix after scale replacement.
In this embodiment, the discrimination matrix is reconstructedThen, to the discriminant matrixConsistency checking by calculating a consistency ratioJudging whether the scale replacement is reasonable or not, and adjusting the judging matrixIs the order of (1) when the consistency ratio isWhen it is determined that the matrix isQualified for consistency check.
Step 4, discriminating matrix based on comprehensive evaluationAnd respectively calculating the scoring results of each well in the research area by using the TOPSIS comprehensive evaluation model, the gray correlation evaluation model, the entropy method comprehensive evaluation model and the rank and ratio method comprehensive evaluation model, wherein the scoring results comprise oil reservoir evaluation scores, well shaft evaluation scores, equipment evaluation scores and comprehensive evaluation scores.
And 5, optimizing scoring results based on a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy method comprehensive evaluation model and a rank sum ratio method comprehensive evaluation model to obtain an optimal comprehensive evaluation model and an optimal comprehensive score of an oil well, wherein the method specifically comprises the following steps of:
and 5.1, carrying out correlation analysis on the TOPSIS comprehensive evaluation model, the gray correlation evaluation model, the entropy method comprehensive evaluation model and the rank sum ratio method comprehensive evaluation model, calculating to obtain the spearman correlation coefficient of each evaluation model, and determining the correlation value among the evaluation models.
Step 5.2, constructing a correlation matrix based on a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy value method comprehensive evaluation model and a Stuffman correlation coefficient of a rank sum ratio method comprehensive evaluation modelCorrelation matrix->Taking the sum of correlation values between the selected evaluation model and other evaluation models as a preferable sample.
The correlation matrixThe method comprises the following steps:
（7）
in the method, in the process of the invention,the sum of correlation coefficients between the TOPSIS comprehensive evaluation model and the other three evaluation models,，；for the sum of the correlation coefficients between the grey correlation evaluation model and the other three evaluation models,，；the sum of correlation coefficients between the comprehensive evaluation model and the other three evaluation models by an entropy method,，；the sum of correlation coefficients between the rank sum ratio method comprehensive evaluation model and the other three evaluation models,，；、are all numbers of the evaluation model, whenOr (b)When the TOPSIS comprehensive evaluation model is expressed, whenOr (b)When representing the grey correlation evaluation model, whenOr (b)3, representing an entropy method comprehensive evaluation model whenOr (b)And 4, representing a rank sum ratio method comprehensive evaluation model.
And 5.3, selecting an evaluation model corresponding to the maximum value of the preferred sample as an optimal comprehensive evaluation model according to the preferred sample of each evaluation model, and obtaining the optimal comprehensive score of the oil well of each current well by using the optimal comprehensive evaluation model.
And 6, verifying the accuracy of the optimal comprehensive evaluation model by using the verification index.
And (3) pre-acquiring verification indexes of the research area well, analyzing and calculating the oil reservoir evaluation score, the well shaft evaluation score, the equipment evaluation score and the correlation coefficient between the comprehensive evaluation score and the verification indexes calculated by the optimal comprehensive evaluation model based on Kendall correlation coefficient, and checking the consistency between the optimal comprehensive evaluation model and the verification indexes.
The Kendall correlation coefficient calculation formula is as follows:
（8）
in the method, in the process of the invention,for the Kendall correlation coefficient,in order to be consistent for the number of pairs,in order to divide the number of pairs into the two,score sequences calculated for optimal comprehensive evaluation modelsThe number of parallel rows in the row is equal,is an authentication indexIs arranged in parallel.
Example 2
In this embodiment, the comprehensive evaluation method based on the oil reservoir-wellbore-equipment evaluation index integration disclosed in embodiment 1 is used to evaluate the field data of a certain oil field.
Step 1, based on oil reservoir data, shaft data and lifting equipment data, obtaining production data of all wells in a research area, constructing a production data set, and preprocessing the production data of all wells in the production data set to obtain an effective production data set.
In this embodiment, 2610 pieces of wellbore basic data, 26178 pieces of well rail related data and 5307 pieces of casing related data are obtained together, wherein the data include 74 pieces of casing damage related data, 13860 pieces of perforation related data and 4825 pieces of equipment basic data, and the data include 985 pieces of downhole operation data, 4825 pieces of equipment efficiency power data, 4825 pieces of energy consumption economic data, 94376 pieces of oil reservoir basic geological data, 4825 pieces of capacity data and 65 pieces of dynamic characteristic data.
Firstly, preprocessing production data of all wells in a research area, and carrying out tag coding on character characteristic data in the production data set to obtain a production data set. The production dataset in this embodiment comprises 65 sub-datasets, including current well category, reservoir top depth, reservoir bottom depth, reservoir middle, depth, original formation pressure, original saturation pressure, original formation temperature, wellhead lateral, wellhead longitudinal, well type, designed well depth, completion method, oil make-up distance, casing make-up distance, artificial bottom hole, bottom hole displacement, displacement azimuth, cement kick-up depth, well cementation quality, drilling purpose, maximum dog leg depth, maximum deflection, dog leg deflection under-influence pump, maximum well slant depth, maximum well slant gradient, drilling fluid mud density, stroke, frequency, oil pressure, casing pressure, back pressure, voltage, current, maximum active power, minimum active power, daily oil production, daily fluid production, daily water content, pump diameter, pump depth, pump efficiency, working fluid depth, travelling valve opening load, travelling valve closing load, fixed valve opening load, and pressure fixed valve shut-off load, up-stroke slope, down-stroke effective stroke, degree of fullness, up-stroke average load, down-stroke average load, maximum load, minimum load, pattern failure type, hydraulic power, motor power, ground efficiency, downhole efficiency, system efficiency, polished rod power, hundred meter ton night power consumption, month oil, month fluid, month water, year oil, year fluid, measured depth, vertical depth, well slant angle, azimuth angle, casing outside diameter, casing wall thickness, casing depth, casing inside diameter, casing total length, sand top depth, sand bottom depth, sand thickness, porosity, permeability, oil saturation, clay content, lithology, perforation density, perforation top depth, perforation bottom depth, perforation thickness, casing loss point, casing loss type, no-repair period, pump inspection period, dynamic liquid level change rate, submergence change rate, daily oil production change rate and water content change rate.
And 2, based on the effective production data set, screening comprehensive evaluation indexes for evaluating oil reservoirs, shafts and equipment according to a comprehensive evaluation index screening principle and feature importance ranking pairs, and constructing a comprehensive evaluation discrimination matrix.
In the embodiment, based on a comprehensive evaluation index screening principle, according to production indexes corresponding to various production data in an effective production data set, preliminary screening is carried out to obtain an evaluation index for evaluating oil reservoir-shaft-equipment, wherein shaft basic data comprises well completion depth, well completion method, cement return depth and well cementation quality, casing data comprises casing damage points, casing outer diameter, casing inner diameter, casing wall thickness and casing lower depth, perforation data comprises perforation total thickness and weighted perforation density, well rail data comprises well type, maximum dog leg degree, maximum deflection and dog leg deflection, pump is influenced, equipment basic data comprises stroke, stroke frequency, oil pressure, casing pressure, back pressure, pump diameter, pump efficiency, filling degree, up-stroke average load, down-stroke average load, maximum load and minimum load, the power data comprises hydraulic power, motor power and polished rod power, the efficiency data comprises ground efficiency, underground efficiency and system efficiency, the economic data comprises hundred-meter night power consumption, production time, production days, fault types, maintenance-free period and pump inspection period, the geological base data comprises sand layer thickness, production thickness, porosity, permeability, oil saturation and clay content, the capacity data comprises daily liquid yield, daily oil yield, daily water content, monthly oil yield, monthly liquid yield, monthly water yield, annual liquid yield, annual oil yield and annual water yield, and the dynamic data comprises working fluid level change rate, submergence change rate, daily oil yield change rate and water content change rate.
After the redundant indexes are removed, the XGBoost model which is trained and verified by the effective production data set is utilized to calculate the Szelman correlation coefficient of each evaluation index and is used as an importance score, so that the correlation of oil pressure and back pressure is 0.96, the correlation of daily oil yield and monthly oil yield is 0.68, the correlation of daily liquid yield and monthly liquid yield is 0.74, the correlation of daily water yield and monthly water yield is 0.77, and the correlation of production thickness and sand layer thickness is 0.94.
The method comprises the steps of performing secondary screening on evaluation indexes according to importance scores of all evaluation indexes, wherein shaft basic data comprise well completion depth, well completion method, cement return depth and well cementation quality, casing data comprise casing damage points, casing outer diameter, casing inner diameter, casing wall thickness and casing depth, perforation data comprise perforation total thickness and weighted perforation density, well rail data comprise well type, maximum dog leg degree, maximum deflection and dog leg deflection influence pump down, equipment basic data comprise stroke, stroke frequency, oil pressure, casing pressure, pump diameter, pump efficiency, full degree, upper stroke average load, lower stroke average load, maximum load and minimum load, power data comprise hydraulic power, motor power and polish rod power, efficiency data comprise ground efficiency, downhole efficiency and system efficiency, economic data comprise hundred-meter-ton night power consumption, production time, production days, fault type, maintenance-free period and pump detection period, and the equipment basic data comprise production thickness, porosity, permeability, oil saturation and clay content, the capacity data comprise daily liquid yield, the daily oil yield, the daily liquid level, the water content, the annual oil content, the water content and the water content changes, and the water content changes.
Ranking according to the importance of each evaluation index, preferably selecting important evaluation indexes, ranking according to the importance, and reserving the first 12 positions of oil reservoir related indexes, wherein the daily oil yield, the production thickness, the daily oil yield change rate, the daily liquid yield, the oil saturation, the porosity, the permeability, the water content change rate, the daily water content, the clay content, the submergence change rate, the working fluid level change rate, the annual oil yield, the annual water yield and the annual liquid yield are sequentially from high to low according to the importance; the related indexes of the shaft are reserved at the first 8 positions, and the quality of well cementation, the cement upward return depth, the maximum dog leg degree, the maximum disturbance degree, the dog leg degree disturbance degree influence pump, casing damage points, the wall thickness of the oil layer casing, the well completion depth, the well completion method, the outer diameter of the oil layer casing, the inner diameter of the oil layer casing, the depth under the casing, the total perforation thickness and the weighted perforation density are sequentially from high to low; the related indexes of the lifting equipment are reserved in the first 15 bits, and the related indexes are sequentially system efficiency, pump efficiency, stroke, motor power, stroke frequency, hundred-meter-ton night power consumption, oil pressure, casing pressure, production time, pump detection period, ground efficiency, maintenance-free period, downhole efficiency, production days, fault type, pump diameter, filling degree, minimum load, downstroke average load, upstroke average load and maximum load from high to low.
The final screening results in comprehensive evaluation indexes including well depth, cement return depth, well cementation quality, casing damage point, oil casing wall thickness, maximum dog leg degree, maximum deflection, pump under the influence of dog leg deflection, stroke frequency, oil pressure, casing pressure, pump efficiency, hydraulic power, motor power, polished rod power, ground efficiency, downhole efficiency, system efficiency, hundred-meter night power consumption, production time, production days, fault type, maintenance-free period, pump detection period, production thickness, porosity, permeability, oil saturation, clay content, daily liquid yield, daily oil yield, daily water content, sinking degree change rate, daily oil yield change rate and water content change rate.
Step 3, determining the weight of each comprehensive evaluation index based on an analytic hierarchy process-entropy weight process, the obtained well depth has index weight of 0.013, cement return depth of 0.013, well cementation quality of 0.088, casing damage point of 0.014, oil layer casing wall thickness of 0.014, maximum dog leg degree of 0.012, maximum deflection of 0.012, pump under influence of dog leg deflection of 0.015 the index weight of the stroke is 0.012, the index weight of the stroke frequency is 0.012, the index weight of the oil pressure is 0.014, the index weight of the casing pressure is 0.012, the index weight of the pump efficiency is 0.018, the index weight of the hydraulic power is 0.012, the index weight of the motor power is 0.012, the index weight of the system efficiency is 0.091, the index weight of the hundred-meter night power consumption is 0.089, the index weight of the production time is 0.018 the number of production days is 0.018, the number of failure types is 0.018, the number of maintenance-free periods is 0.018, the number of pump cycles is 0.024, the number of production thicknesses is 0.027, the number of porosities is 0.025, the number of permeability is 0.028, the number of oil saturation is 0.024, the number of clay content is 0.022, the number of daily liquid production is 0.028, the number of daily oil production is 0.190, the number of daily water content is 0.018, the number of polish rod power is 0.011, the number of ground efficiency is 0.014, the number of downhole efficiency is 0.009, the number of sink rate is 0.018, the number of oil production rate is 0.018, and the number of water content rate is 0.019.
And 4, respectively calculating the scoring results of each well in the research area by using a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy method comprehensive evaluation model and a rank sum ratio method comprehensive evaluation model, wherein the scoring results comprise oil reservoir evaluation scores, well bore evaluation scores, equipment evaluation scores and comprehensive evaluation scores.
And 5, optimizing the scoring results based on the TOPSIS comprehensive evaluation model, the gray correlation evaluation model, the entropy method comprehensive evaluation model and the rank sum ratio method comprehensive evaluation model to obtain an optimal comprehensive evaluation model and an optimal comprehensive score of the oil well.
In this embodiment, the results of the oil reservoir optimal comprehensive evaluation model are shown in table 1.
Table 1 results of the optimal comprehensive evaluation model of the oil reservoir;
。
the results of the wellbore optimal comprehensive evaluation model are shown in table 2.
Table 2 results of the wellbore optimal comprehensive evaluation model;
。
the results of the optimal comprehensive evaluation model of the equipment are shown in table 3.
Table 3 the results of the optimal comprehensive evaluation model of the equipment;
。
the results of the comprehensive evaluation index optimal comprehensive evaluation model are shown in table 4.
Table 4 comprehensive evaluation index optimal comprehensive evaluation model results;
。
and 6, verifying the accuracy of the optimal comprehensive evaluation model by using the verification index.
In this embodiment, the evaluation scores of the wells in the research area are obtained by respectively evaluating the wells in the research area by using an oil reservoir optimal comprehensive evaluation model, a well bore optimal comprehensive evaluation model, an equipment optimal comprehensive evaluation model and a comprehensive evaluation index optimal comprehensive evaluation model, wherein fig. 2 is a schematic diagram of the oil reservoir evaluation scores of the wells in the research area, fig. 3 is a schematic diagram of the well bore evaluation scores of the wells in the research area, fig. 4 is a schematic diagram of the equipment evaluation scores of the wells in the research area, and fig. 5 is a schematic diagram of the comprehensive evaluation index evaluation scores of the wells in the research area.
The analysis of the graphs 2-5 shows that the comprehensive evaluation index evaluation score has strong daily oil production correlation with the verification index thereof, the Kendall correlation value is 0.80, and the comprehensive evaluation index evaluation score is integrally increased from left to right and is consistent with the daily oil production trend; meanwhile, the correlation value of the oil reservoir evaluation score and the verification index thereof is 0.77, and the correlation value of the equipment evaluation score and the verification index thereof is 0.65, which shows that the oil reservoir score and the equipment score and the verification index thereof have good correlation, and as can be seen from the combination of FIG. 3, the higher the well cementation quality relation between the well shaft evaluation score and the verification index thereof is, the higher the well shaft score is, the well cementation quality is 0, the character type is qualified for the well cementation quality, the well shaft score is low, the well cementation quality of a plurality of wells is 1, and the character type is unqualified for the well cementation quality, thereby verifying that the evaluation result of the method has higher accuracy.
In conclusion, the method can accurately evaluate the oil well production condition in multiple directions, is beneficial to on-site workers to improve the cognition of the current oil well production condition, and provides directional information for oil well optimization.
It should be understood that the above description is not intended to limit the invention to the particular embodiments disclosed, but to limit the invention to the particular embodiments disclosed, and that the invention is not limited to the particular embodiments disclosed, but is intended to cover modifications, adaptations, additions and alternatives falling within the spirit and scope of the invention.
Claims (8)
1. The comprehensive evaluation method based on the oil reservoir-shaft-equipment evaluation index integration is characterized by comprising the following steps of:
step 1, acquiring production data of all wells in a research area based on oil reservoir data, shaft data and lifting equipment data, constructing a production data set, and preprocessing the production data of each well in the production data set to obtain an effective production data set;
step 2, based on the effective production data set, screening comprehensive evaluation indexes for evaluating oil reservoirs, shaft and equipment according to a comprehensive evaluation index screening principle and feature importance ranking pairs, and constructing a comprehensive evaluation discrimination matrix;
step 3, determining the weight of each comprehensive evaluation index based on an analytic hierarchy process-entropy weight process;
Step 4, based on the comprehensive evaluation discrimination matrix, calculating the scoring results of each well in the research area respectively by using a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy comprehensive evaluation model and a rank sum ratio comprehensive evaluation model, wherein the scoring results comprise oil reservoir evaluation scores, shaft evaluation scores, equipment evaluation scores and comprehensive evaluation scores;
step 5, optimizing scoring results based on the TOPSIS comprehensive evaluation model, the gray correlation evaluation model, the entropy method comprehensive evaluation model and the rank sum ratio method comprehensive evaluation model to obtain an optimal comprehensive evaluation model and an optimal comprehensive score of the oil well;
step 6, verifying the accuracy of the optimal comprehensive evaluation model by using the verification index;
the step 2 comprises the following steps:
step 2.1, based on a comprehensive evaluation index screening principle, primarily screening evaluation indexes for evaluating oil reservoirs, shaft and equipment according to production indexes corresponding to various production data in an effective production data set;
step 2.2, carrying out correlation analysis on the evaluation indexes, and calculating the spearman correlation coefficient of each evaluation index to be used as an importance score;
step 2.3, constructing an XGBoost model, training the XGBoost model by using an effective production data set, verifying, and determining the importance of each evaluation index by using the verified XGBoost model;
Step 2.4, constructing a comprehensive evaluation index matrix by using the production data corresponding to each comprehensive evaluation index, and normalizing the comprehensive evaluation index matrix to obtain a comprehensive evaluation discrimination matrix；
The step 5 comprises the following steps:
step 5.1, carrying out correlation analysis on a TOPSIS comprehensive evaluation model, a gray correlation evaluation model, an entropy method comprehensive evaluation model and a rank sum ratio method comprehensive evaluation model, calculating a spearman correlation coefficient of each evaluation model, and determining a correlation value among the evaluation models;
step 5.2, constructing a correlation matrix by taking the sum of correlation values between the selected evaluation model and other evaluation models as a preferred sampleAs shown in formula (7):
（7）
in the method, in the process of the invention,the sum of correlation coefficients between the TOPSIS comprehensive evaluation model and the other three evaluation models,，/>；/>for the sum of the correlation coefficients between the grey correlation evaluation model and the other three evaluation models, ++>，/>；/>For the sum of correlation coefficients between the entropy method comprehensive evaluation model and the other three evaluation models, ++>，/>；/>For the sum of correlation coefficients between the rank sum ratio comprehensive evaluation model and the other three evaluation models, ++>，/>；/>、/>Are all numbers of the evaluation model, when +. >Or->At the same time, the TOPSIS comprehensive evaluation model is represented, when +.>Or->When ∈K represents gray correlation evaluation model, when ∈K->Or->3, representing an entropy method comprehensive evaluation model, when +.>Or->4, representing a rank sum ratio method comprehensive evaluation model;
and 5.3, selecting an evaluation model corresponding to the maximum value of the preferred sample as an optimal comprehensive evaluation model according to the preferred sample of each evaluation model, and obtaining the optimal comprehensive score of the oil well of each current well by using the optimal comprehensive evaluation model.
2. The comprehensive evaluation method based on oil reservoir-wellbore-equipment evaluation index integration according to claim 1, wherein in the step 1, the method comprises the following steps:
step 1.1, acquiring production data of all wells in a research area, constructing a production data set, determining that each production data is digital characteristic data or character characteristic data according to the type of each production data in the production data set, performing data cleaning on the production data in the production data set, removing abnormal oil well production data in the oil well production data set, and performing blank value filling on the production data missing in the production data set;
step 1.2, based on the production data in the production data set after data processing, the oil reservoir data, the shaft data and the lifting equipment data in the production data set are fused, and the production data of the same well are combined to form a sub-data set according to the well number;
And 1.3, performing tag coding on the text type characteristic data in the production data set, performing standardized processing on various production data in the production data set after converting the text type characteristic data assignment into digital type characteristic data, and obtaining an effective production data set.
3. The comprehensive evaluation method based on oil reservoir-well bore-equipment evaluation index integration according to claim 2, wherein after searching abnormal production data in the production data set to determine data abnormal wells, deleting the production data of all the data abnormal wells in the production data set for data cleaning;
searching missing production data in the production data set after data cleaning, determining a well with a blank value of the production data as a data missing well, acquiring a well number of the data missing well, and judging the missing production data as digital characteristic data or character characteristic data;
when the missing production data is digital characteristic data, filling blank values of the data missing wells according to the well numbers, if the same type of production data exists in the wells positioned in the same row with the data missing wells, filling the production data of the data missing wells by using the average value of the same type of production data of all the wells positioned in the same row, and if the same type of production data does not exist in the wells positioned in the same row with the data missing wells, filling the blank values of the data missing wells by using the average value of the same type of production data of all the wells in the research area;
And when the missing production data is character feature data, filling the blank value of the production data of the data missing well according to the well number, if the same type of production data exists in the wells positioned in the same row with the data missing well, filling the production data of the data missing well by using the production data corresponding to the mode of the same type of production data in all the wells in the same row, and if the same type of production data does not exist in the wells positioned in the same row with the data missing well, filling the blank value of the data missing well by using the production data corresponding to the mode of the same type of production data of all the wells in the research area.
4. The comprehensive evaluation method based on oil reservoir-wellbore-equipment evaluation index integration according to claim 1, wherein in the step 2.3, the method specifically comprises the following steps:
step 2.3.1, dividing production data in an effective production data set into an evaluation index and a target variable, setting an accuracy value of an XGBoost model, and dividing the effective production data set into a test set and a training set;
step 2.3.2, training an XGBoost model by using production data in a training set, acquiring production data corresponding to each evaluation index by using the XGBoost model, calculating importance scores of each evaluation index, outputting according to descending order of the importance scores, acquiring the calculation precision of the current XGBoost model in each training process, comparing with a preset precision value, entering the step 2.3.3 if the calculation precision of the XGBoost model reaches the preset precision value, otherwise, repeating the step 2.3.2 to continuously train the XGBoost model;
2.3.3, verifying that the calculation accuracy of the XGBoost model reaches a preset accuracy value by using a test set to obtain a verified XGBoost model;
step 2.3.4, inputting the production data in the effective production data set into the validated XGBoost model, calculating the importance scores of all the evaluation indexes by using the XGBoost model, outputting the importance scores according to descending order, obtaining the importance ranking of all the evaluation indexes, and selecting the importance ranking before selectingThe name evaluation index is used as a comprehensive evaluation index.
5. The method for comprehensive evaluation based on reservoir-wellbore-equipment evaluation index integration according to claim 1, wherein the step 3 comprises the following steps:
step 3.1, constructing a hierarchical structure model, wherein the hierarchical structure model comprises a target layer, a criterion layer and a scheme layer, wherein the target layer is internally provided with ranking of oil well production conditions in a research area, the criterion layer is internally provided with the comprehensive evaluation indexes obtained by screening in the step 2, and the scheme layer is internally provided with a well number;
step 3.2, constructing a discrimination matrix by adopting an entropy weight methodCalculating the entropy weight of each comprehensive evaluation index;
and 3.3, constructing a discrimination matrix based on an analytic hierarchy process, coupling an entropy weight method with the analytic hierarchy process, and replacing the entropy weight method weight of each comprehensive evaluation index through a scale to obtain the weight of each comprehensive evaluation index.
6. The comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration according to claim 5, wherein the judgment matrix constructed based on entropy weight methodThe method comprises the following steps:
（1）
in the method, in the process of the invention,for the number of lines->And->Is an integer>For the total number of samples->Is the column number and->Is an integer>The total number of the comprehensive evaluation indexes is calculated; />To distinguish the->Line->Column sample data corresponding to the +.>Line->Production data of columns；
Normalized processing discrimination matrix based on linear proportional conversion methodObtaining a normalization matrix;
if the comprehensive evaluation index in the judgment matrix is a positive index, a positive index transformation formula is utilized to transform to obtain a processed sample data value, and if the comprehensive evaluation index in the judgment matrix is a negative index, a negative index transformation formula is utilized to transform to obtain a processed sample data value;
the transformation formula of the forward index is as follows:
（2）
in the method, in the process of the invention,to distinguish the->Line->Sample data value after processing column sample data, < >>As a function of the minimum value +.>Is a maximum function;
the transformation formula of the negative index is as follows:
（3）
and calculating the entropy weight of each comprehensive evaluation index, as shown in a formula (4):
（4）
Wherein,
（5）
in the method, in the process of the invention,is->Entropy weight of comprehensive evaluation index, < ->Is->Information entropy of each comprehensive evaluation index, +.>Is->The first part of the comprehensive evaluation index>Specific gravity of the individual sample data;
constructing a discrimination matrix based on an analytic hierarchy process, and coupling an entropy weight method with the analytic hierarchy process to obtain the discrimination matrix after scale replacementThe method comprises the following steps:
（6）。
7. the comprehensive evaluation method based on oil reservoir-wellbore-equipment evaluation index integration according to claim 5, wherein in the step 4, the TOPSIS comprehensive evaluation model calculation process is as follows: determining positive ideal solutions and negative ideal solutions by taking each well in a research area as an evaluation object, and then calculating the distances between each evaluation object and the positive ideal solutions and the negative ideal solutions to obtain the relative proximity of each evaluation object to the ideal solutions, so as to determine the TOPSIS comprehensive evaluation model scores of each evaluation object;
the gray correlation evaluation model calculation process comprises the following steps: taking each well in a research area as an evaluation object, firstly selecting optimal sample data of each comprehensive evaluation index, constructing an optimal index set, and respectively calculating gray correlation coefficients corresponding to each comprehensive evaluation index in each evaluation object to obtain gray correlation of each evaluation object to an ideal solution so as to determine a gray correlation evaluation model score of each evaluation object;
The calculation process of the entropy method comprehensive evaluation model is as follows: taking each well in the research area as an evaluation object, and calculating the target score of each evaluation object based on the entropy weight method weight of each comprehensive evaluation index calculated in the step 3.2, thereby determining the entropy method comprehensive evaluation model score of each evaluation object;
the calculation process of the rank sum ratio method comprehensive evaluation model is as follows: using each well in the research area as an evaluation object, calculating the rank of each comprehensive evaluation index in the evaluation object based on the comprehensive evaluation discrimination matrix to obtain a rank matrixAnd then, calculating the rank and the ratio of each evaluation object, thereby determining the rank and the ratio method comprehensive evaluation model score of each evaluation object.
8. The comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration according to claim 1, wherein in the step 6, verification indexes of the research area well are obtained in advance, oil reservoir evaluation scores, shaft evaluation scores, equipment evaluation scores and correlation coefficients between the comprehensive evaluation scores and the verification indexes calculated by the optimal comprehensive evaluation model are calculated based on Kendall correlation coefficient analysis, and consistency between the optimal comprehensive evaluation model and the verification indexes is checked;
The Kendall correlation coefficient calculation formula is as follows:
（8）
in the method, in the process of the invention,for Kendall correlation coefficient, +.>For the same number of pairs, add>For branching the number of pairs->Score sequence calculated for optimal comprehensive evaluation model +.>The number of parallel rows in->Is verification index->Is arranged in parallel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410016502.7A CN117541082B (en) | 2024-01-05 | 2024-01-05 | Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410016502.7A CN117541082B (en) | 2024-01-05 | 2024-01-05 | Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117541082A CN117541082A (en) | 2024-02-09 |
| CN117541082B true CN117541082B (en) | 2024-04-05 |
Family
ID=89796171
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410016502.7A Active CN117541082B (en) | 2024-01-05 | 2024-01-05 | Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117541082B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2766437A1 (en) * | 2009-08-12 | 2011-02-17 | Exxonmobil Upstream Research Company | Optimizing well management policy |
| CN105469196A (en) * | 2015-11-18 | 2016-04-06 | 山东科技大学 | Comprehensive evaluation method and comprehensive evaluation system for evaluating mine construction project process |
| CN109377062A (en) * | 2018-10-31 | 2019-02-22 | 中国石油大学(华东) | A kind of oil field Service Component evaluation method based on improvement entropy weight gray relative analysis method |
| CN112561356A (en) * | 2020-12-21 | 2021-03-26 | 西安石油大学 | Comprehensive evaluation method for production dynamic mode and productivity of shale oil fracturing horizontal well |
| CN113537585A (en) * | 2021-07-09 | 2021-10-22 | 中海石油（中国）有限公司天津分公司 | Oil field production increasing measure recommendation method based on random forest and gradient lifting decision tree |
| CN116362447A (en) * | 2021-12-24 | 2023-06-30 | 中国石油天然气集团有限公司 | Drilling construction difficulty evaluation method based on analytic hierarchy process |
| CN116384618A (en) * | 2021-12-21 | 2023-07-04 | 中国石油天然气股份有限公司 | Method and system for profile control and flooding well selection of large well spacing oil reservoir |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109933745B (en) * | 2019-01-29 | 2022-01-28 | 西南石油大学 | Hydrate drilling and production risk intelligent judgment method based on fuzzy judgment |
| CN111441767B (en) * | 2020-05-11 | 2022-05-20 | 中国石油大学(华东) | Oil reservoir production dynamic prediction method and device |
| CN111861774B (en) * | 2020-06-22 | 2021-06-04 | 中国石油大学(华东) | Oil reservoir production machine learning method based on parallel agent model |
-
2024
- 2024-01-05 CN CN202410016502.7A patent/CN117541082B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2766437A1 (en) * | 2009-08-12 | 2011-02-17 | Exxonmobil Upstream Research Company | Optimizing well management policy |
| CN105469196A (en) * | 2015-11-18 | 2016-04-06 | 山东科技大学 | Comprehensive evaluation method and comprehensive evaluation system for evaluating mine construction project process |
| CN109377062A (en) * | 2018-10-31 | 2019-02-22 | 中国石油大学(华东) | A kind of oil field Service Component evaluation method based on improvement entropy weight gray relative analysis method |
| CN112561356A (en) * | 2020-12-21 | 2021-03-26 | 西安石油大学 | Comprehensive evaluation method for production dynamic mode and productivity of shale oil fracturing horizontal well |
| CN113537585A (en) * | 2021-07-09 | 2021-10-22 | 中海石油（中国）有限公司天津分公司 | Oil field production increasing measure recommendation method based on random forest and gradient lifting decision tree |
| CN116384618A (en) * | 2021-12-21 | 2023-07-04 | 中国石油天然气股份有限公司 | Method and system for profile control and flooding well selection of large well spacing oil reservoir |
| CN116362447A (en) * | 2021-12-24 | 2023-06-30 | 中国石油天然气集团有限公司 | Drilling construction difficulty evaluation method based on analytic hierarchy process |
Non-Patent Citations (3)
| Title |
|---|
| Optimized production profie using a coupled reservoir network model;A. Hoffann等;《Journal of Petroleum Exploration and Production Technology》;20190124;第9卷;第2123–2137页 * |
| 候选注气油藏筛选的指标综合权重优化新方法;熊钰, 熊玉滨, 陈天钢, 李拥军;西南石油学院学报;20040430(第02期);全文 * |
| 油藏-井筒-举升设备动态耦合模型及应用;马高强;《中国优秀博士学位论文全文数据库 工程科技I辑》;20230215(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117541082A (en) | 2024-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104533400B (en) | Method for reconstructing logging curve | |
| CN104298883B (en) | A kind of method for building up of the hydrocarbon source rock hydrocarbon producing rate plate in oil and gas resource evaluation | |
| CN105626009A (en) | Fracture-cavern type carbonate oil reservoir single well water injection oil substituting effect quantitative evaluation method | |
| CN109214026A (en) | A kind of shale gas horizontal well initial productivity prediction technique | |
| CN107895092B (en) | Inter-well communication quantitative evaluation method based on complex nonlinear injection-production modeling | |
| CN113570226A (en) | Method for evaluating occurrence probability grade of tunnel water inrush disaster in fault fracture zone | |
| CN104899411A (en) | Method and system for establishing reservoir capacity prediction model | |
| CN115030707A (en) | Rapid evaluation method of oil shale dessert | |
| CN117541082B (en) | Comprehensive evaluation method based on oil reservoir-shaft-equipment evaluation index integration | |
| CN112052977A (en) | Oil reservoir reserve prediction method based on deep space-time attention network | |
| CN114648153A (en) | Dense oil dessert optimization method and system based on gray matrix weighting method | |
| CN115705452A (en) | Novel recovery ratio prediction method for middle and later stages of integrated sandstone reservoir development | |
| CN107818189B (en) | Method for calculating recoverable resource amount of shale gas | |
| CN111080167A (en) | Underground space resource quality assessment method for urban planning | |
| CN116011234B (en) | Pressure front sleeve change risk level judgment method integrating geomechanics and Bayes | |
| CN114562236B (en) | Geological engineering integrated lost circulation real-time early warning method based on integrated learning model | |
| Feder | Machine-learning approach determines spatial variation in shale decline curves | |
| CN117522173B (en) | Natural gas hydrate depressurization exploitation productivity prediction method based on deep neural network | |
| CN114676631B (en) | Shale gas reservoir fracturing horizontal well fracturing transformation parameter recommendation method | |
| CN107944749B (en) | Shale gas block development potential evaluation method based on relative preference index | |
| CN115936465A (en) | Method for calculating water cone critical yield of fractured-vuggy carbonate reservoir and application thereof | |
| CN116304921A (en) | Intelligent interpretation method for reservoir diagenetic facies logging | |
| CN117951859A (en) | Method for predicting capacity of horizontal well of tight conglomerate oil reservoir based on drilling and logging data | |
| CN114418252A (en) | Comprehensive evaluation screening method for geothermal field | |
| CN116263106A (en) | Data-driven casing well screening method |
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 |