CN115271182A - Offshore oilfield water flooding recovery rate prediction method - Google Patents

Offshore oilfield water flooding recovery rate prediction method Download PDF

Info

Publication number
CN115271182A
CN115271182A CN202210812608.9A CN202210812608A CN115271182A CN 115271182 A CN115271182 A CN 115271182A CN 202210812608 A CN202210812608 A CN 202210812608A CN 115271182 A CN115271182 A CN 115271182A
Authority
CN
China
Prior art keywords
oil
reservoir
reservoirs
namely
meters
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210812608.9A
Other languages
Chinese (zh)
Inventor
吴春新
刘英宪
王少鹏
张宏友
马奎前
罗宪波
侯东梅
邓琪
刘美佳
原建伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CNOOC China Ltd Tianjin Branch
Original Assignee
CNOOC China Ltd Tianjin Branch
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CNOOC China Ltd Tianjin Branch filed Critical CNOOC China Ltd Tianjin Branch
Priority to CN202210812608.9A priority Critical patent/CN115271182A/en
Publication of CN115271182A publication Critical patent/CN115271182A/en
Pending legal-status Critical Current

Links

Images

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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B43/00—Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells
    • E21B43/16—Enhanced recovery methods for obtaining hydrocarbons
    • E21B43/20—Displacing by water
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; 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
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
    • Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mining & Mineral Resources (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Geology (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Strategic Management (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Biophysics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Fluid Mechanics (AREA)
  • Molecular Biology (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geochemistry & Mineralogy (AREA)
  • Biomedical Technology (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Primary Health Care (AREA)
  • Marine Sciences & Fisheries (AREA)

Abstract

A method for predicting the recovery ratio of water flooding of an offshore oilfield comprises the following steps: firstly, the following steps: determining an analog oil field according to the geological oil reservoir characteristics of a research object; II, secondly, the method comprises the following steps: performing history fitting on the analog oilfield production dynamic data; thirdly, the method comprises the following steps: predicting the recovery ratio of the analogy oil field, and carrying out the sensitivity analysis of the recovery ratio; fourthly, the method comprises the following steps: constructing a machine learning model based on the recovery ratio prediction result; fifthly: and predicting the recovery ratio of the research object by using the trained machine learning model. The invention not only considers the particularity of offshore oilfield development; in addition, when the recovery ratio is quantitatively predicted, a machine learning model is also constructed, so that subjective factors are avoided, and the method has strong objectivity; meanwhile, the problem of few sample points in the conventional machine learning is avoided during machine learning; the problem of difficult prediction of the offshore oilfield water flooding recovery efficiency is solved.

Description

Offshore oilfield water flooding recovery rate prediction method
Technical Field
The invention belongs to the field of oil and gas field exploration and development, and particularly relates to a method for predicting the water flooding recovery ratio of an offshore oil field.
Background
Recovery factor is an important parameter for evaluating the development effect of an oil field, and is always a concern of reservoir workers in the design and adjustment work of a development scheme.
As natural energy of the Bohai sea oil field is insufficient, 86% of reserves need to be injected with water to develop and supplement energy; and because the offshore oil field has higher investment cost, reasonable recovery ratio and various measure development effects need to be determined, and engineering reservation is made. Therefore, the research on the offshore oilfield water flooding recovery efficiency calculation model has important significance for providing evaluation basis for developing the oilfield by water injection.
Currently, the prediction methods for recovery mainly include: static and dynamic methods, wherein the static method is suitable for undeveloped fields and developed early fields, and mainly comprises: a class comparison method and an empirical formula method; the dynamic rule is suitable for the oil field with certain development time, rich production dynamic data and certain development rule, and mainly comprises the following steps: water flooding curve method, decreasing curve method, numerical simulation method, etc. However, the current methods for predicting recovery efficiency are not suitable to some extent in offshore applications, such as: analog method in static method: it has strong subjectivity in quantitative prediction of recovery ratio; empirical formula method: generally from land fields, there are relatively many measures and relatively high well pattern densities compared to offshore fields. And the dynamic method: more dynamic data is required to be produced, and offshore oil fields are limited by cost, and relatively less data is obtained. Therefore, it is urgently needed to establish a recovery prediction method for offshore oil fields to guide recovery prediction of offshore oil fields.
Disclosure of Invention
The invention aims to provide a method for predicting the water-drive recovery ratio of an offshore oil field, which aims to solve the technical problem of predicting the water-drive recovery ratio of the offshore oil field.
In order to achieve the purpose, the specific technical scheme of the offshore oilfield water flooding recovery prediction method is as follows:
a method for predicting the recovery ratio of water flooding in an offshore oilfield comprises the following steps: the method comprises the following steps:
the first step is as follows: determining an analog oil field according to the geological oil reservoir characteristics of a research object;
the second step: performing history fitting on the analog oilfield production dynamic data;
the third step: predicting the recovery ratio of the analogy oil field, and carrying out the sensitivity analysis of the recovery ratio;
the fourth step: constructing a machine learning model based on the recovery ratio prediction result;
the fifth step: and predicting the recovery ratio of the research object by using the trained machine learning model.
Further, in the first step, the study objects are: combining well drilling, well logging and seismic data to determine the geological reservoir characteristics of the research object, wherein the main parameters are used for representing the geological reservoir characteristics; the method comprises the following steps: the method comprises the following steps of measuring an oil-bearing horizon, a sedimentary facies, a reservoir type, reservoir lithology, a driving type, reservoir burial depth, effective thickness, porosity, permeability and formation crude oil viscosity, wherein the characteristic parameters of the oil-bearing horizon, the sedimentary facies, the reservoir type, the reservoir lithology and the driving type are measured by adopting qualitative indexes, and the other 5 characteristic parameters of the reservoir burial depth, the effective thickness, the porosity, the permeability and the formation crude oil viscosity are measured by adopting quantitative indexes.
Further, in the first step, when selecting the analog oil field, the research object and the analog oil field are required to be the same with the parameters measured by the qualitative indexes, the same oil field is difficult to find for the parameters measured by the quantitative indexes, and the parameters are selected according to the following steps:
1. firstly, qualitatively characterizing the parameters according to national standards, row standards or enterprise standards, and requiring that the qualitative division results of the parameters measured by quantitative indexes are the same for a research object and an analog oil field when the analog oil field is selected;
2. according to the oil reservoir burial depth, the oil reservoir is divided into 5 types: one type is shallow reservoirs, namely: the oil deposit burial depth is less than 500 meters; the second category is medium-shallow reservoirs, namely: the oil deposit burial depth is more than or equal to 500 meters and less than 2000 meters; the three types are medium-deep oil reservoirs, namely: the oil deposit burial depth is more than or equal to 2000 meters and less than 3500 meters; the four categories are deep reservoirs, namely: the oil deposit depth is greater than or equal to 3500 meters and is less than 4500 meters, and five types of oil deposits are ultra-deep oil deposits, namely: the oil deposit burial depth is more than or equal to 4500 meters;
3. the oil reservoirs are divided into 4 types according to the thickness of the oil reservoirs: one type is thin layer reservoirs, namely: the thickness of the oil deposit is less than 5 meters; the second category is medium-heavy reservoir, namely: the thickness of the oil reservoir is more than or equal to 5 meters and less than 20 meters; three types are thick-layer reservoirs, namely: the thickness of the oil reservoir is more than or equal to 20 meters and less than 40 meters; four types are extra-thick reservoirs, namely: the thickness of the oil reservoir is more than or equal to 40 meters;
4. oil reservoirs are classified into 4 categories according to porosity: one type is an ultra-low porosity reservoir, namely: the porosity of the oil reservoir is less than 10 percent; the second category is low porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 10 percent and less than 15 percent; three categories are medium porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 15 percent and less than 25 percent; four categories are high porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 25 percent;
5. the reservoirs were classified into 5 categories according to permeability: one type is an ultra-low permeability reservoir, namely: the oil reservoir permeability is less than 5 millidarcy; the second category is low permeability reservoirs, namely: the oil reservoir permeability is more than or equal to 5 millidarcy and less than 50 millidarcy; three types are medium permeability reservoirs, namely: the oil reservoir permeability is more than or equal to 50 millidarcy and less than 500 millidarcy; the four categories are high permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 500 millidarcies and less than 1000 millidarcies; the five types are ultra-high permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 1000 millidarcy;
6. oil reservoirs are classified into 4 categories according to the viscosity of the crude oil in the stratum: one category is low-oil reservoirs, namely: the viscosity of the crude oil in the stratum is less than 5 mPas; the second category is medium-viscosity oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 5 mPas and less than 20 mPas; the three types are high-viscosity oil reserves, namely: the viscosity of the crude oil in the stratum is more than or equal to 20 mPas and less than 50 mPas; the four types are heavy oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 50 mPas;
and determining an analog oil field according to the geological oil reservoir characteristics of the research object, and selecting a production oil field with the same or similar horizon, sedimentary facies, oil reservoir type, reservoir lithology, driving type, oil reservoir burial depth, effective thickness, porosity, permeability and formation crude oil viscosity as the analog object.
Further, in the second step, a history fitting process is performed on the analog oilfield production dynamic data: due to the limitation of understanding the geological condition of the oil reservoir during modeling, the numerical simulation model cannot truly reflect the actual condition of the oil reservoir, and the calculated values and the actual values of the main dynamic indexes of pressure, oil production and water content in oil field development are matched by repeatedly and continuously adjusting the static parameters of the oil reservoir.
Further, in the second step, historical data of a plurality of oil fields are input into corresponding geological models, and historical fitting work is carried out, wherein the historical fitting work comprises the following steps: oil field fitting and single well fitting.
Further, in the third step, on the basis of the history fitting of the second step, recovery factor prediction is carried out, and sensitivity analysis is carried out on several oil fields by adjusting model parameters.
Further, in the third step, the recovery ratio of the simulated oil field is predicted, and recovery ratio sensitivity analysis is carried out: on the basis of historical fitting, numerical simulation research is carried out, the recovery ratio of the existing model is predicted, sensitivity analysis of parameters such as horizontal permeability, horizontal and vertical permeability ratio, the skin, the viscosity of crude oil in stratum, well pattern density, production pressure difference, water injection time, extraction liquid multiple and the like on the recovery ratio is carried out, and the water flooding recovery ratio under different conditions is predicted.
Further, in the fourth step, the recovery factor prediction results and the sensitivity analysis results of the plurality of oil field basic models in the third step are input into a machine learning model, and the selected machine learning model is as follows: a BP neural network; constructing a machine learning model based on the recovery prediction result: and (3) forming a sample set by recovery ratio prediction results of all the analog oil fields and recovery ratios predicted by sensitivity analysis, randomly selecting a part of samples as a training set, using the rest of samples as a test set, selecting parameters of the sensitivity analysis in the third step as characteristic values of recovery ratio prediction, training the training set by using a machine learning model, carrying out recovery ratio prediction on the test set by using the trained machine learning model, calculating the prediction precision of the test set, stopping training if the test precision meets an expected requirement, and adjusting related parameters of the machine learning model until the prediction precision of the test set meets the expected requirement if the test precision does not meet the expected requirement.
Further, in the fifth step, the trained machine learning model is used for predicting the recovery ratio of the research object: and inputting the characteristic parameters of the research object into the machine learning model trained in the fourth step, wherein the characteristic parameters are the parameters of the sensitivity analysis in the third step, and obtaining a recovery factor prediction result.
The offshore oilfield water flooding recovery prediction method has the following advantages:
1. the offshore oil field is selected as the analog oil field, the particularity of offshore oil field development is fully considered, and the applicability is stronger compared with the prediction recovery ratio of an empirical formula;
2. when the recovery ratio is quantitatively predicted, a machine learning model is constructed, machine learning samples are from an analog oil field, but compared with a conventional analog method, subjective factors are avoided to a certain extent, and the method has stronger objectivity;
3. when the machine learning is used, the method adopts the result of the recovery factor sensitivity analysis, and avoids the problem of few sample points in the conventional machine learning.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic of the oilfield pressure fitting of the present invention; (which is the actual pattern on the screen)
FIG. 3 is a schematic diagram of the oilfield production fitting of the present invention; (which is the actual pattern on the screen)
FIG. 4 is a schematic representation of an oilfield hydration fit of the present invention; (which is the actual pattern on the screen)
FIG. 5 is a schematic diagram of oilfield single well pressure fitting of the present invention; (which is the actual pattern on the screen)
FIG. 6 is a schematic of an oilfield single well production fit of the present invention; (which is the actual pattern on the screen)
FIG. 7 is a schematic representation of the oilfield single well water content fitting of the present invention. (which is the actual pattern on the screen)
Detailed Description
In order to better understand the purpose, structure and function of the present invention, the following describes a method for predicting the recovery ratio of water flooding in offshore oil field in further detail with reference to the attached drawings.
As shown in fig. 1 to 7, the present invention employs the following steps:
the first step is as follows: determining an analog oil field according to the geological oil reservoir characteristics of a research object;
it specifically combines well drilling, logging, well logging and seismic data, carries out geological research to confirm the geology oil deposit characteristic in subject X oil field, be used for the main parameter of characterization geology oil deposit characteristic, include: the method comprises the following steps of measuring an oil-bearing horizon, a sedimentary facies, an oil reservoir type, reservoir lithology, a driving type, an oil reservoir burial depth, effective thickness, porosity, permeability and formation crude oil viscosity, wherein 5 characteristic parameters of the oil-bearing horizon, the sedimentary facies, the oil reservoir type, the reservoir lithology and the driving type are measured by adopting qualitative indexes, and the other 5 characteristic parameters of the oil reservoir burial depth, the effective thickness, the porosity, the permeability and the formation crude oil viscosity are measured by adopting quantitative indexes.
1) When selecting an analog oil field, requiring the research object and the analog oil field to have the same parameters measured by the qualitative indexes, and difficultly finding the same oil field for the parameters used for measuring the quantitative indexes, firstly, qualitatively characterizing the parameters according to national standards, row standards or enterprise standards, and when selecting the analog oil field, requiring the research object and the analog oil field to have the same qualitative division results of the parameters measured by the quantitative indexes, wherein the qualitative division results of the quantitative parameters of the embodiment are as follows;
2) According to the oil reservoir burial depth, the oil reservoir is divided into 5 types: one type is shallow reservoirs, namely: the oil deposit burial depth is less than 500 meters; the second category is medium-shallow reservoirs, namely: the oil deposit burial depth is more than or equal to 500 meters and less than 2000 meters; the three types are medium-deep reservoirs, namely: the oil deposit burial depth is more than or equal to 2000 meters and less than 3500 meters; the four categories are deep reservoirs, namely: the oil deposit depth is more than or equal to 3500 meters, and is less than 4500 meters, and five types are ultra-deep oil deposits, namely: the oil deposit burial depth is more than or equal to 4500 meters;
3) The oil reservoirs are divided into 4 types according to the thickness of the oil reservoirs: one type is thin layer reservoirs, namely: the thickness of the oil reservoir is less than 5 m; the second category is medium-heavy reservoir, namely: the thickness of the oil reservoir is more than or equal to 5 meters and less than 20 meters; three types are thick-layer reservoirs, namely: the thickness of the oil reservoir is more than or equal to 20 meters and less than 40 meters; four types are extra-thick reservoirs, namely: the thickness of the oil reservoir is more than or equal to 40 meters;
4) Oil reservoirs are classified into 4 categories according to porosity: one type is an ultra-low porosity reservoir, namely: the oil deposit porosity is less than 10%; the second category is low porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 10 percent and less than 15 percent; three categories are medium porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 15 percent and less than 25 percent; four categories are high porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 25 percent.
5) The reservoirs were classified into 5 categories according to permeability: one type is an ultra-low permeability reservoir, namely: the oil reservoir permeability is less than 5 millidarcy; the second category is low permeability reservoirs, namely: the oil reservoir permeability is more than or equal to 5 millidarcy and less than 50 millidarcy; three categories are medium permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 50 millidarcies and less than 500 millidarcies; the four categories are high permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 500 millidarcies and less than 1000 millidarcies; the five types are ultra-high permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 1000 millidarcy;
6) Oil reservoirs are classified into 4 categories according to the viscosity of the formation crude oil: one is a low-viscosity oil reservoir, namely: the viscosity of the crude oil in the stratum is less than 5 mPas; the second category is medium viscous oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 5 mPas and less than 20 mPas; three categories are high viscosity oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 20 mPas and less than 50 mPas, four types are heavy oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 50 mPas. Wherein, the oil-bearing horizon of the X oil field of the research object is: n is a radical of1mLThe sedimentary phases are: shallow water delta, the reservoir type is: lithology-architecture, reservoir lithology is: sandstone, the driving type is: artificial water injection, the oil reservoir burial depth is: 1672 to 1753 meters of oil deposit, belonging to middle and shallow oil deposit; the effective thickness is: 7.8 m, belonging to medium-thick oil reservoir; the porosity was: 31.5 percent, belonging to a high-porosity oil reservoir; the permeability is: 1471.4 millidarcy, belonging to ultra-high permeability oil reservoir; the formation crude oil viscosity is: 158 mpa sec, which belongs to heavy oil reservoir.
And according to the geological oil reservoir characteristics of the research object, determining the analog oil fields as follows: A. b, C and D, wherein,
the oil-bearing layer of the A oil field is as follows: n is a radical of hydrogen1mLThe deposition phase is as follows: shallow water delta, the reservoir types are: lithology-architecture, reservoir lithology is: sandstone, the drive type is: artificial water injection, the oil reservoir burial depth is: 1203-1263 m, belonging to middle and shallow oil reservoirs; the effective thickness of the A oil field is as follows: 6 meters, 8 meters, 10 meters, 11 meters, 11.9 meters, 12 meters, 14 meters, 16 meters, 18 meters, 20 meters, and the embodiment is preferably as follows: 11.9 meters, namely: the thickness of the oil deposit is more than or equal to 5 meters and less than 20 meters; belongs to a class II thick-layer oil reservoir; the porosity was: 27.7%, belonging to high porosity oil reservoir; the permeability of the A oil field is: 1000, 1200, 1400, 1600, 1691, 1692.0, 1800, 2000 millidarcy; the preferred embodiment of the present invention:1691.0 millidarcy, the permeability of the A oil reservoir is more than or equal to 1000 millidarcy, and the A oil reservoir belongs to five types of ultra-high permeability oil reservoirs; the formation crude oil viscosity is: 135 mpa second, belonging to heavy oil reservoir.
The oil-bearing layer of the oil field B is as follows: n is a radical of1mLThe deposition phase is as follows: shallow water delta, the reservoir type is: lithology-architecture, reservoir lithology is: sandstone, the driving type is: artificial water injection, the oil reservoir burial depth is as follows: 1420-1485 meters of the oil pool, belonging to middle and shallow oil pool; the effective thickness of the oil field B is as follows: 6 meters, 8 meters, 9.7 meters, 10 meters, 11 meters, 12 meters, 14 meters, 16 meters, 18 meters, and the embodiment is preferably as follows: 9.7 meters, namely: the thickness of the oil deposit is more than or equal to 5 meters and less than 20 meters; belongs to a class II thick-layer oil reservoir; the porosity is 28.8 percent, and the oil reservoir belongs to a high-porosity oil reservoir; permeability of 1332.0 millidarcy; belonging to ultra-high permeability oil reservoir; the viscosity of the crude oil taken from the stratum of the oil field B is as follows: 50 mpa sec, 80 mpa sec, 100 mpa sec, 120 mpa sec, 140 mpa sec, 159 mpa sec, 180 mpa sec, 200 mpa sec, 220 mpa sec; the present embodiment is preferably: 159 mpa sec, i.e.: the viscosity of the crude oil in the stratum is more than or equal to 50 mPas, and the oil belongs to four types of heavy oil reservoirs. Oil-containing horizon of C field is N1mLThe deposition phase is as follows: shallow water delta, the reservoir type is: lithology-architecture, reservoir lithology is: sandstone, the drive type is: artificial water injection, the oil reservoir burial depth is: 1455-1483 m, which belongs to middle and shallow oil reservoirs; the effective thickness of the C oil field is as follows: 6 meters, 8 meters, 8.1 meters, 8.2 meters, 10 meters, 11 meters, 12 meters, 14 meters, 16 meters, 18 meters, and the embodiment is preferably: 8.1 meters, namely: the thickness of the oil deposit is more than or equal to 5 meters and less than 20 meters, and the oil deposit belongs to two types of thick-layer oil deposits; the porosity was: 29.9%, belonging to high porosity oil reservoir; the permeability is: 1502.9 millidarcy, belonging to ultra-high permeability reservoirs; the formation crude oil viscosity is: 120 millipascal seconds, belonging to heavy oil reservoir.
D, the oil-bearing layer of the oil field is as follows: n is a radical of hydrogen1mLThe deposition phase is as follows: shallow water delta, the reservoir type is: lithology-architecture, reservoir lithology is: sandstone, the driving type is: artificial water injection, the oil reservoir burial depth is: 1150-1263 m, belonging to middle and shallow oil reservoirs; the effective thickness of the oil field is as follows: 6 meters, 8 meters, 10 meters, 11 meters, 11.2 meters, 11.6 meters,12 meters, 14 meters, 16 meters, 18 meters, the preferred embodiment is: 11.2 meters, namely: the thickness of the oil deposit is more than or equal to 5 meters and less than 20 meters; belongs to a class II thick-layer oil reservoir; the porosity was: 29.3 percent, belonging to high-porosity oil reservoirs; the permeability is as follows: 1104.0 millidarcy, belonging to ultra-high permeability reservoirs; the formation crude oil viscosity is: 130m mPas, belonging to heavy oil reservoir.
The geologic reservoir characterization parameters for subject X field and analogous fields a, B, C, and D are shown in table 1 below:
table 1: research object and analogous oil field geological reservoir characteristic parameters
Figure BDA0003741080620000071
Figure BDA0003741080620000081
In the first step, according to the geological oil reservoir characteristics of the research object, the analog oil field is determined: selecting a producing oilfield with the same or similar horizon, sedimentary facies, reservoir type, reservoir lithology, driving type, reservoir burial depth, effective thickness, porosity, permeability and formation crude oil viscosity as an analogy object.
The second step is that: performing history fitting on the analog oilfield production dynamic data;
carrying out history fitting process on the analog oilfield production dynamic data: due to the limitation of understanding on the geological condition of the oil reservoir during modeling, the numerical simulation model cannot truly reflect the actual condition of the oil reservoir, and the calculated values and the actual values of main dynamic indexes of pressure, oil production and water content in oil field development are matched by repeatedly and continuously adjusting the static parameters of the oil reservoir.
Firstly, inputting historical data of the oil fields A, B, C and D into corresponding geological models, and carrying out historical fitting work, wherein the historical fitting work comprises the following steps: the oil field fitting and single well fitting are carried out in two ways, taking the oil field A as an example, and the history fitting process and the history fitting result are described as follows:
1. inputting the oil field development parameters A into the numerical simulation model, fitting by continuously optimizing and changing the static parameters and the skin of the numerical simulation model, wherein the fitting result is shown in figures 2 to 7, and the horizontal permeability of the numerical simulation model of the oil field A after fitting is as follows: 1691.0 millidarcy, horizontal to vertical permeability ratio: 10. the epidermis is: 5. the formation crude oil viscosity is: 135m mPa s;
2. after fitting is obtained by using a similar method, the horizontal permeability of the oil field B numerical simulation model is as follows: 1332.0 millidarcy, horizontal to vertical permeability ratios: 20. the epidermis is: 6. the formation crude oil viscosity is: 159 mPa s;
the horizontal permeability of the numerical simulation model of the C oil field is as follows: 1502.9 millidarcy, horizontal to vertical permeability ratio: 22. the epidermis is: 7. the formation crude oil viscosity is: 120 millipascal seconds;
the horizontal permeability of the numerical simulation model of the oil field is as follows: 1104.0 millidarcy, horizontal to vertical permeability ratio: 18. the epidermis is: 5. the formation crude oil viscosity is: 130 mpa sec.
The third step: predicting the recovery ratio of the analogous oil field and developing the sensitivity analysis of the recovery ratio;
and developing recovery ratio prediction on the basis of the second step of history fitting, and developing sensitivity analysis on the oil fields A, B, C and D by adjusting model parameters, wherein the recovery ratio prediction result and the sensitivity analysis result of the basic model of the oil field A are shown in the following table 2:
table 2: a oil field basic model recovery ratio prediction result and sensitivity analysis result
Figure BDA0003741080620000091
B field base model recovery prediction and sensitivity analysis as shown in table 3 below:
table 3: b oil field basic model recovery ratio prediction result and sensitivity analysis result
Figure BDA0003741080620000101
C field basemodel recovery prediction and sensitivity analysis, as shown in table 4 below:
table 4: c oil field basic model recovery ratio prediction result and sensitivity analysis result
Figure BDA0003741080620000111
D field base model recovery prediction and sensitivity analysis, as shown in table 5 below:
table 5: d oil field basic model recovery ratio prediction result and sensitivity analysis result
Figure BDA0003741080620000121
In the third step, the recovery ratio of the analog oil field is predicted, and the recovery ratio sensitivity analysis is carried out: on the basis of historical fitting, numerical simulation research is carried out, the recovery ratio of the existing model is predicted, sensitivity analysis of parameters such as horizontal permeability, horizontal and vertical permeability ratio, the skin, the viscosity of crude oil in stratum, well pattern density, production pressure difference, water injection time, extraction liquid multiple and the like on the recovery ratio is carried out, and the water flooding recovery ratio under different conditions is predicted.
The fourth step: constructing a machine learning model based on the recovery ratio prediction result;
inputting the oil field basic model recovery prediction results and sensitivity analysis results in the third step A, B, C and D into a machine learning model, wherein the machine learning model selected in the embodiment is as follows: 33 sample points are related to each oil field of the BP neural network, 132 sample points are summed, 112 sample points are selected as a training set, the rest 20 sample points are selected as a test set, and the number of model output nodes is as follows: 1, representing recovery ratio, and hidden layer nodes are as follows: 5, the input layer and hidden layer transfer functions are: logsig, the transfer function of the output layer is: purelin, the training function is: trangd, the learning function is: learnd, learning rate is: 0.05, the maximum number of iterations is: 20000, the accuracy of the interpolation model is measured by using the root mean square error, the average relative error and the average absolute error, the measurement accuracy requires that the root mean square error, the average relative error and the average absolute error are all less than 10%, and from the prediction result of the machine learning model, the root mean square error, the average relative error and the average absolute error are respectively as follows: 7.25%, 6.28% and 2.86%, and the prediction accuracy of the test set reaches the expected requirement.
In the fourth step, a machine learning model is constructed based on the recovery factor prediction result: forming a sample set by the recovery ratio prediction results and the sensitivity analysis results of all the analog oil fields, randomly selecting part of samples as a training set, using the rest samples as a test set, selecting the parameters of the sensitivity analysis in the third step as the characteristic values of the recovery ratio prediction, training the training set by using a machine learning model, carrying out the recovery ratio prediction on the test set by using the trained machine learning model, calculating the prediction precision of the test set, stopping training if the test precision meets the expected requirement, and adjusting the relevant parameters of the machine learning model if the test precision does not meet the expected requirement until the prediction precision of the test set meets the expected requirement.
The fifth step: predicting the recovery ratio of the research object by using the trained machine learning model;
inputting the recovery factor prediction parameters of the X oil field of the research object into the machine learning model constructed in the fourth step for prediction, wherein the horizontal permeability set by the basic parameters of the X oil field is as follows: 1471.4 millidarcy, horizontal to vertical permeability ratio: 15. the epidermis is: 5. the formation crude oil viscosity is: 158 mpa sec, well pattern density: 3.1 ports per square kilometer, and the production pressure difference is as follows: 1.2 MPa, synchronous water injection and 4.5 times of extraction time. Finally, the recovery of the subject X field was 28.2%.
And in the fifth step, predicting the recovery ratio of the research object by using the trained machine learning model: and inputting the characteristic parameters of the research object into the machine learning model trained in the fourth step, wherein the characteristic parameters are the parameters of the sensitivity analysis in the third step, and obtaining a recovery factor prediction result.
The above-mentioned unexplained technologies are prior art and will not be described in detail.
It is to be understood that the present invention has been described with reference to certain embodiments, and that various changes in the features and embodiments, or equivalent substitutions may be made therein by those skilled in the art without departing from the spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.

Claims (9)

1. The method for predicting the recovery ratio of water flooding in offshore oil field is characterized by comprising the following steps: the method comprises the following steps:
the first step is as follows: determining an analog oil field according to the geological oil reservoir characteristics of a research object;
the second step: performing history fitting on the analog oilfield production dynamic data;
the third step: predicting the recovery ratio of the analogy oil field, and carrying out the sensitivity analysis of the recovery ratio;
the fourth step: constructing a machine learning model based on the recovery ratio prediction result;
the fifth step: and predicting the recovery ratio of the research object by using the trained machine learning model.
2. The offshore oilfield water flooding recovery prediction method of claim 1, wherein in the first step, the research objects are: combining drilling, logging, well logging and seismic data to determine geological reservoir characteristics of a study object, the main parameters for characterizing the geological reservoir characteristics comprise: the method comprises the following steps of measuring an oil-bearing horizon, a sedimentary facies, a reservoir type, reservoir lithology, a driving type, reservoir burial depth, effective thickness, porosity, permeability and formation crude oil viscosity, wherein the characteristic parameters of the oil-bearing horizon, the sedimentary facies, the reservoir type, the reservoir lithology and the driving type are measured by adopting qualitative indexes, and the other 5 characteristic parameters of the reservoir burial depth, the effective thickness, the porosity, the permeability and the formation crude oil viscosity are measured by adopting quantitative indexes.
3. The offshore oilfield water flooding recovery prediction method of claim 1 or 2, wherein in the first step, the analogy oilfield is selected according to the following steps:
1. firstly, dividing the reservoir types of a research object according to national standard, row standard or enterprise standard reservoir classification related standards; when selecting the analog oil field, requiring that the qualitative indexes of the characteristic parameters of the research object and the analog oil field are the same, and requiring that the oil reservoir types divided by the quantitative indexes of the characteristic parameters of the research object and the analog oil field are the same;
2. according to the oil reservoir burial depth, the oil reservoir is divided into 5 types: one type is shallow reservoirs, namely: the oil deposit burial depth is less than 500 meters; the second category is medium-shallow reservoirs, namely: the oil deposit burial depth is more than or equal to 500 meters and less than 2000 meters; the three types are medium-deep oil reservoirs, namely: the oil deposit burial depth is more than or equal to 2000 meters and less than 3500 meters; the four types are deep reservoirs, namely: the oil deposit depth is greater than or equal to 3500 meters and is less than 4500 meters, and five types of oil deposits are ultra-deep oil deposits, namely: the oil deposit burial depth is more than or equal to 4500 meters;
3. the oil reservoirs are divided into 4 types according to the thickness of the oil reservoirs: one type is thin layer reservoirs, namely: the thickness of the oil deposit is less than 5 meters; the second category is medium-heavy reservoir, namely: the thickness of the oil deposit is more than or equal to 5 meters and less than 20 meters; three types are thick-layer reservoirs, namely: the thickness of the oil deposit is more than or equal to 20 meters and less than 40 meters; the four types are extra-thick reservoir, namely: the thickness of the oil reservoir is more than or equal to 40 meters;
4. the reservoirs are classified according to porosity into 4 classes: one type is an ultra-low porosity reservoir, namely: the porosity of the oil reservoir is less than 10 percent; the second category is low porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 10 percent and less than 15 percent; three categories are medium porosity reservoirs, namely: the porosity of the oil reservoir is more than or equal to 15 percent and less than 25 percent; four categories are high porosity reservoirs, namely: reservoir porosity greater than or equal to 25%;
5. the reservoirs were classified into 5 categories according to permeability: one type is an ultra-low permeability reservoir, namely: the oil reservoir permeability is less than 5 millidarcy; the second category is low permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 5 millidarcies and less than 50 millidarcies; three categories are medium permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 50 millidarcies and less than 500 millidarcies; the four categories are high permeability reservoirs, i.e.: the oil reservoir permeability is greater than or equal to 500 millidarcies and less than 1000 millidarcies; the five types are ultra-high permeability reservoirs, namely: the oil reservoir permeability is greater than or equal to 1000 millidarcy;
6. oil reservoirs are classified into 4 categories according to the viscosity of the crude oil in the stratum: one is a low-viscosity oil reservoir, namely: the viscosity of the crude oil in the stratum is less than 5 mPas; the second category is medium viscous oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 5 mPas and less than 20 mPas; three categories are high viscosity oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 20 mPas and less than 50 mPas, four types are heavy oil reservoirs, namely: the viscosity of the crude oil in the stratum is more than or equal to 50 mPas.
4. The offshore oilfield water drive recovery prediction method of claim 1, wherein in the second step, the analogous oilfield production dynamics data is subjected to a history fitting process: due to the limitation of understanding on the geological condition of the oil reservoir during modeling, the numerical simulation model cannot truly reflect the actual condition of the oil reservoir, and the calculated values and the actual values of main dynamic indexes of pressure, oil production and water content in oil field development are matched by repeatedly and continuously adjusting the static parameters of the oil reservoir.
5. The offshore oilfield water flooding recovery prediction method of claim 1 or 4, wherein in the second step, historical data of a plurality of oil fields are input into corresponding geological models, and history fitting work is performed, wherein the history fitting work comprises: oil field fitting and single well fitting.
6. The offshore oilfield water flooding recovery prediction method of claim 1, wherein in the third step, a recovery prediction is developed based on the second step history fit and a sensitivity analysis is developed for several fields by adjusting model parameters.
7. The offshore oilfield water flooding recovery prediction method of claim 1 or 6, wherein in the third step, the recovery of the simulated oilfield is predicted and a recovery sensitivity analysis is performed: on the basis of historical fitting, carrying out numerical simulation research, predicting the recovery ratio of the existing model, carrying out sensitivity analysis on parameters such as horizontal permeability, horizontal and vertical permeability ratio, skin, stratum crude oil viscosity, well pattern density, production pressure difference, water injection opportunity, extraction liquid multiple and the like on the recovery ratio, and predicting the water flooding recovery ratio under different conditions.
8. The offshore oilfield water drive recovery prediction method of claim 1, wherein in the fourth step, the multiple oilfield base model recovery prediction results and the sensitivity analysis results of the third step are input into a machine learning model, and the machine learning model is selected as follows: a BP neural network; constructing a machine learning model based on the recovery prediction result: and (3) forming a sample set by recovery ratio prediction results of all the analog oil fields and recovery ratios predicted by sensitivity analysis, randomly selecting a part of samples as a training set, using the rest of samples as a test set, selecting parameters of the sensitivity analysis in the third step as characteristic values of recovery ratio prediction, training the training set by using a machine learning model, carrying out recovery ratio prediction on the test set by using the trained machine learning model, calculating the prediction precision of the test set, stopping training if the test precision meets an expected requirement, and adjusting related parameters of the machine learning model until the prediction precision of the test set meets the expected requirement if the test precision does not meet the expected requirement.
9. The offshore oilfield water flooding recovery prediction method of claim 1, wherein in the fifth step, the well-trained machine learning model is used to predict the recovery ratio of the study: and inputting the characteristic parameters of the research object into the machine learning model trained in the fourth step, wherein the characteristic parameters are the parameters of the sensitivity analysis in the third step, and obtaining a recovery factor prediction result.
CN202210812608.9A 2022-07-12 2022-07-12 Offshore oilfield water flooding recovery rate prediction method Pending CN115271182A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210812608.9A CN115271182A (en) 2022-07-12 2022-07-12 Offshore oilfield water flooding recovery rate prediction method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210812608.9A CN115271182A (en) 2022-07-12 2022-07-12 Offshore oilfield water flooding recovery rate prediction method

Publications (1)

Publication Number Publication Date
CN115271182A true CN115271182A (en) 2022-11-01

Family

ID=83765153

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210812608.9A Pending CN115271182A (en) 2022-07-12 2022-07-12 Offshore oilfield water flooding recovery rate prediction method

Country Status (1)

Country Link
CN (1) CN115271182A (en)

Similar Documents

Publication Publication Date Title
CN109447532B (en) Oil reservoir inter-well connectivity determination method based on data driving
US8359184B2 (en) Method, program and computer system for scaling hydrocarbon reservoir model data
CN105631529B (en) Method for predicting water breakthrough time of boundary water gas reservoir
CN102041995A (en) System for monitoring complicated oil deposit flooding conditions
CN104899411B (en) A kind of reservoir productivity prediction model method for building up and system
CN110954944A (en) Fault trap oil-containing height earthquake prediction method
Xuequn et al. Multi-parameter quantitative assessment of 3D geological models for complex fault-block oil reservoirs
CN115330060A (en) Shale oil horizontal well productivity calculation method based on reservoir and engineering factor analysis
Ali et al. A semi-analytical method for history matching and improving geological models of layered reservoirs: CGM analytical method
CN113361111B (en) Partitioning method of low-permeability reservoir characteristic model
CN115271182A (en) Offshore oilfield water flooding recovery rate prediction method
CN110851982B (en) Method for analyzing feasibility of oxygen reduction air flooding of medium-low permeability oil reservoir
CN116011268A (en) Quantitative description method of dominant seepage channel
Gharbi Use of reservoir simulation for optimizing recovery performance
RU2701761C1 (en) Control method of oil production at mature separate oil deposit
CN112502677A (en) Water injection development effect evaluation method based on multiple linear regression
CN111428375A (en) Novel high-water-content oilfield layer system recombination and division method
Wang Waterflooding Sandstone Reservoirs: Methods, Design and Analysis
Aslam et al. Closing the Loop on a History Match for a Permian EOR Field Using Relative Permeability Data Uncertainty
Li et al. Application of Petrel software in reserve estimation in Xinli-Xinbei oilfield, Jilin Province, China
Zandkarimi et al. Static model construction for lower Shuaiba Formation (Lower Dariyan)
Yin et al. Estimation of reservoir and remaining oil prediction based on flow unit analysis
Hossain et al. Application of Petrel Software in Reserve Estimation of Titas Gas Field (B & C Sand)
Khamees EVALUATING UNCERTAINTY IN THE VOLUME OF OIL IN PLACE IN MISHRIF RESERVOIR
Bond History Matching

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