CN116911216B - Reservoir oil well productivity factor assessment and prediction method - Google Patents
Reservoir oil well productivity factor assessment and prediction method Download PDFInfo
- Publication number
- CN116911216B CN116911216B CN202311167331.XA CN202311167331A CN116911216B CN 116911216 B CN116911216 B CN 116911216B CN 202311167331 A CN202311167331 A CN 202311167331A CN 116911216 B CN116911216 B CN 116911216B
- Authority
- CN
- China
- Prior art keywords
- oil
- reservoir
- oil reservoir
- data
- analysis
- 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
- 239000003129 oil well Substances 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 87
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 49
- 239000011148 porous material Substances 0.000 claims abstract description 28
- 238000009826 distribution Methods 0.000 claims abstract description 23
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 20
- 238000012544 monitoring process Methods 0.000 claims abstract description 18
- 230000002068 genetic effect Effects 0.000 claims abstract description 16
- 230000007246 mechanism Effects 0.000 claims abstract description 15
- 238000000556 factor analysis Methods 0.000 claims abstract description 14
- 239000012530 fluid Substances 0.000 claims description 45
- 230000006870 function Effects 0.000 claims description 42
- 238000005516 engineering process Methods 0.000 claims description 35
- 238000004458 analytical method Methods 0.000 claims description 34
- 238000005457 optimization Methods 0.000 claims description 33
- 230000003993 interaction Effects 0.000 claims description 32
- 238000011161 development Methods 0.000 claims description 28
- 238000003012 network analysis Methods 0.000 claims description 23
- 239000000126 substance Substances 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000013210 evaluation model Methods 0.000 claims description 10
- 238000013473 artificial intelligence Methods 0.000 claims description 9
- 230000035699 permeability Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000000611 regression analysis Methods 0.000 claims description 9
- 238000010206 sensitivity analysis Methods 0.000 claims description 9
- 238000004088 simulation Methods 0.000 claims description 9
- 238000004519 manufacturing process Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 239000011435 rock Substances 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 238000011002 quantification Methods 0.000 claims description 5
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000013178 mathematical model Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000002591 computed tomography Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000013209 evaluation strategy Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 230000008303 genetic mechanism Effects 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 claims description 3
- 238000013486 operation strategy Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000002921 genetic algorithm search Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 description 4
- 238000012417 linear regression Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009396 hybridization Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012067 mathematical method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- 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
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V11/00—Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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"
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Mining & Mineral Resources (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Fluid Mechanics (AREA)
- Molecular Biology (AREA)
- Geology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Mathematical Optimization (AREA)
- Computer Graphics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Geochemistry & Mineralogy (AREA)
- Physiology (AREA)
- Algebra (AREA)
- Development Economics (AREA)
Abstract
The invention relates to the technical field of oil well productivity assessment, in particular to a reservoir oil well productivity factor assessment and prediction method, which comprises the following steps: acquiring microscopic pore and crack structure information of an oil well by utilizing multi-source geological and engineering data, and further analyzing the spatial distribution characteristics of an oil reservoir by combining an earthquake method and an electric method; introducing a multiphase flow mechanism and reservoir engineering dynamic monitoring, analyzing the motion characteristics of oil and water gas under microscopic and macroscopic dimensions, and combining with the underground pressure and temperature distribution of a reservoir to form an omnibearing main control factor analysis framework; according to the invention, complex and nonlinear relations among oil reservoir main control factors are captured, so that more accurate oil well productivity prediction is realized, and parameters of the RBF network can be adaptively adjusted by utilizing global searching capability of a genetic algorithm.
Description
Technical Field
The invention relates to the technical field of oil well productivity assessment, in particular to a reservoir oil well productivity factor assessment and prediction method.
Background
The oil well productivity prediction is a key link of oil field development and management, and directly affects the economic benefit and the resource utilization rate of the oil field. However, the complexity, nonlinearity, and variability of reservoir systems make well productivity prediction a challenge.
Conventional well productivity prediction methods rely primarily on empirical formulas and linear regression analysis, which, while performing well in some situations, do not capture well the complex, nonlinear relationships between reservoir master factors.
Advanced mathematical and computational methods are required to achieve more accurate well productivity predictions. Fuzzy logic can handle fuzzy, uncertain information of reservoirs, but building an effective fuzzy model has a certain complexity. Nonlinear regression analysis, particularly Radial Basis Function (RBF) networks, can capture complex relationships, but selecting appropriate basis functions and parameter adjustments is a challenge. Furthermore, optimization algorithms such as genetic algorithms are very effective in parameter optimization, but their execution efficiency and stability need to be further improved.
The existing oil well productivity prediction method often lacks depth integration of oil reservoir geological characteristics and actual operation data, and also lacks an effective self-adaptive adjustment mechanism. Limitations of the prior art lead to difficulties in prediction accuracy and stability to meet practical needs.
In response to the challenges and shortcomings described above, a comprehensive, dynamic, real-time well productivity prediction method is needed. The method should combine fuzzy logic, nonlinear regression analysis and adaptive optimization algorithm, and can adaptively adjust parameters and optimize prediction accuracy.
By analyzing the primary control factors of the oil reservoir, such as geological characteristics, lithology, fluid properties and the like, and combining the actual operation data of the oil well, the invention aims to construct a more advanced, flexible and accurate oil well productivity prediction model. The model realizes comprehensive capture of complexity and variability of the oil reservoir by combining the fuzzy logic, the radial basis function network and the genetic algorithm, and provides self-adaptive adjustment of prediction parameters and optimization of prediction accuracy.
Disclosure of Invention
Based on the above purpose, the invention provides a reservoir oil well productivity factor evaluation and prediction method.
A method for evaluating and predicting the productivity factors of a reservoir oil well comprises the following steps:
step one: acquiring microscopic pore and crack structure information of an oil well by utilizing multi-source geological and engineering data, and further analyzing the spatial distribution characteristics of an oil reservoir by combining an earthquake method and an electric method;
step two: introducing a multiphase flow mechanism and reservoir engineering dynamic monitoring, analyzing the motion characteristics of oil and water gas under microscopic and macroscopic dimensions, and combining with the underground pressure and temperature distribution of a reservoir to form an omnibearing main control factor analysis framework;
step three: through deep learning and complex network analysis technology, the interaction of an underground physical field, a chemical field and a mechanical field is synthesized, a multi-scale multi-factor main control factor evaluation model is constructed, and accurate quantification and sensitivity analysis of oil well productivity are realized;
step four: according to main control factor analysis, a dynamic and real-time oil well productivity prediction model is constructed by combining actual operation data and geological characteristics of an oil well by using fuzzy logic, nonlinear regression analysis and a self-adaptive optimization algorithm, parameters are adaptively adjusted, and prediction accuracy is optimized;
step five: the system is linked with a field operation system, the automatic control and intelligent optimization of the oil well productivity are realized through the Internet of things and artificial intelligence technology, and the optimal operation of the oil well is realized through continuous learning and optimizing of an operation strategy by an intelligent algorithm;
step six: through the visual interface, the virtual reality technology is combined, the operation state of the oil well and the influence of the main control factors are displayed in real time, and oil field management staff can intuitively understand the operation state of the oil well.
Further, the first step specifically includes:
carrying out fine analysis on a rock sample of the oil deposit by adopting a high-resolution X-ray CT scanning and nuclear magnetic resonance imaging technology and combining ground and underground data to reveal microstructure characteristics of pores, cracks and oil-water contact of the oil deposit;
using an acoustic logging tool, performing nondestructive evaluation on permeability, elastic modulus and pore structure of an oil reservoir by analyzing propagation characteristics of acoustic waves in rock, and providing correlation between microstructure and macroscopic physical characteristics of the oil reservoir;
combining the seismic exploration and the ground-electric method, adopting three-dimensional seismic inversion and resistivity imaging technology to construct a three-dimensional spatial model of the oil reservoir, analyzing the spatial distribution characteristics of the oil reservoir, revealing the conditions of broken blocks, folding and broken geological structures of the oil reservoir, and providing spatial references for the development of the oil reservoir;
and integrating and excavating geological and engineering data through an artificial intelligence algorithm, identifying key control factors of an oil reservoir, constructing a comprehensive description model of the oil reservoir, and providing accurate input for subsequent main control factor analysis and yield prediction.
Further, the second step specifically includes:
analyzing the flow characteristics of oil and water vapor in pores and cracks by a multiphase fluid simulation technology based on a finite volume method, revealing the flow mechanism of fluid under different permeabilities and pore structures and the influence of interaction and phase change processes, and specifically comprising the following steps: collecting basic geology and engineering data of related oil reservoirs; performing three-dimensional oil reservoir meshing by using a finite volume method to capture the pore and fracture geometry of the oil reservoir; defining viscosity, density and pressure fluid property parameters of oil, water and gas according to the acquired data; constructing a mathematical model for describing the flow of oil, water and gas in pores and cracks based on Darcy and non-Darcy flow equations; adding a submodel for describing interactions between fluids and phase changes between oil, water, and gas; solving the flow equation by using a finite volume method to obtain the speed, pressure distribution and phase distribution of the fluid in the pores and the cracks; analyzing the flow mechanism of the fluid under different permeabilities and pore structures through solving the results, and how the interaction and phase change process affect the oil well productivity; based on flow mechanism analysis, combining actual operation data of the oil well, and constructing a dynamic and real-time oil well productivity prediction model;
the method utilizes an oil reservoir simulator and fluid dynamics software to analyze the large-scale flow characteristics of the oil reservoir by combining the temperature, pressure and lithology parameters of the oil reservoir, and simulates the pressure distribution, temperature change and fluid saturation in the oil field development process, and specifically comprises the following steps: collecting and setting basic parameters of temperature, pressure and lithology of the related oil reservoirs; selecting an oil reservoir simulator and fluid dynamics software according to the required analysis precision and range, and initializing an oil reservoir and fluid model according to the collected parameters in the selected simulator and software; running simulation, namely simulating pressure distribution, temperature change and fluid saturation in the oilfield development process; analyzing the simulation result, and providing corresponding optimization strategies or adjusting the existing development plans aiming at different development stages;
combining dynamic monitoring data of the oil reservoir, wherein the dynamic monitoring data comprise wellhead pressure, temperature and yield, real-time monitoring information, and dynamic correction of an oil reservoir model is realized through data assimilation and real-time updating technologies;
the method comprises the steps of utilizing a field monitoring and sensing technology and combining internet of things equipment to monitor and analyze the underground pressure and the underground temperature of an oil reservoir in real time;
by analyzing the interaction of the fluid at the microscopic level and the macroscopic level, the fluid conveying characteristics of the oil reservoir are revealed, and a basis is provided for the subsequent oil well productivity prediction and intelligent control.
Further, the third step specifically includes:
adopting a recurrent neural network algorithm (RNN) to process and analyze underground physical, chemical and mechanical data of an oil reservoir, extracting modes and relations in the data, and providing data support for the evaluation of main control factors;
constructing a multi-scale and multi-dimensional interaction model of an oil reservoir by using a complex network analysis technology based on social network analysis, analyzing complex interactions and dependency relations among elements in the oil reservoir, and revealing weights and influences among main control factors specifically comprises: collecting data of various elements in an oil reservoir, interactions and dependency relations of the elements, and preprocessing the data; selecting a social network analysis tool for complex network analysis based on the collected data, and constructing a multi-scale and multi-dimensional network model for describing complex interactions and dependency relations among elements in the oil reservoir; analyzing and quantifying the weight and influence of each node in the network by using a social network analysis algorithm; according to the analysis result, identifying factors which have dominant influence on the oil well productivity, and carrying out deep analysis on the factors; optimizing the development strategy of the oil well and performing more accurate productivity prediction by combining the identified main control factors and the corresponding weights and influences thereof;
the method combines the data of the underground physical field, the chemical field and the mechanical field to construct a comprehensive oil reservoir characteristic analysis framework, accurately reflects the complexity and the variability of the oil reservoir, and specifically comprises the following steps: collecting data of a physical field, a chemical field and a mechanical field related to an oil reservoir; preprocessing and standardizing data of physical, chemical and mechanical fields by utilizing a data fusion technology; constructing a comprehensive oil reservoir characteristic analysis framework according to the collected and preprocessed data; evaluating the variability and complexity of the reservoir within the constructed framework using a multi-factor analysis method; the weight of each influence factor in a physical field, a chemical field and a mechanical field is determined through a sensitivity analysis and optimization algorithm so as to accurately reflect the influence of the influence factors on the oil well productivity;
analyzing the influence degree and uncertainty of each main control factor on the oil well production energy by a sensitivity analysis and uncertainty quantification method so as to identify key control factors and optimize an evaluation strategy;
by utilizing the existing geological, engineering and production data and combining artificial intelligence, an evaluation model of the main control factors of the oil well productivity is established and trained, and accurate evaluation and prediction are carried out on different development stages of the oil reservoir.
Further, the recurrent neural network algorithm is used for processing and analyzing time series data of the oil reservoir, and the specific form comprises the following steps:
the hidden layer is updated as follows:
output layer calculation:
wherein (1)>Is the hidden layer state at time t, +.>Is the input of the time t at which,、/>、/>is a weight matrix, < >>、/>Is a bias term.
Further, the social network analysis is used for analyzing complex interactions and dependency relationships among elements in the oil reservoir, and the analysis model comprises:
adjacency matrix representation:
node degree centrality calculation:
wherein (1)>Is node->Neighbor node set,/->Is the total number of nodes in the network。
Further, the step four specifically includes:
fuzzy logic analysis: processing fuzzy and uncertain information of oil reservoirs by using fuzzy logic, including geological characteristics, lithology and fluid properties, constructing a fuzzy rule and a fuzzy reasoning system, and converting qualitative description of main control factors into quantifiable model input;
nonlinear regression analysis: capturing complex and nonlinear relations among oil reservoir main control factors by adopting a Radial Basis Function (RBF) network;
and (3) constructing an adaptive optimization algorithm: parameter adjustment and optimization are carried out on the model by utilizing a genetic algorithm, and the prediction accuracy and stability of the model are ensured through continuous iteration and self-adaptive adjustment;
dynamic and real-time prediction model construction: combining fuzzy logic, nonlinear regression and an adaptive optimization algorithm to construct a dynamic and real-time oil well productivity prediction model;
actual operation data and geological characteristics are fused: the actual operation data of the oil well, including the pressure, the temperature and the flow of the oil well, are combined with geological characteristics such as lithology, porosity and saturation, so that the comprehensiveness and the accuracy of the prediction model are ensured.
Further, the Radial Basis Function (RBF) network is used to solve the nonlinear regression problem. The output of the RBF network is a function based on the euclidean distance of the input to the center of each basis function, and is specifically formed by:
RBF layer: the layer calculates the distance between the input vector and the center of the basis function, applies a radial basis function, and based on a Gaussian function, the calculation mode is as follows:
wherein (1)>Is an input vector, +.>Is->The center of the individual basis functions,is the standard deviation;
output layer: this layer is a linear combination of radial basis functions for predicting the output:
wherein (1)>Is a weight of->Is a bias item->Is the number of basis functions.
Further, the genetic algorithm searches for an optimal solution by simulating natural selection and genetic mechanism based on a heuristic optimization algorithm, and the basic steps include:
initializing: randomly generating a set of solutions (called a population);
selecting: evaluating the quality of each solution according to the fitness function, classifying the solutions into a first grade and a second grade according to the quality, and selecting the first grade to enter the next generation;
crossover (hybridization): generating child solutions by combining parent solutions;
variation: randomly changing the child solutions with a specific probability;
termination condition: stopping when a predetermined number of iterations is reached or a solution satisfying the condition is found;
parameters (e.g., basis function center, standard deviation, weight) of the RBF network are adaptively adjusted using genetic algorithms to find an optimal well productivity prediction model.
Further, the system also comprises a data preprocessing module for cleaning, sorting and converting oil well data to adapt to the input requirements of the evaluation model.
The invention has the beneficial effects that:
according to the invention, by combining a Radial Basis Function (RBF) network and a genetic algorithm, the complex and nonlinear relation between oil reservoir main control factors can be captured, so that more accurate oil well productivity prediction is realized, and parameters of the RBF network, such as a basis function center, standard deviation, weight and the like, can be adaptively adjusted by utilizing the global searching capability of the genetic algorithm, so that the prediction model is ensured to be consistent with actual operation data and geological characteristics of an oil well all the time.
The invention allows different mathematical and calculation methods to be flexibly selected and combined so as to adapt to specific requirements and conditions of different oil fields and oil reservoirs, and can continuously optimize the prediction accuracy so as to meet the actual requirements of oil field development and management.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only of the invention and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method according to an embodiment of the invention;
fig. 2 is a schematic diagram of a master factor evaluation model according to an embodiment of the invention.
Detailed Description
The present invention will be further described in detail with reference to specific embodiments in order to make the objects, technical solutions and advantages of the present invention more apparent.
It is to be noted that unless otherwise defined, technical or scientific terms used herein should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The terms "first," "second," and the like, as used herein, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
1-2, a method for estimating and predicting the productivity factor of a reservoir oil well comprises the following steps:
step one: acquiring microscopic pore and crack structure information of an oil well by utilizing multi-source geological and engineering data, and further analyzing the spatial distribution characteristics of an oil reservoir by combining an earthquake method and an electric method;
step two: introducing a multiphase flow mechanism and reservoir engineering dynamic monitoring, analyzing the motion characteristics of oil and water gas under microscopic and macroscopic dimensions, and combining with the underground pressure and temperature distribution of a reservoir to form an omnibearing main control factor analysis framework;
step three: through deep learning and complex network analysis technology, the interaction of an underground physical field, a chemical field and a mechanical field is synthesized, a multi-scale multi-factor main control factor evaluation model is constructed, and accurate quantification and sensitivity analysis of oil well productivity are realized;
step four: according to main control factor analysis, a dynamic and real-time oil well productivity prediction model is constructed by combining actual operation data and geological characteristics of an oil well by using fuzzy logic, nonlinear regression analysis and a self-adaptive optimization algorithm, parameters are adaptively adjusted, and prediction accuracy is optimized;
step five: the system is linked with a field operation system, the automatic control and intelligent optimization of the oil well productivity are realized through the Internet of things and artificial intelligence technology, and the optimal operation of the oil well is realized through continuous learning and optimizing of an operation strategy by an intelligent algorithm;
step six: through a visual interface and combining with a virtual reality technology, the operation state of the oil well and the influence of main control factors are displayed in real time, so that oil field management staff can intuitively understand the operation state of the oil well;
through the comprehensive implementation of the steps, the method not only can accurately evaluate the productivity of the oil well, but also realizes the omnibearing intelligent evaluation and control of the productivity of the oil well by introducing advanced data analysis technology, self-adaptive algorithm and intelligent control means, fills the blank of the prior art, and realizes technical effects which are difficult to realize in the development of the oil field.
The first step specifically comprises:
carrying out fine analysis on a rock sample of the oil deposit by adopting a high-resolution X-ray CT scanning and nuclear magnetic resonance imaging technology and combining ground and underground data to reveal microstructure characteristics of pores, cracks and oil-water contact of the oil deposit;
using an acoustic logging tool, performing nondestructive evaluation on permeability, elastic modulus and pore structure of an oil reservoir by analyzing propagation characteristics of acoustic waves in rock, and providing correlation between microstructure and macroscopic physical characteristics of the oil reservoir;
combining the seismic exploration and the ground-electric method, adopting three-dimensional seismic inversion and resistivity imaging technology to construct a three-dimensional spatial model of the oil reservoir, analyzing the spatial distribution characteristics of the oil reservoir, revealing the conditions of broken blocks, folding and broken geological structures of the oil reservoir, and providing spatial references for the development of the oil reservoir;
integrating and excavating geological and engineering data through an artificial intelligence algorithm, identifying key control factors of an oil reservoir, constructing a comprehensive description model of the oil reservoir, and providing accurate input for subsequent main control factor analysis and yield prediction;
by fusing multisource data and adopting advanced geological exploration and data analysis technology, comprehensive understanding and accurate description are provided for microscopic and macroscopic characteristics of oil reservoirs, and innovation and perfection of a method for evaluating main control factors and predicting productivity of oil wells in tight sandstone reservoirs are further promoted.
The second step specifically comprises:
analyzing the flow characteristics of oil and water vapor in pores and cracks by a multiphase fluid simulation technology based on a finite volume method, revealing the flow mechanism of fluid under different permeabilities and pore structures, and the influences of interaction and phase change processes specifically comprise: collecting basic geology and engineering data of related oil reservoirs; performing three-dimensional oil reservoir meshing by using a finite volume method to capture the pore and fracture geometry of the oil reservoir; defining viscosity, density and pressure fluid property parameters of oil, water and gas according to the acquired data; constructing a mathematical model for describing the flow of oil, water and gas in pores and cracks based on Darcy and non-Darcy flow equations; adding a submodel for describing interactions between fluids and phase changes between oil, water, and gas; solving the flow equation by using a finite volume method to obtain the speed, pressure distribution and phase distribution of the fluid in the pores and the cracks; analyzing the flow mechanism of the fluid under different permeabilities and pore structures through solving the results, and how the interaction and phase change process affect the oil well productivity; based on flow mechanism analysis, combining actual operation data of the oil well, and constructing a dynamic and real-time oil well productivity prediction model;
the method utilizes an oil reservoir simulator and fluid dynamics software to analyze the large-scale flow characteristics of the oil reservoir by combining the temperature, pressure and lithology parameters of the oil reservoir, and simulates the pressure distribution, temperature change and fluid saturation in the oil field development process, and specifically comprises the following steps: collecting and setting basic parameters of temperature, pressure and lithology of the related oil reservoirs; selecting an oil reservoir simulator and fluid dynamics software according to the required analysis precision and range, and initializing an oil reservoir and fluid model according to the collected parameters in the selected simulator and software; running simulation, namely simulating pressure distribution, temperature change and fluid saturation in the oilfield development process; analyzing the simulation result, and providing corresponding optimization strategies or adjusting the existing development plans aiming at different development stages;
combining dynamic monitoring data of an oil reservoir, wherein the dynamic monitoring data comprise wellhead pressure, temperature, yield and real-time monitoring information, and dynamic correction of an oil reservoir model is realized through data assimilation and real-time updating technologies, so that the accuracy and reliability of prediction are improved;
the on-site monitoring and sensing technology is used, and the real-time monitoring and analysis of the underground pressure and the underground temperature of the oil reservoir are combined with the Internet of things equipment, so that the safety and the high efficiency of the oil reservoir development process are ensured;
the interaction of the fluid on the microscopic level and the macroscopic level is analyzed, so that the fluid conveying characteristic of the oil reservoir is revealed, and a basis is provided for the subsequent oil well productivity prediction and intelligent control;
the method comprehensively utilizes multiphase flow mechanism, dynamic monitoring of oil reservoir engineering and advanced numerical simulation technology, realizes accurate analysis of oil reservoir fluid under different scales, provides a comprehensive main control factor analysis framework, and further promotes innovation and accuracy of a main control factor evaluation and productivity prediction method of an oil well of a tight sandstone reservoir.
The third step specifically comprises:
adopting a recurrent neural network algorithm (RNN) to process and analyze underground physical, chemical and mechanical data of an oil reservoir, extracting modes and relations in the data, and providing data support for the evaluation of main control factors;
constructing a multi-scale and multi-dimensional interaction model of an oil reservoir by using a complex network analysis technology based on social network analysis, analyzing complex interactions and dependency relations among elements in the oil reservoir, and revealing weights and influences among main control factors, wherein the method specifically comprises the following steps of: collecting data of various elements in an oil reservoir, interactions and dependency relations of the elements, and preprocessing the data; selecting a social network analysis tool for complex network analysis based on the collected data, and constructing a multi-scale and multi-dimensional network model for describing complex interactions and dependency relations among elements in the oil reservoir; analyzing and quantifying the weight and influence of each node in the network by using a social network analysis algorithm; according to the analysis result, identifying factors which have dominant influence on the oil well productivity, and carrying out deep analysis on the factors; optimizing the development strategy of the oil well and performing more accurate productivity prediction by combining the identified main control factors and the corresponding weights and influences thereof;
by combining data of underground physical fields (such as pressure and temperature), chemical fields (such as chemical reactions of oil, water and gas components) and mechanical fields (such as stress and deformation of rock), a comprehensive oil reservoir characteristic analysis framework is constructed, and complexity and variability of an oil reservoir are accurately reflected, and the method specifically comprises the following steps: collecting data of a physical field, a chemical field and a mechanical field related to an oil reservoir; preprocessing and standardizing data of physical, chemical and mechanical fields by utilizing a data fusion technology; constructing a comprehensive oil reservoir characteristic analysis framework according to the collected and preprocessed data; evaluating the variability and complexity of the reservoir within the constructed framework using a multi-factor analysis method; the weight of each influence factor in a physical field, a chemical field and a mechanical field is determined through a sensitivity analysis and optimization algorithm so as to accurately reflect the influence of the influence factors on the oil well productivity;
analyzing the influence degree and uncertainty of each main control factor on the oil well production energy by a sensitivity analysis and uncertainty quantification method so as to identify key control factors and optimize an evaluation strategy;
establishing and training an evaluation model of oil well productivity main control factors by utilizing the existing geological, engineering and production data and combining artificial intelligence, and accurately evaluating and predicting different development stages of an oil reservoir;
through deep learning and the introduction of complex network analysis technology, the multi-scale and multi-factor main control factor evaluation of the oil reservoir is realized, the limitation of the traditional method in the aspect of processing the complexity of the oil reservoir is broken through, the innovation and deepening of the main control factor evaluation and productivity prediction method of the tight sandstone reservoir oil well are promoted, and the development efficiency and economic benefit of the oil well are improved.
The recurrent neural network algorithm is used for processing and analyzing time series data of the oil reservoir, and the specific forms comprise:
the hidden layer is updated as follows:
output layer calculation:
wherein (1)>Is the hidden layer state at time t, +.>Is the input of the time t at which,、/>、/>is a weight matrix, < >>、/>Is a bias term;
in the present invention, RNNs can be used to analyze the dynamic behavior of the reservoir and the trend of the master factor over time, for example by analyzing the time series of pressure and temperature, to predict future performance of the reservoir and to adjust the development strategy accordingly.
Social network analysis is used to analyze complex interactions and dependencies between elements within a reservoir, and the analysis model includes:
adjacency matrix representation:
node degree centrality calculation:
wherein (1)>Is node->Neighbor node set,/->Is the total number of nodes in the network.
In the invention, the physical field, the chemical field and the mechanical field of the oil reservoir are regarded as nodes in the network, and the interaction between the physical field, the chemical field and the mechanical field forms edges. Through social network analysis, interactions and dependency relationships among the factors can be deeply understood, so that deeper and comprehensive analysis is provided for main control factor evaluation and oil well productivity prediction.
The fourth step specifically comprises:
fuzzy logic analysis: processing fuzzy and uncertain information of oil reservoirs by using fuzzy logic, including geological characteristics, lithology and fluid properties, constructing a fuzzy rule and a fuzzy reasoning system, and converting qualitative description of main control factors into quantifiable model input;
nonlinear regression analysis: capturing complex and nonlinear relations among reservoir main control factors by adopting a Radial Basis Function (RBF) network, and constructing a complex mathematical model among reservoir pressure, temperature and productivity through nonlinear regression analysis;
and (3) constructing an adaptive optimization algorithm: parameter adjustment and optimization are carried out on the model by utilizing a genetic algorithm, and the prediction accuracy and stability of the model are ensured through continuous iteration and self-adaptive adjustment, for example, the parameters of a nonlinear regression model can be self-adaptively adjusted by utilizing the genetic algorithm, so that the actual condition of an oil reservoir can be reflected more accurately;
dynamic and real-time prediction model construction: combining fuzzy logic, nonlinear regression and a self-adaptive optimization algorithm to construct a dynamic and real-time oil well productivity prediction model, wherein the model can be dynamically adjusted according to real-time operation data and geological characteristics of an oil well to realize real-time optimization of productivity prediction;
actual operation data and geological characteristics are fused: the actual operation data of the oil well, including the pressure, the temperature and the flow of the oil well, are combined with geological characteristics such as lithology, porosity and saturation, so that the comprehensiveness and the accuracy of the prediction model are ensured, and the reliability and the practicability of the prediction model are improved through the deep integration of geological and engineering data.
Through the steps, the invention constructs a dynamic and real-time oil well productivity prediction model, fully considers the complexity and the variability of oil reservoirs, and realizes the self-adaptive adjustment of prediction parameters and the optimization of prediction accuracy by applying advanced fuzzy logic, nonlinear regression and self-adaptive optimization technology. The step provides powerful technical support for intelligent management and optimization development of the oil well, and has important practical value and innovation significance.
Radial Basis Function (RBF) networks are used to solve the nonlinear regression problem. The output of the RBF network is a function based on the euclidean distance of the input to the center of each basis function, and is specifically formed by:
RBF layer: the layer calculates the distance between the input vector and the center of the basis function, applies a radial basis function, and based on a Gaussian function, the calculation mode is as follows:
wherein (1)>Is an input vector, +.>Is->The center of the individual basis functions,is the standard deviation;
output layer: this layer is a linear combination of radial basis functions for predicting the output:
wherein (1)>Is a weight of->Is a bias item->Is the number of basis functions;
complex, nonlinear relationships between reservoir master factors, such as modeling reservoir pressure, temperature, and capacity, are captured through the RBF network.
The genetic algorithm is based on a heuristic optimization algorithm, and searches for an optimal solution by simulating natural selection and a genetic mechanism, and the basic steps comprise:
initializing: randomly generating a set of solutions (called a population);
selecting: evaluating the quality of each solution according to the fitness function, classifying the solutions into a first grade and a second grade according to the quality, and selecting the first grade to enter the next generation;
crossover (hybridization): generating child solutions by combining parent solutions;
variation: randomly changing the child solutions with a specific probability;
termination condition: stopping when a predetermined number of iterations is reached or a solution satisfying the condition is found;
in the invention, the parameters (such as a basis function center, standard deviation and weight) of the RBF network are adaptively adjusted by using a genetic algorithm so as to find an optimal oil well productivity prediction model;
the combination of the two methods can realize accurate prediction of the oil well productivity, can adaptively adjust parameters and optimize prediction accuracy, is tightly combined with actual operation data and geological characteristics of the oil well, and has high practical value and innovation significance.
The system also comprises a data preprocessing module for cleaning, sorting and converting the oil well data to adapt to the input requirements of the evaluation model.
In order to verify the effect of the present invention, the following experimental design was performed.
Experimental test
Purpose(s)
It was verified whether the well productivity prediction method by combining Radial Basis Function (RBF) network, genetic algorithm and fuzzy logic could provide more accurate, flexible and adaptive prediction results.
Testing oil fields and data sets
Oil field: testing is performed in a large oilfield having a variety of lithologies and reservoir characteristics.
Data set: historical production data, geologic characteristics, lithology parameters, fluid properties, etc. are included, for a total of 1000 wells of data over 10 years.
Method and Experimental setup
And (3) establishing a model: a nonlinear regression model is built by adopting a radial basis function network, fuzzy logic is used for processing uncertainty, and a genetic algorithm is used for parameter optimization.
Training and verification: training was performed using 80% of the data and 20% of the data were validated.
Comparison reference: comparison is made with conventional linear regression analysis and empirical formulas.
Evaluation index: the prediction accuracy is mainly estimated using the Mean Square Error (MSE) and the decision coefficient (R).
Results
The method of the invention.
MSE：0.045
R²：0.98
Traditional linear regression analysis
MSE：0.12
R²：0.85
Empirical formula
MSE：0.15
R²：0.80
Analysis
From experimental results, the method is superior to the traditional method in terms of MSE and R, shows the advantages of the method in terms of capturing complex relations among oil reservoir main control factors, and can be optimized for specific requirements and conditions of different oil fields and oil reservoirs through the self-adaptive adjustment of a genetic algorithm.
Those of ordinary skill in the art will appreciate that: the discussion of any of the embodiments above is merely exemplary and is not intended to suggest that the scope of the invention is limited to these examples; the technical features of the above embodiments or in the different embodiments may also be combined within the idea of the invention, the steps may be implemented in any order and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.
The present invention is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. of the present invention should be included in the scope of the present invention.
Claims (10)
1. The method for evaluating and predicting the productivity factors of the reservoir oil well is characterized by comprising the following steps:
step one: acquiring microscopic pore and crack structure information of an oil well by utilizing multi-source geological and engineering data, and further analyzing the spatial distribution characteristics of an oil reservoir by combining an earthquake method and an electric method;
step two: introducing a multiphase flow mechanism and reservoir engineering dynamic monitoring, analyzing the motion characteristics of oil and water gas under microscopic and macroscopic dimensions, and combining with the underground pressure and temperature distribution of a reservoir to form an omnibearing main control factor analysis framework;
step three: through deep learning and complex network analysis technology, the interaction of an underground physical field, a chemical field and a mechanical field is synthesized, a multi-scale multi-factor main control factor evaluation model is constructed, and the oil well productivity is accurately quantified and subjected to sensitivity analysis;
step four: according to main control factor analysis, a dynamic and real-time oil well productivity prediction model is constructed by combining actual operation data and geological characteristics of an oil well by using fuzzy logic, nonlinear regression analysis and a self-adaptive optimization algorithm, parameters are adaptively adjusted, and prediction accuracy is optimized;
step five: the system is linked with a field operation system, the automatic control and intelligent optimization of the oil well productivity are realized through the Internet of things and artificial intelligence technology, and the optimal operation of the oil well is realized through continuous learning and optimizing of an operation strategy by an intelligent algorithm;
step six: through the visual interface, the virtual reality technology is combined, the operation state of the oil well and the influence of the main control factors are displayed in real time, and oil field management staff can intuitively understand the operation state of the oil well.
2. The method for estimating and predicting capacity factors of a reservoir well according to claim 1, wherein said step one specifically comprises:
carrying out fine analysis on a rock sample of the oil reservoir by adopting an X-ray CT scanning and nuclear magnetic resonance imaging technology and combining ground and underground data to reveal microstructure characteristics of pores, cracks and oil-water contact of the oil reservoir;
using an acoustic logging tool, performing nondestructive evaluation on permeability, elastic modulus and pore structure of an oil reservoir by analyzing propagation characteristics of acoustic waves in rock, and providing correlation between microstructure and macroscopic physical characteristics of the oil reservoir;
combining the seismic exploration and the ground-electric method, adopting three-dimensional seismic inversion and resistivity imaging technology to construct a three-dimensional spatial model of the oil reservoir, analyzing the spatial distribution characteristics of the oil reservoir, revealing the conditions of broken blocks, folding and broken geological structures of the oil reservoir, and providing spatial references for the development of the oil reservoir;
and integrating and excavating geological and engineering data through an artificial intelligence algorithm, identifying key control factors of an oil reservoir, constructing a comprehensive description model of the oil reservoir, and providing accurate input for subsequent main control factor analysis and yield prediction.
3. The method for estimating and predicting the capacity factor of a reservoir well according to claim 1, wherein the second step comprises:
analyzing the flow characteristics of oil and water vapor in pores and cracks by a multiphase fluid simulation technology based on a finite volume method, revealing the flow mechanism of fluid under different permeabilities and pore structures and the influence of interaction and phase change processes, and specifically comprising the following steps: collecting basic geology and engineering data of related oil reservoirs; performing three-dimensional oil reservoir meshing by using a finite volume method to capture the pore and fracture geometry of the oil reservoir; defining viscosity, density and pressure fluid property parameters of oil, water and gas according to the acquired data; constructing a mathematical model for describing the flow of oil, water and gas in pores and cracks based on Darcy and non-Darcy flow equations; adding a submodel for describing interactions between fluids and phase changes between oil, water, and gas; solving the flow equation by using a finite volume method to obtain the speed, pressure distribution and phase distribution of the fluid in the pores and the cracks; analyzing the flow mechanism of the fluid under different permeabilities and pore structures through solving the results, and how the interaction and phase change process affect the oil well productivity; based on flow mechanism analysis, combining actual operation data of the oil well, and constructing a dynamic and real-time oil well productivity prediction model;
the method utilizes an oil reservoir simulator and fluid dynamics software to analyze the large-scale flow characteristics of the oil reservoir by combining the temperature, pressure and lithology parameters of the oil reservoir, and simulates the pressure distribution, temperature change and fluid saturation in the oil field development process, and specifically comprises the following steps: collecting and setting the temperature, pressure and lithology basic parameters of an oil reservoir; selecting an oil reservoir simulator and fluid dynamics software according to the required analysis precision and range, and initializing an oil reservoir and fluid model according to the collected parameters in the selected simulator and software; simulating pressure distribution, temperature change and fluid saturation in the oil field development process; analyzing the simulation result, and providing corresponding optimization strategies or adjusting the existing development plans aiming at different development stages;
combining dynamic monitoring data of the oil reservoir, wherein the dynamic monitoring data comprise wellhead pressure, temperature and yield, real-time monitoring information, and dynamic correction of an oil reservoir model is realized through data assimilation and real-time updating technologies;
the method comprises the steps of utilizing a field monitoring and sensing technology and combining internet of things equipment to monitor and analyze the underground pressure and the underground temperature of an oil reservoir in real time;
by analyzing the interaction of the fluid at the microscopic level and the macroscopic level, the fluid conveying characteristics of the oil reservoir are revealed, and a basis is provided for the subsequent oil well productivity prediction and intelligent control.
4. The method for estimating and predicting the capacity factor of a reservoir well according to claim 1, wherein said step three specifically comprises:
adopting a recurrent neural network algorithm to process and analyze underground physical, chemical and mechanical data of the oil reservoir, extracting modes and relations in the data, and providing data support for the evaluation of main control factors;
constructing a multi-scale and multi-dimensional interaction model of an oil reservoir by using a complex network analysis technology based on social network analysis, analyzing complex interactions and dependency relations among elements in the oil reservoir, and revealing weights and influences among main control factors, wherein the method specifically comprises the following steps of: collecting data of each element in the oil reservoir, interaction and dependency relationship of each element, and preprocessing the data; selecting a social network analysis tool for complex network analysis based on the collected data, and constructing a multi-scale and multi-dimensional network model for describing complex interactions and dependency relations among elements in the oil reservoir; analyzing and quantifying the weight and influence of each node in the network by using a social network analysis algorithm; according to the analysis result, identifying factors which have dominant influence on the oil well productivity, and carrying out deep analysis on the factors; optimizing the development strategy of the oil well and carrying out accurate productivity prediction by combining the identified main control factors and the corresponding weights and influences thereof;
the method combines the data of the underground physical field, the chemical field and the mechanical field to construct a comprehensive oil reservoir characteristic analysis framework, accurately reflects the complexity and the variability of the oil reservoir, and specifically comprises the following steps: collecting data of a physical field, a chemical field and a mechanical field related to an oil reservoir; preprocessing and standardizing data of physical, chemical and mechanical fields by utilizing a data fusion technology; constructing a comprehensive oil reservoir characteristic analysis framework according to the collected and preprocessed data; evaluating the variability and complexity of the reservoir within the constructed framework using a multi-factor analysis method; the weight of each influence factor in a physical field, a chemical field and a mechanical field is determined through a sensitivity analysis and optimization algorithm so as to accurately reflect the influence of the influence factors on the oil well productivity;
analyzing the influence degree and uncertainty of each main control factor on the oil well production energy by a sensitivity analysis and uncertainty quantification method so as to identify key control factors and optimize an evaluation strategy;
by utilizing the existing geological, engineering and production data and combining artificial intelligence, an evaluation model of the main control factors of the oil well productivity is established and trained, and accurate evaluation and prediction are carried out on different development stages of the oil reservoir.
5. The method for estimating and predicting the capacity factor of a reservoir well according to claim 4, wherein the recurrent neural network algorithm is used for processing and analyzing time series data of the reservoir, and the specific form comprises:
the hidden layer is updated as follows:
output layer calculation:
wherein (1)>Is the hidden layer state at time t, +.>Is the input of time t, < >>、、/>Is a weight matrix, < >>、/>Is a bias term.
6. The method for evaluating and predicting energy production factors of a reservoir well according to claim 5, wherein the social network analysis is used for analyzing complex interactions and dependencies between elements in the reservoir, and the analysis model comprises:
adjacency matrix representation:
node degree centrality calculation: />Wherein (1)>Is node->Neighbor node set,/->Is the total number of nodes in the network.
7. The method for estimating and predicting the capacity factor of a reservoir well according to claim 1, wherein said step four specifically comprises:
fuzzy logic analysis: processing fuzzy and uncertain information of oil reservoirs by using fuzzy logic, including geological characteristics, lithology and fluid properties, constructing a fuzzy rule and a fuzzy reasoning system, and converting qualitative description of main control factors into quantifiable model input;
nonlinear regression analysis: capturing complex and nonlinear relations among oil reservoir main control factors by adopting a radial basis function network;
and (3) constructing an adaptive optimization algorithm: parameter adjustment and optimization are carried out on the model by utilizing a genetic algorithm, and the prediction accuracy and stability of the model are ensured through continuous iteration and self-adaptive adjustment;
dynamic and real-time prediction model construction: combining fuzzy logic, nonlinear regression and an adaptive optimization algorithm to construct a dynamic and real-time oil well productivity prediction model;
actual operation data and geological characteristics are fused: the actual operation data of the oil well, including the pressure, the temperature and the flow of the oil well, are combined with geological characteristics such as lithology, porosity and saturation, so that the comprehensiveness and the accuracy of the prediction model are ensured.
8. The method for estimating and predicting capacity factors of a reservoir well according to claim 7, wherein the radial basis function network is used for solving the nonlinear regression problem, and the output of the RBF network is a function based on the euclidean distance between the input and the center of each basis function, and is specifically formed as follows:
RBF layer: the layer calculates the distance between the input vector and the center of the basis function, applies a radial basis function, and based on a Gaussian function, the calculation mode is as follows:
wherein (1)>Is an input vector, +.>Is->Center of individual basis functions, +.>Is the standard deviation;
output layer: this layer is a linear combination of radial basis functions for predicting the output:
wherein (1)>Is a weight of->Is a bias item->Is the number of basis functions.
9. The method for evaluating and predicting the productivity of a reservoir well according to claim 8, wherein the genetic algorithm searches for the optimal solution by simulating natural selection and genetic mechanism based on a heuristic optimization algorithm, and the basic steps include:
initializing: randomly generating a set of solutions;
selecting: evaluating the quality of each solution according to the fitness function, classifying the solutions into a first grade and a second grade according to the quality, and selecting the first grade to enter the next generation;
crossing: generating child solutions by combining parent solutions;
variation: randomly changing the child solutions with a specific probability;
termination condition: stopping when a predetermined number of iterations is reached or a solution satisfying the condition is found;
and (3) adaptively adjusting parameters of the RBF network by using a genetic algorithm so as to find an optimal oil well productivity prediction model.
10. The method of claim 1, further comprising a data preprocessing module for cleaning, sorting and converting well data to adapt to the input requirements of the evaluation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311167331.XA CN116911216B (en) | 2023-09-12 | 2023-09-12 | Reservoir oil well productivity factor assessment and prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311167331.XA CN116911216B (en) | 2023-09-12 | 2023-09-12 | Reservoir oil well productivity factor assessment and prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116911216A CN116911216A (en) | 2023-10-20 |
| CN116911216B true CN116911216B (en) | 2023-11-24 |
Family
ID=88351453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311167331.XA Active CN116911216B (en) | 2023-09-12 | 2023-09-12 | Reservoir oil well productivity factor assessment and prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116911216B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118228614B (en) * | 2024-05-24 | 2024-07-16 | 中国石油大学(华东) | Multi-agent model-based intelligent extraction method for fractured gas reservoir |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668251A (en) * | 2020-11-30 | 2021-04-16 | 中国石油大学(华东) | Method for simulating fluid flow in multiple nested media of fractured reservoir |
| CA3123281A1 (en) * | 2021-06-25 | 2022-12-25 | Cenovus Energy Inc. | Apparatus and method for oil production forecasting |
| CN115630564A (en) * | 2022-09-27 | 2023-01-20 | 呼和 | Data prediction method and platform of energy big data platform and application |
-
2023
- 2023-09-12 CN CN202311167331.XA patent/CN116911216B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668251A (en) * | 2020-11-30 | 2021-04-16 | 中国石油大学(华东) | Method for simulating fluid flow in multiple nested media of fractured reservoir |
| CA3123281A1 (en) * | 2021-06-25 | 2022-12-25 | Cenovus Energy Inc. | Apparatus and method for oil production forecasting |
| CN115630564A (en) * | 2022-09-27 | 2023-01-20 | 呼和 | Data prediction method and platform of energy big data platform and application |
Non-Patent Citations (1)
| Title |
|---|
| 致密油藏水平井立体动用程度预测;郁林军 等;石油化工应用;第第41卷卷(第第7期期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116911216A (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1984860B1 (en) | Methods, systems, and computer-readable media for fast updating of oil and gas field production models with physical and proxy simulators | |
| US8504341B2 (en) | Methods, systems, and computer readable media for fast updating of oil and gas field production models with physical and proxy simulators | |
| US10621500B2 (en) | Completion design optimization using machine learning and big data solutions | |
| Chen et al. | A committee machine with empirical formulas for permeability prediction | |
| US8855986B2 (en) | Iterative method and system to construct robust proxy models for reservoir simulation | |
| WO2016154383A1 (en) | Probabalistic modeling and analysis of hydrocarbon-containing reservoirs | |
| Nasrabadi et al. | Well placement optimization: A survey with special focus on application for gas/gas-condensate reservoirs | |
| AU2010229934A1 (en) | A system and method for characterizing fractures in a subsurface reservoir | |
| AU2011283109A1 (en) | Systems and methods for predicting well performance | |
| Wang et al. | Fast history matching and optimization using a novel physics-based data-driven model: an application to a diatomite reservoir | |
| EP3469498B1 (en) | Automatic calibration for modeling a field | |
| Kalantari-Dahaghi et al. | Numerical simulation and multiple realizations for sensitivity study of shale gas reservoir | |
| WO2017011469A1 (en) | Ensemble based decision making | |
| CN116911216B (en) | Reservoir oil well productivity factor assessment and prediction method | |
| Han et al. | Comprehensive analysis for production prediction of hydraulic fractured shale reservoirs using proxy model based on deep neural network | |
| Khalili et al. | Reservoir modeling & simulation: Advancements, challenges, and future perspectives | |
| CN117574755B (en) | Hierarchical multistage optimization method for fracturing parameters of horizontal well of shale reservoir well factory | |
| Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
| Gudmundsdottir et al. | Inferring interwell connectivity in fractured geothermal reservoirs using neural networks | |
| WO2019216892A1 (en) | Optimized methodology for automatic history matching of a petroleum reservoir model with ensemble kalman filter | |
| Tom et al. | Volumetric approach based DOE and ANN models for estimating reservoirs oil in place | |
| Zhuang et al. | Co-Optimization of CO2-EOR Strategies Considering the Spatio-Temporal Sequence Prediction of CO2 Flooding and Sequestration | |
| Rezaei et al. | Utilizing a Global Sensitivity Analysis and Data Science to Identify Dominant Parameters Affecting the Production of Wells and Development of a Reduced Order Model for the Eagle Ford Shale | |
| Dong et al. | Assisted history-matching based on multiobjective surrogate reservoir model for tight oil reservoirs | |
| Yan et al. | An Innovative Method for Comprehensive Optimization of Hydraulic Fracturing Parameters to Enhance Production in Tight Oil Reservoirs |
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 |