CN109558967A - Oilfield development program optimization method based on self-teaching mechanism - Google Patents
Oilfield development program optimization method based on self-teaching mechanism Download PDFInfo
- Publication number
- CN109558967A CN109558967A CN201811285474.XA CN201811285474A CN109558967A CN 109558967 A CN109558967 A CN 109558967A CN 201811285474 A CN201811285474 A CN 201811285474A CN 109558967 A CN109558967 A CN 109558967A
- Authority
- CN
- China
- Prior art keywords
- module
- neural network
- monte carlo
- obtains
- recovery
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000007246 mechanism Effects 0.000 title claims abstract description 10
- 238000004519 manufacturing process Methods 0.000 claims abstract description 21
- 238000011084 recovery Methods 0.000 claims abstract description 21
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000004088 simulation Methods 0.000 claims abstract description 10
- 210000005036 nerve Anatomy 0.000 claims abstract description 5
- 238000011161 development Methods 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 5
- 238000003062 neural network model Methods 0.000 abstract description 3
- 238000013459 approach Methods 0.000 abstract description 2
- 238000005457 optimization Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010027336 Menstruation delayed Diseases 0.000 description 1
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000013016 learning Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/045—Combinations of networks
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention proposes a kind of oilfield development program optimization methods based on self-teaching mechanism, it is characterized in that, neural network reservoir modeling module, Monte Carlo manufacturing parameter analog module, RNN result infer training module and prediction module, the following steps are included: in neural network reservoir modeling module, by historical data training nerve, neural network reservoir simulator step is obtained;In Monte Carlo manufacturing parameter analog module, by monte carlo search tree method, the simulation for carrying out each attribute in production process obtains tens of thousands of a recovery schemes by simulating different values to each attribute variable;Network, which is trained, is inferred to RNN result by historical production data, obtains a recovery scheme estimator;In the recovery scheme input recovery scheme estimator that simulation is generated, optimal exploitation scheme is returned.This method goes to approach the process of unknown physics using neural network model, gets rid of the constraint that reservoir numerical simulation simplifies modeling formula, increases accuracy rate.Meanwhile large increase has been obtained in terms of training speed, it can choose the best alternatives in a large amount of development plan.
Description
Technical field
The present invention relates to deep learnings and intensified learning, and in particular to a kind of oil field development based on self-teaching mechanism
Scheme optimization method
Background technique
The selection of traditional oil field recovery scheme is exactly to grind according to achievement and necessary pilot production data is visited in detail in synthesis
To the oil field with industrial value on the basis of studying carefully, from the actual conditions in oil field and production law, to improve final harvesting
For the purpose of rate, reasonable development plan is formulated.With the fast development of deep learning, in identification, classification and natural language processing
Widely apply and obtain preferable accuracy in equal fields;
Have in recent years closest to technology of the invention:
(1), single LSTM model: shot and long term memory models (long-short term memory) are a kind of special
RNN model is proposed to solve the problems, such as RNN model gradient disperse.Its table in reservoir numerical simulation substitution work
Now preferably, but model is more single, and accuracy waits considering.
(2) VMD-BPNN built-up pattern: variation mode decomposition (Variational Mode Decomposition, VMD)
It is a kind of new adaptive signal processing method, is substantially multiple adaptive wiener filter groups, there is good robustness.
BP (back propagation) neural network is to be proposed by the scientist headed by Rumelhart and McClelland for 1986
Concept is a kind of multilayer feedforward neural network according to the training of error backpropagation algorithm.The basic data of the model is more single
One, accuracy waits considering.
In recent years, as late period of oil field, Cut of Oilfield increase, economic benefit is reduced, and whole nation oil at present
Field average recovery ratio is still in reduced levels, and have greatly improved space, needs to carry out development plan optimization.Based on self-teaching
The oilfield development program optimization method of mechanism is gone the process for approaching unknown physics using neural network model, gets rid of oil reservoir
Numerical simulation simplifies the constraint of modeling formula, increases accuracy rate.Meanwhile large increase has been obtained in terms of training speed, energy
It is enough to choose the best alternatives in a large amount of development plan.
Summary of the invention
To solve shortcoming and defect in the prior art, the invention proposes a kind of oil fields based on self-teaching mechanism to open
Originating party case optimization method, which is characterized in that neural network reservoir modeling module, Monte Carlo manufacturing parameter analog module, RNN knot
Fruit infers training module and prediction module, comprising the following steps:
Step (1) obtains neural network oil reservoir by historical data training nerve in neural network reservoir modeling module
Simulator;
Step (2), in Monte Carlo manufacturing parameter analog module, by monte carlo search tree method, produced
The simulation of each attribute in journey obtains tens of thousands of a recovery schemes by simulating different values to each attribute variable;
Step (3) infers that network is trained to RNN result by historical production data, obtains a recovery scheme and pushes away
Disconnected device;
The result that step (1) generates is input in step (2) by step (4), simulates multiple exploitations by step (2)
Scheme is input in the recovery scheme estimator that step (3) obtains;
Step (5) returns to optimal exploitation scheme;
Step (6) then determines surpassing in step (1) and step (3) if it is the training stage by the way of cross matching
Parameter.
Beneficial effects of the present invention:
(1) it because time for calculating each scheme of traditional numerical simulator is very long, is substituted by neural network
Traditional reservoir numerical simulation finds rule from data, is gone to approach unknown object with the neural network model of not physical meaning
The process of reason.The speed of service is fast, and precision is high, solves the problems, such as many of traditional reservoir modeling.
(2) in traditional development plan formulation process, oilfield engineering teacher has to pre-select according to the experience of oneself several
Alternative.After to several program simulations, according to as a result, selecting optimal scheme, human factor influences too much, and considers
Not comprehensive enough, predetermined speed of neural network reservoir modeling is exceedingly fast, so not needing Petroleum Engineer's pre-program, it is only necessary to
Different values is provided to each parameter with Monte Carlo simulation, is finally combined into tens of thousands of or even hundreds of thousands production decision.
(3) in oil reservoir recovery process, the Parameters variation of different moments is larger, and for this time series data, we are adopted
Network is inferred with RNN to realize the selection to recovery scheme.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is that the present invention is based on the flow charts of the oilfield development program optimization method of self-teaching mechanism;
Fig. 2 is that the present invention is based on the structure charts of the oilfield development program optimization method of self-teaching mechanism;
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in Figure 1, the flow chart of the oilfield development program optimization method based on self-teaching mechanism includes four modules:
Neural network reservoir modeling module, Monte Carlo manufacturing parameter analog module, RNN result infer training module and prediction module.
Hidden layer in structure chart, so to find their pattern class, will consider the pass between entirety since each relationship by objective (RBO) is close
System, so using bidirectional circulating neural network (Bidirectional RNNs).
Below with reference to Fig. 1 and Fig. 2, the detailed process of the equipment state prediction method based on deep learning is carried out specifically
It is bright:
Step (1) obtains neural network oil reservoir by historical data training nerve in neural network reservoir modeling module
Simulator
Step (2), in Monte Carlo manufacturing parameter analog module, by monte carlo search tree method, produced
The simulation of each attribute in journey obtains tens of thousands of a recovery schemes by simulating different values to each attribute variable;
Step (3) infers that network is trained to RNN result by historical production data, obtains a recovery scheme and pushes away
Disconnected device;
The result that step (1) generates is input in step (2) by step (4), simulates multiple exploitations by step (2)
Scheme is input in the recovery scheme estimator that step (3) obtains;
Step (5) returns to optimal exploitation scheme;
Step (6) then determines surpassing in step (1) and step (3) if it is the training stage by the way of cross matching
Parameter.
The invention proposes a kind of mining optimization methods based on monte carlo search tree, which is characterized in that neural network
Reservoir modeling module, Monte Carlo manufacturing parameter analog module, RNN result infer training module and prediction module, including following
Step: neural network reservoir simulator step is obtained by historical data training nerve in neural network reservoir modeling module;
The mould of each attribute in production process is carried out by monte carlo search tree method in Monte Carlo manufacturing parameter analog module
It is quasi-, by simulating different values to each attribute variable, obtain tens of thousands of a recovery schemes;Pass through historical production data pair
RNN result infers that network is trained, and obtains a recovery scheme estimator;The recovery scheme that simulation is generated inputs exploitation side
In case estimator, optimal exploitation scheme is returned.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (1)
1. a kind of oilfield development program optimization method based on self-teaching mechanism, which is characterized in that neural network reservoir modeling
Module, Monte Carlo manufacturing parameter analog module, RNN result infer training module and prediction module, comprising the following steps:
Step (1) obtains neural network reservoir modeling by historical data training nerve in neural network reservoir modeling module
Device
Step (2) is carried out in production process in Monte Carlo manufacturing parameter analog module by monte carlo search tree method
The simulation of each attribute obtains tens of thousands of a recovery schemes by simulating different values to each attribute variable;
Step (3) infers that network is trained to RNN result by historical production data, obtains a recovery scheme estimator;
The result that step (1) generates is input in step (2) by step (4), simulates multiple recovery schemes by step (2),
It is input in the recovery scheme estimator that step (3) obtains;
Step (5) returns to optimal exploitation scheme;
Step (6) then determines hyper parameter in step (1) and step (3) if it is the training stage by the way of cross matching.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811285474.XA CN109558967A (en) | 2018-10-30 | 2018-10-30 | Oilfield development program optimization method based on self-teaching mechanism |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811285474.XA CN109558967A (en) | 2018-10-30 | 2018-10-30 | Oilfield development program optimization method based on self-teaching mechanism |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109558967A true CN109558967A (en) | 2019-04-02 |
Family
ID=65865473
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811285474.XA Pending CN109558967A (en) | 2018-10-30 | 2018-10-30 | Oilfield development program optimization method based on self-teaching mechanism |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109558967A (en) |
-
2018
- 2018-10-30 CN CN201811285474.XA patent/CN109558967A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106779151B (en) | A kind of line of high-speed railway wind speed multi-point multi-layer coupling prediction method | |
CN102622418B (en) | Prediction device and equipment based on BP (Back Propagation) nerve network | |
CN102622515B (en) | A kind of weather prediction method | |
CN110400006A (en) | Oil well output prediction technique based on deep learning algorithm | |
Maschio et al. | Bayesian history matching using artificial neural network and Markov Chain Monte Carlo | |
CN103310286A (en) | Product order prediction method and device with time series characteristics | |
CN105631554B (en) | A kind of oil well oil liquid moisture content multi-model prediction technique based on time series | |
CN107909206A (en) | A kind of PM2.5 Forecasting Methodologies based on deep structure Recognition with Recurrent Neural Network | |
CN105279555A (en) | Self-adaptive learning neural network implementation method based on evolutionary algorithm | |
Karkevandi-Talkhooncheh et al. | Application of hybrid adaptive neuro-fuzzy inference system in well placement optimization | |
CN106779148A (en) | A kind of method for forecasting wind speed of high speed railway line of multi-model multiple features fusion | |
Teixeira et al. | Machine learning models to support reservoir production optimization | |
CN109934422A (en) | Neural network wind speed prediction method based on time series data analysis | |
CN108595803A (en) | Shale gas well liquid loading pressure prediction method based on recurrent neural network | |
CN109101712A (en) | Product model designing system and method based on figure network | |
Abdalla et al. | Optimizing the multilayer feed-forward artificial neural networks architecture and training parameters using genetic algorithm | |
CN106529185A (en) | Historic building displacement combined prediction method and system | |
CN104680025A (en) | Oil pumping unit parameter optimization method on basis of genetic algorithm extreme learning machine | |
CN104680023B (en) | Oil pumper parameter optimization method based on multiobjective decision-making | |
Zhu et al. | Structural safety monitoring of high arch dam using improved ABC-BP model | |
CN102279852A (en) | Automatic oil deposit history fitting system and method based on genetic algorithm | |
CN104951871A (en) | Intelligent evaluation method and system for power simulation training | |
CN112539054A (en) | Production optimization method for ground pipe network and underground oil reservoir complex system | |
CN109558967A (en) | Oilfield development program optimization method based on self-teaching mechanism | |
Fattakhov et al. | Using artificial neural networks for analyzing efficiency of advanced recovery methods |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190402 |