CN115204530A - Oil reservoir prediction method based on Fourier neural operator and cyclic neural network - Google Patents
Oil reservoir prediction method based on Fourier neural operator and cyclic neural network Download PDFInfo
- Publication number
- CN115204530A CN115204530A CN202211125454.2A CN202211125454A CN115204530A CN 115204530 A CN115204530 A CN 115204530A CN 202211125454 A CN202211125454 A CN 202211125454A CN 115204530 A CN115204530 A CN 115204530A
- Authority
- CN
- China
- Prior art keywords
- fourier
- operator
- neural
- distribution
- representing
- 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.)
- Granted
Links
- 230000001537 neural effect Effects 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 29
- 125000004122 cyclic group Chemical group 0.000 title claims description 4
- 238000009826 distribution Methods 0.000 claims abstract description 66
- 230000035699 permeability Effects 0.000 claims abstract description 39
- 230000015654 memory Effects 0.000 claims abstract description 23
- 230000000306 recurrent effect Effects 0.000 claims abstract description 13
- 210000005036 nerve Anatomy 0.000 claims abstract description 9
- 238000003860 storage Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 6
- 238000012804 iterative process Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 2
- 238000013136 deep learning model Methods 0.000 claims description 2
- 230000004927 fusion Effects 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 7
- 229910000831 Steel Inorganic materials 0.000 description 4
- 239000010959 steel Substances 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003129 oil well Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
Abstract
The embodiment of the invention discloses an oil reservoir prediction method based on a Fourier neural operator and a recurrent neural network, which comprises the following steps: acquiring current permeability distribution of a target area; extracting a first feature of the current permeability distribution through a trained encoder, wherein the encoder comprises a first convolutional layer, a plurality of first Fourier operators and a second convolutional layer which are sequentially connected; evolving the time dimension based on the first features through a trained convolution long-time and short-time memory neural network to obtain second features of a plurality of time points in the future, wherein each second feature is used for representing the oil reservoir distribution feature of the corresponding time point; and restoring each second characteristic into the oil reservoir distribution of the corresponding time point through a trained decoder, wherein the decoder comprises a third convolution layer, a plurality of second Fourier nerve operators and a fourth convolution layer which are connected in sequence. The embodiment improves the prediction accuracy of the large oil field.
Description
Technical Field
The embodiment of the invention relates to the field of reservoir simulation, in particular to a reservoir prediction method based on a Fourier neural operator and a recurrent neural network.
Background
The oil reservoir prediction means that the future oil reservoir change trend is predicted according to the current oil reservoir data, and effective guidance is provided for oil field development and oil reservoir exploitation. Most of the current reservoir predictions use some commercial software or deep learning based network models.
The commercial software has the problems of slow calculation speed, expensive payment and the like. The existing oil reservoir prediction model based on deep learning mainly aims at that the small oil reservoir effect is good, and the oil reservoir scale and the number of oil wells are limited; and the existing model mainly depends on convolution to calculate, so that the precision in oil reservoir prediction is limited and the generalization is not high.
Disclosure of Invention
The embodiment of the invention provides an oil reservoir prediction method based on a Fourier neural operator and a cyclic neural network, which improves the oil reservoir prediction precision of a large-scale oil field.
In a first aspect, an embodiment of the present invention provides a reservoir prediction method based on a fourier neural operator and a recurrent neural network, including:
acquiring current permeability distribution of a target area;
extracting a first characteristic of the current permeability distribution through a trained encoder, wherein the encoder comprises a first convolutional layer, a plurality of first Fourier operators and a second convolutional layer which are sequentially connected;
carrying out evolution on the time dimension based on the first characteristics by a trained convolution long-time memory neural network to obtain second characteristics of a plurality of time points in the future, wherein each second characteristic is used for representing the oil deposit distribution characteristics of the corresponding time point;
and restoring each second characteristic into the oil reservoir distribution at the corresponding time point through a trained decoder, wherein the decoder comprises a third convolutional layer, a plurality of second Fourier neural operators and a fourth convolutional layer which are sequentially connected.
In a second aspect, an embodiment of the present invention provides an electronic device, including:
one or more processors;
a memory for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the above-described fourier neural operator and recurrent neural network-based reservoir prediction method.
In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the above-mentioned reservoir prediction method based on fourier nerve operators and a recurrent neural network.
According to the embodiment of the invention, the oil reservoir control equation is simulated by combining the Fourier operator and the neural network, and the derivation process of solving the Fourier operator is completely carried out according to the solution of partial differential equation, so that the method is more consistent with the process of numerical solution and has higher prediction precision; and because the convolution used in the data space is less, the calculation speed is faster; in addition, the method can predict the pressure and the saturation at the same time, and does not need to predict the pressure and the saturation twice, so that the prediction efficiency is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic diagram of a reservoir prediction model based on Fourier neural operators and a recurrent neural network according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of an encoder according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a decoder according to an embodiment of the present invention.
FIG. 4 is a flowchart of a reservoir prediction method based on Fourier neural operators and a recurrent neural network according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a working condition of an actual example according to an embodiment of the present invention.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should also be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The embodiment of the invention provides an oil reservoir prediction method based on FNO (Fourier Neural Operator) and a recurrent Neural network. To illustrate the method, a reservoir prediction model that implements the method is preferably described. Fig. 1 is a schematic diagram of a reservoir prediction model provided in an embodiment of the present invention, which is used for simulating reservoir distribution changes under different oil permeability conditions, so as to simulate the influence of geological changes on the mining process.
As shown in fig. 1, the model includes: an encoder, a ConvSLTM (Convolutional Long Short Term Memory) Convolutional Long-and-Short time Memory neural network and a decoder. Inputting the current permeability distribution m representing the current geological condition into a model, firstly extracting features through an encoder, then carrying out time evolution through a convolution time memory neural network, and finally reconstructing pressure distribution and saturation distribution through a decoder.
Specifically, the encoder includes a first convolution layer, a plurality of first fourier nerve operators, and a second convolution layer connected in sequence, as shown in fig. 2.
ConvLSTM was developed from fully-connected LSTM, and the output of fully-connected LSTM through the fully-connected network is used as the input of LSTM, so that the data passing through the fully-connected network can be well represented in time series by means of the recurrent neural network LSTM, but the capability of the fully-connected network in capturing the spatial characteristics of the data is still insufficient, and the generation of ConvLSTM solves the problem well. ConvLSTM has a similar chain-like structure to a standard recurrent neural network, while controlling information transfer through a gating unit. The ConvLSTM gate control unit has four interaction layers including an input gate, an output gate, a forgetting gate, and a tanh layer. The gate is used for controlling how much information of the previous step can be transmitted to the current time step, and simultaneously determining how much current information is distributed to be transmitted to the next time step. The forgetting gate layer determines which parts in the hidden state of the previous step are transmitted to the current step; the input gate layer decides which new information needs to be added to the current state; the tanh layer creates a new candidate value that can be added to the current state; the final output value is determined by the output gate and the current state value. The ConvLSTM network and its variants can efficiently extract spatial and temporal information.
The decoder includes a third convolutional layer, a plurality of second fourier operators, and a fourth convolutional layer, which are sequentially connected, as shown in fig. 3. In the decoding process, the output of the last layer of the second FNO of each layer is spliced with the output of the first FNO of each layer in the encoder and is used as the input of the second FNO of each layer. As shown in FIG. 1, it is assumed that the first FNO and the second FNO have three layers, respectively, and the first FNO of the three layers outputs the characteristic F 1 、F 2 And F 3 Then F will be 3 After being spliced with the output of the ConvLSTM, the output is input into a first layer and a second layer of FNO; f is to be 2 After being spliced with the output of the first layer of second FNO, the output is input into the second layer of second FNO; f is to be 3 And after being spliced with the output of the second FNO of the second layer, the output of the second FNO of the third layer is input into the second FNO of the third layer.
The functions and arrangement of the modules will be described in detail in the following embodiments.
Based on the model schematic, fig. 4 is a flowchart of an oil reservoir prediction method based on a fourier neural operator and a recurrent neural network, which is provided by the embodiment of the invention and is suitable for predicting the oil reservoir distribution under certain geological conditions. The method is executed by an electronic device, and specifically includes the following steps, as shown in fig. 4:
and S110, acquiring the current permeability distribution of the target area.
The target area refers to a geographical area to be studied, such as a certain oil field. The permeability distribution refers to the distribution of permeability composition at each location within the target area. As shown in FIG. 5, one 191 is provided
313
The 3D practical example of 7 is the permeability distribution and well site distribution along the X direction. Where PERMX represents permeability and black dots represent well sites. The current permeability distribution refers to a data volume composed of permeability distributions at the current time point.
More specifically, assume that the target area has X number of two-dimensional geographic grids
Y, X and Y represent the number of grids in two perpendicular directions parallel to the ground, respectively, and then the current permeability distribution has a dimension of [ X, Y,1]]The three-dimensional data volume of (1). When N calculations are predicted simultaneously, the current permeability distribution has a dimension of [ N, X, Y,1]]N is a natural number. In particular, X and Y are arbitrary natural numbers, and are not limited by the dimensions of the model training samples.
And S120, extracting the first characteristic of the current permeability distribution through the trained encoder.
The main role of the encoder is to reduce the dimensionality (i.e., dimensionality) of the input data and to extract features. Firstly, inputting the current permeability distribution into the first convolution layer for feature extraction to obtain an initial feature. Optionally, the first winding layer comprises 32 and 3 winding layers
3, after the current permeability distribution is calculated by the first convolution layer, the obtained initial characteristic dimension is [ M, X, Y,64 ]]。
Then, inputting the initial features into the plurality of first Fourier operators, gradually performing dimension reduction in a Fourier space, and extracting depth features, wherein an iterative process of each first Fourier operator is as follows:
wherein W and
are all the weight parameters that are trained well,
input data representing a current first fourier neural operator,lrepresent
Is indexed by the dimension of the object to be scanned,
presentation pair
As a result of the fourier transform being performed,
an index representing a wave of a different frequency,
to represent
Is indexed by the dimension of the object to be scanned,
to represent
The total number of dimensions of (a) is,
which represents the inverse of the fourier transform,
representing an activation function representing input data of a current first Fourier neural operator;
is 1 in 64 steps of 2
1 convolution, and the dimension of each first Fourier neural operator output feature is 1/2 of the input feature.
Specifically, the implementation of the fourier neural operator is divided into three steps: transforming the data into Fourier space by Fourier transform, and taking the higher frequency data in the Fourier spaceAnd calculating the wave bands, and mapping the data to an original space by adopting Fourier inverse change. In the embodiment, when inverse fourier transform is used, the output size of the inverse fourier transform is directly changed to 1/2 of the original size, so that the features can be evolved in the fourier space while the dimension of the input data is reduced. The calculation of each fourier neural operator is as follows: by using
Iterating the numerical solution to solve a differential equation, specifically, the iterative equation is as follows:
wherein,xrepresenting input variables, D represents the dimension of the variables, and the input of each layer of FNO is
Thus can be used forxAs a virtual variable, the number of variables,
a linear differential operator is represented by a linear differential operator,
it is indicated that one of the parameters,
expressing unknown variables the solution to differential equations is converted into a solver function by equation (1)
。
Wherein,
is a parameter of
The neural network operator of (2) needs to be learned from the data. If operator pairs are removed
Is dependent on, order
Then the above equation becomes the convolution calculation:
since the convolution can be multiplied after fourier transform and then inverse transformed back to the original space, equation (3) evolves:
wherein,
representing a fourier transform. In fact, now that
Is a parameter to be learned, and
the space learning is not directly learned in the Fourier space, but is followed by Fourier transform, so that a convolution operator defined in the Fourier space is directly used
In place of the above formula
：
The final equation (5) can be converted to equation (6):
for the
，
，
Is about the frequency domain
As a function of (c). And for a periodic function, expanding the periodic function by using Fourier series, and selecting a limited number of series expansion terms to approximate the whole process by truncation of the series in the data calculation process. That is, a finite pair of frequency waves is selected in the frequency domain
And (6) optimizing. Assume that the number of selections is:
parameterization directly with truncated Fourier frequency
Finally, finally
Become (a)
) Is not less than the tensor sum
Is irrelevant (i.e. the
。
The hyper-parameters of the model need to be determined at the time of model construction. The final equation (6) becomes:
due to the fact thatxAs a virtual variable, equation (8) becomes:
this is an iterative process of fourier neural operators, in which,
and
are weight parameters obtained by gradient descent optimization. From the above calculations, it can be seen that the calculation of the fourier neural operator is more consistent with the solution of the partial differential equation, and therefore can achieve higher accuracy with fewer features as an encoder.
It is worth mentioning that the present embodiment performs down-sampling of dimensions while performing evolution in the fourier space.
Are 64 convolution operators of size 1x1, step size 2, with parameter size 64,
the size of the recovered space becomes 1/2 of the original size. Thus, after each Fourier operator operation, the output becomes 1/2 of the original.
In one embodiment, the current permeability profile is assumed to be [ M, X, Y,1]]M represents the number of examples, X and Y represent the number of grids of the target area in two perpendicular directions parallel to the ground, respectively; as shown in fig. 2, there are 3 of the first fourier operators. The first coiled layer comprises 32 and 3
3, after the current permeability distribution is calculated by the first convolution layer, the obtained initial characteristic dimension is [ M, X, Y,64 ]](ii) a After the initial features pass through a first Fourier neural operator, the dimension of the output features is [ M, X/2,Y/2,64]The number of parameters of the model is
，
Representing the maximum fourier series selected in the first fourier neural operator,
. After the output characteristic of the first Fourier neural operator passes through the second first Fourier neural operator, the dimension of the output characteristic is [ M, X/4,Y/4,64]The number of parameters of the model is
；
Represents a second first Fourier operatorMaximum fourier series taken. After the output characteristic of the second first Fourier neural operator passes through the third first Fourier neural operator, the output depth characteristic dimension is [ M, X/8,Y/8,64]The number of parameters of the model is
，
Representing the maximum fourier series selected in the third first fourier neural operator,
。
and after the depth feature is obtained, inputting the depth feature into the second convolution layer for feature fusion to obtain the first feature of the current permeability distribution. Optionally, the second convolutional layer comprises 32 and 3 convolutional layers
3, the number of parameters of the second convolution layer is 32
3
3。
S130, carrying out time dimension evolution on the basis of the first characteristics through a trained convolution long-time and short-time memory neural network to obtain second characteristics of a plurality of time points in the future, wherein each second characteristic is used for representing the oil reservoir distribution characteristics of the corresponding time point.
The main roles of ConvLSTM are: based on the first feature, the time dimension is then evolved. Since the first feature after dimensionality reduction still contains the geospatial information, the ConvLSTM is adopted for feature extraction, and T (T represents the number of future time points needing to be calculated) different features are obtained and used for recovering the pressure and the saturation. For ease of distinction and description, the features of the encoder output are referred to as first features and the features of the ConvLSTM output are referred to as second features.
And S140, restoring each second characteristic into the oil reservoir distribution at the corresponding time point through the trained decoder.
The reservoir distribution refers to the distribution of reservoir data components at each location within the target region. Wherein the reservoir data at each location includes at least one of: oil phase pressure, oil phase saturation, water phase pressure, and water phase saturation. The oil reservoir distribution at a certain time point is a three-dimensional data body with the dimensionality [ X, Y, M ], wherein M represents the channel number of the oil reservoir data, and X and Y respectively represent the geography grid dimensionality. When N calculations are predicted simultaneously, the current reservoir distribution is a four-dimensional data volume with the size [ N, X, Y, M ].
The decoder is mainly used for up-sampling the second characteristics corresponding to each time point obtained by ConvLSTM to obtain the final oil reservoir distribution. Optionally, first, each second feature is input into the third convolution layer to perform feature extraction. As shown in FIG. 3, the first convolution layer includes 64 3 with step size 1
3 convolution kernels for feature extraction of the second feature inputs, the number of parameters of the first convolution layer being 64
3
3。
Then, inputting the output features of the third convolutional layer into the plurality of second fourier neural operators, and performing dimension lifting and feature evolution in a fourier space step by step, wherein an iterative process of each second fourier neural operator includes: and (3) adopting interpolation up-sampling to change the dimension of the output characteristic into 2 times of the input characteristic, and adopting Fourier transform and inverse Fourier transform to evolve the up-sampled characteristic. As shown in fig. 3, 3 second fourier operators are used for feature dimension promotion and evolution. In each Fourier operator, firstly, an interpolation up-sampling mode is adopted to change the input into 2 times of the original input, then, 64 convolution kernels with the step length of 2 and the 1x1 are adopted to carry out convolution calculation, and the quantity of the parameters of the part is 64.
Accordingly, in the fourier section,
the size of the recovered space also becomes 2 times of the original size. Thus, after each second fourier neural operator operation, the overall output becomes 2 times the original output. Also exemplified in the above embodiments, the permeability profile is [ M, X, Y,1 [ ]]Each second feature is [ M, X/8,Y/8,32]M represents the number of examples. The fourth convolutional layer comprises 64 3 with step size 1
3 convolution kernel with parameter number of 64
3
3, after each second feature passes through the fourth convolution layer, the dimension of the output feature is [ M, X/8,Y/8,64]. After the output characteristic of the fourth convolution layer passes through the first Fourier nerve operator, the dimension of the output characteristic is [ M, X/4,Y/4,64]The number of parameters of the model is
，
Representing the maximum fourier series selected in the first and second fourier operators,
. The output characteristic of the first and second Fourier operatorsAfter the second Fourier operator, the output characteristic dimension is [ M, X/2,Y/2,64]The number of parameters of the model is
，
Representing the maximum fourier series selected in the second fourier neural operator,
. The output characteristic dimension of the second Fourier operator is [ M, X, Y,64 ] after the output characteristic of the second Fourier operator passes through the third second Fourier operator]The number of parameters of the model is
，
Representing a selected maximum Fourier series of said third second Fourier neural operator,
。
and finally, inputting the output characteristics of the plurality of second Fourier operators into the fourth convolution layer for characteristic extraction to obtain the oil reservoir distribution of the corresponding time point. The fourth convolution layer comprises 2 and 3 layers
3, number of parameters is 2
3
3。
In addition, the fourier nerve operators in the encoder and the decoder are multiple, so that the down-sampling, the feature extraction and the up-sampling are performed step by step, and the excessive information loss is avoided.
In the embodiment, the reservoir control equation is simulated by using a mode of combining the Fourier operator and the neural network, and because the derivation process of solving the Fourier operator is completely carried out according to the solution of the partial differential equation, the method is more consistent with the process of numerical solution and has higher prediction precision; and because the convolution used in the data space is less, the calculation speed is faster; in addition, the method can predict the pressure and the saturation at the same time, and does not need to predict the pressure and the saturation twice, so that the prediction efficiency is further improved. Based on the simulation result of the full physical model, the average relative error of the method in the aspects of pressure and saturation prediction is 4.1 percent and 2.1 percent, 0.2s is needed for completing one-time prediction, and the prediction results of the pressure and the saturation are obtained at one time.
In particular, the present embodiment implements data upsampling and downsampling while evolving features in fourier space using fourier transforms with different input and output dimensions in the encoder and decoder. The method is particularly suitable for scenes taking certain geological condition distribution (embodied as permeability distribution) as a prediction premise, and the permeability fields are different due to the change of geological conditions and need to be adjusted through terrain fitting, so that the characteristic extraction, the up-sampling and the down-sampling processes are integrated into the same operation, the calculated amount can be reduced, the fitting of different geological condition distribution is facilitated, and the application range of the model is improved.
Optionally, resNet or densnet is also included before and after ConvLSTM to improve model performance. For example, based directly on the last layer output of the fourier, a convolution of 3 × 3 is used for the calculation, and 3 layers of ReseNet or densnet are used.
On the basis of the above embodiment and the following embodiment, the present embodiment refines the training process of the entire model. Optionally, before the trained encoder extracts the first feature of the current permeability distribution, the method further includes the following steps:
first, a plurality of examples of different permeability distributions are randomly generated, and LandSim simulation is adoptedThe software simulates the examples and extracts the pressure distribution, saturation distribution and well oil and water production. As shown at 191 in FIG. 5
313
7, a plurality of samples with different permeabilities are randomly generated on the basis of the 3D actual sample, and software simulation and data extraction are carried out.
The extracted data is then normalized and divided into a training set and a test set. Optionally, the data is normalized by a maximum and minimum method, and the normalized data is normalized according to the ratio of 8:2 into training and test sets.
Meanwhile, in order to better adjust the importance of each part of output, the present embodiment divides the total loss function into two parts, namely pressure loss and saturation, calculates the pressure loss by using a 1 norm, calculates the saturation loss by using a 2 norm, and constructs the following loss function:
wherein,
denotes saturation loss, N t Denotes the number of samples, n b Represents the total number of grids per sample, T represents the number of future time points,
representing the true saturation of grid j at time point t,
represents grid j at timePredicted saturation at the intermediate point t;
it is indicated that the pressure loss is,
representing the true pressure at time point t of grid j,
represents the predicted pressure at time t for grid j;
which represents the overall loss of the power,
and
representing weights, 4 and 1 can be chosen, respectively.
And finally, training a deep learning model formed by the encoder, the convolution long-time and short-time memory neural network and the decoder by using the training samples and the loss function. Specifically, the number of epochs is 100 by adopting an ADAM optimizer, the initial learning rate is 0.0002, and the learning rate is attenuated to 0.8 times of the original learning rate every 10 epochs.
And after the training is finished, the model parameters are stored as corresponding model files. In the prediction stage, a model is loaded, then the initial permeability field of the oil reservoir is used as input, the distribution of pressure and saturation at a series of moments is obtained through model calculation, and further the relevant information of the well is calculated.
According to the embodiment, large-scale actual oil field data are used as samples, an oil reservoir prediction model based on the Fourier neural operator and the circulating neural network is trained, and model accuracy is improved. The model uses less convolution in the data space and the training speed is faster. Compared with other technical schemes, the model can achieve better precision and higher speed.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 6, the electronic device includes a processor 50, a memory 51, an input device 52, and an output device 53; the number of processors 50 in the device may be one or more, and one processor 50 is taken as an example in fig. 6; the processor 50, the memory 51, the input device 52 and the output device 53 in the apparatus may be connected by a bus or other means, as exemplified by the bus connection in fig. 6.
The memory 51 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the method, device, and storage medium for determining parameters and predicting concentration of orthotropic plates of steel box girders in the embodiments of the present invention. The processor 50 executes various functional applications and data processing of the device by running software programs, instructions and modules stored in the memory 51, namely, the method, the device and the storage medium for determining the steel box girder orthotropic plate parameters and predicting the concentration are realized.
The memory 51 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 51 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 51 may further include memory located remotely from the processor 50, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 52 is operable to receive input numeric or character information and to generate key signal inputs relating to user settings and function controls of the apparatus. The output device 53 may include a display device such as a display screen.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the method, the equipment and the storage medium for determining the steel box girder orthotropic plate parameters and predicting the concentration of the steel box girder orthotropic plate of any embodiment are realized.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing. Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, or the like, as well as conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the technical solutions of the embodiments of the present invention.
Claims (10)
1. A reservoir prediction method based on Fourier neural operators and a recurrent neural network is characterized by comprising the following steps:
acquiring current permeability distribution of a target area;
extracting a first characteristic of the current permeability distribution through a trained encoder, wherein the encoder comprises a first convolutional layer, a plurality of first Fourier operators and a second convolutional layer which are sequentially connected;
carrying out evolution on the time dimension based on the first characteristics by a trained convolution long-time memory neural network to obtain second characteristics of a plurality of time points in the future, wherein each second characteristic is used for representing the oil deposit distribution characteristics of the corresponding time point;
and restoring each second characteristic into the oil reservoir distribution of the corresponding time point through a trained decoder, wherein the decoder comprises a third convolution layer, a plurality of second Fourier nerve operators and a fourth convolution layer which are connected in sequence.
2. The method of claim 1, wherein the reservoir profile comprises at least one of: oil phase pressure distribution, oil phase saturation distribution, water phase pressure distribution and water phase saturation distribution.
3. The method of claim 1, wherein the permeability distribution and the reservoir distribution refer to a two-dimensional geographic distribution, wherein the number of two-dimensional geographic grids is any number.
4. The method of claim 1, wherein extracting, by the trained encoder, the first feature of the current permeability distribution comprises:
inputting the current permeability distribution into the first convolution layer for feature extraction to obtain initial features;
inputting the initial features into the plurality of first Fourier nerve operators, gradually reducing dimensions in a Fourier space and extracting depth features, wherein the iteration process of each first Fourier nerve operator is as follows:
wherein W and
are all the weight parameters that are trained well,
input data representing a current first fourier neural operator,lto represent
Is indexed by the dimension of the object to be scanned,
presentation pair
As a result of the fourier transform being performed,kan index representing a wave of a different frequency,
to represent
Is indexed by the dimension of the object to be scanned,
to represent
The total number of dimensions of (a) is,
which represents the inverse of the fourier transform,
it is shown that the activation function is,
input data representing a current first fourier neural operator;
is 1 in 64 steps of 2
1 convolution, wherein the dimension of each first Fourier neural operator output feature is 1/2 of the input feature;
and inputting the depth features into the second convolution layer for feature fusion to obtain the first features of the current permeability distribution.
5. The method of claim 4, wherein the current permeability distribution is a tensor of [ M, X, Y,1], M representing the number of examples, X and Y representing the number of grids of the target area in two perpendicular directions parallel to the ground, respectively; there are 3 of the first Fourier operators;
the first coiled layer comprises 32 and 3
3, after the current permeability distribution is calculated by the first convolution layer, the obtained initial characteristic dimension is [ M, X, Y,64 ]]；
After the initial features pass through a first Fourier neural operator, the dimension of the output features is [ M, X/2,Y/2,64]Wherein the number of parameters of the first Fourier operator is
，
Representing a selected maximum Fourier series of said first Fourier neural operator,
；
after the output characteristic of the first Fourier neural operator passes through the second first Fourier neural operator, the dimension of the output characteristic is [ M, X/4,Y/4,64]Wherein the second first Fourier neural operator has a number of parameters of
，
Representing the maximum Fourier series selected in said second first Fourier neural operator,
；
after the output characteristic of the second first Fourier neural operator passes through the third first Fourier neural operator, the output depth characteristic dimension is [ M, X/8,Y/8,64]Wherein the number of parameters of the third first Fourier neural operator is
，
Representing the maximum fourier series selected in the third first fourier neural operator,
；
the second convolution layer comprises 32 and 3
3, the number of parameters of the second convolutional layer is 32
3
3。
6. The method of claim 1, wherein the restoring, by the trained decoder, each second feature to the reservoir distribution at the corresponding time point comprises:
inputting each second feature into the third convolutional layer for feature extraction;
inputting the output features of the third convolutional layer into the plurality of second fourier neural operators, and performing dimension lifting and feature evolution gradually in a fourier space, wherein an iterative process of each second fourier neural operator comprises: adopting interpolation up-sampling to change the dimension of the output characteristic into 2 times of the input characteristic, and adopting Fourier transform and inverse Fourier transform to evolve the up-sampled characteristic;
and inputting the output characteristics of the plurality of second Fourier nerve operators into the fourth convolution layer for characteristic extraction, so as to obtain the oil reservoir distribution of the corresponding time point.
7. The method of claim 6, wherein the permeability distribution is a tensor of [ M, X, Y,1], each second feature is a tensor of [ M, X/8,Y/8,32], M represents an arithmetic number, and X and Y represent the number of meshes of the target area in two perpendicular directions parallel to the ground, respectively; 3 of the second Fourier operators;
the fourth convolutional layer comprises 64 3 with step size 1
3 convolution kernel with parameter number of 64
3
3, after each second feature passes through the fourth convolution layer, the dimension of the output feature is [ M, X/8,Y/8,64]；
After the output characteristic of the fourth convolution layer passes through the first Fourier neural operator, the dimension of the output characteristic is [ M, X/4,Y/4,64 ]]Wherein the number of parameters of the first and second Fourier operators is
，
Representing a selected maximum Fourier series of said first second Fourier neural operator,
；
the output characteristic dimension of the first Fourier neural operator and the output characteristic dimension of the second Fourier operator are [ M, X/2,Y/2,64 ] after the output characteristic of the first Fourier neural operator and the second Fourier operator pass through]Wherein the number of parameters of the second Fourier operator is
，
Representing a selected maximum Fourier series in said second Fourier neural operator,
；
the output characteristic dimension of the second Fourier operator is [ M, X, Y,64 ] after the output characteristic of the second Fourier operator passes through the third second Fourier operator]Wherein the third second Fourier operator has a parameter number of
；
Representing the maximum selected in the third second Fourier neural operatorA Fourier series of the series of Fourier transforms,
；
the fourth convolution layer comprises 2 and 3 layers
3, number of parameters is 2
3
3。
8. The method of claim 1, wherein prior to said extracting, by the trained encoder, the first feature of the current permeability distribution, further comprising:
randomly generating a plurality of examples with different permeability distributions, simulating the examples by using LandSim simulation software, and extracting pressure distribution, saturation distribution and oil and water production of a well;
normalizing the extracted data, and dividing the normalized data into a training set and a test set;
the following loss function was constructed:
wherein,
denotes the loss of saturation, N t Denotes the number of samples, n b The total number of grids per sample is represented,Tthe number of future points in time is indicated,
representing the true saturation of grid j at time point t,
representing the predicted saturation of grid j at time point t;
it is indicated that the pressure loss is,
representing the true pressure at time point t of grid j,
represents the predicted pressure at time t for grid j;
which represents the overall loss of the power,
representing a weight;
and training a deep learning model formed by the encoder, the convolution long-time and short-time memory neural network and the decoder by using the training samples in the training set and the testing set and the loss function.
9. An electronic device, comprising:
one or more processors;
a memory for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method for reservoir prediction based on fourier neural operators and cyclic neural networks as recited in any of claims 1-8.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method for reservoir prediction based on fourier neural operators and recurrent neural networks as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211125454.2A CN115204530B (en) | 2022-09-16 | 2022-09-16 | Oil reservoir prediction method based on Fourier neural operator and cyclic neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211125454.2A CN115204530B (en) | 2022-09-16 | 2022-09-16 | Oil reservoir prediction method based on Fourier neural operator and cyclic neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115204530A true CN115204530A (en) | 2022-10-18 |
| CN115204530B CN115204530B (en) | 2023-05-23 |
Family
ID=83572133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211125454.2A Active CN115204530B (en) | 2022-09-16 | 2022-09-16 | Oil reservoir prediction method based on Fourier neural operator and cyclic neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115204530B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101951595B1 (en) * | 2018-05-18 | 2019-02-22 | 한양대학교 산학협력단 | Vehicle trajectory prediction system and method based on modular recurrent neural network architecture |
| CN110717270A (en) * | 2019-10-10 | 2020-01-21 | 南京特雷西能源科技有限公司 | Oil and gas reservoir simulation method based on data |
| CN112541572A (en) * | 2020-11-25 | 2021-03-23 | 中国石油大学(华东) | Residual oil distribution prediction method based on convolutional encoder-decoder network |
| CN113052371A (en) * | 2021-03-16 | 2021-06-29 | 中国石油大学(华东) | Residual oil distribution prediction method and device based on deep convolutional neural network |
| CN114152977A (en) * | 2020-09-07 | 2022-03-08 | 中国石油化工股份有限公司 | Reservoir parameter prediction method and device based on geological feature constraint and storage medium |
| CN114282725A (en) * | 2021-12-24 | 2022-04-05 | 山东大学 | Construction of transient oil reservoir agent model based on deep learning and oil reservoir prediction method |
| CN114462262A (en) * | 2021-12-07 | 2022-05-10 | 中国海洋石油集团有限公司 | History fitting prediction method based on dual dimensionality of time and space |
| CN114492213A (en) * | 2022-04-18 | 2022-05-13 | 中国石油大学(华东) | Wavelet neural operator network model-based residual oil saturation and pressure prediction method |
| CN114492211A (en) * | 2022-04-15 | 2022-05-13 | 中国石油大学(华东) | Residual oil distribution prediction method based on autoregressive network model |
| CN114693005A (en) * | 2022-05-31 | 2022-07-01 | 中国石油大学(华东) | Three-dimensional underground oil reservoir dynamic prediction method based on convolution Fourier neural network |
-
2022
- 2022-09-16 CN CN202211125454.2A patent/CN115204530B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101951595B1 (en) * | 2018-05-18 | 2019-02-22 | 한양대학교 산학협력단 | Vehicle trajectory prediction system and method based on modular recurrent neural network architecture |
| CN110717270A (en) * | 2019-10-10 | 2020-01-21 | 南京特雷西能源科技有限公司 | Oil and gas reservoir simulation method based on data |
| CN114152977A (en) * | 2020-09-07 | 2022-03-08 | 中国石油化工股份有限公司 | Reservoir parameter prediction method and device based on geological feature constraint and storage medium |
| CN112541572A (en) * | 2020-11-25 | 2021-03-23 | 中国石油大学(华东) | Residual oil distribution prediction method based on convolutional encoder-decoder network |
| CN113052371A (en) * | 2021-03-16 | 2021-06-29 | 中国石油大学(华东) | Residual oil distribution prediction method and device based on deep convolutional neural network |
| CN114462262A (en) * | 2021-12-07 | 2022-05-10 | 中国海洋石油集团有限公司 | History fitting prediction method based on dual dimensionality of time and space |
| CN114282725A (en) * | 2021-12-24 | 2022-04-05 | 山东大学 | Construction of transient oil reservoir agent model based on deep learning and oil reservoir prediction method |
| CN114492211A (en) * | 2022-04-15 | 2022-05-13 | 中国石油大学(华东) | Residual oil distribution prediction method based on autoregressive network model |
| CN114492213A (en) * | 2022-04-18 | 2022-05-13 | 中国石油大学(华东) | Wavelet neural operator network model-based residual oil saturation and pressure prediction method |
| CN114693005A (en) * | 2022-05-31 | 2022-07-01 | 中国石油大学(华东) | Three-dimensional underground oil reservoir dynamic prediction method based on convolution Fourier neural network |
Non-Patent Citations (3)
| Title |
|---|
| 王宇: "致密油藏压裂水平井产能模型与应用研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑(月刊)》 * |
| 薛婷 等: "一种确定水驱气藏动态储量及水体能量的新方法", 《大庆石油地质与开发》 * |
| 马承杰: "基于循环神经网络的油藏产量预测方法", 《内蒙古石油化工》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115204530B (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113052371B (en) | Residual oil distribution prediction method and device based on deep convolutional neural network | |
| CN111523713B (en) | Method and device for predicting saturation distribution of residual oil in oil field | |
| Sreekanth et al. | Stochastic and robust multi-objective optimal management of pumping from coastal aquifers under parameter uncertainty | |
| US20170338802A1 (en) | Actually-measured marine environment data assimilation method based on sequence recursive filtering three-dimensional variation | |
| CN111832227B (en) | Shale gas saturation determination method, device and equipment based on deep learning | |
| NO20230580A1 (en) | Estimating reservoir production rates using machine learning models for wellbore operation control | |
| CN115186936A (en) | Optimal well pattern construction method for oil field based on GNN model | |
| Kumar et al. | Ensemble-based assimilation of nonlinearly related dynamic data in reservoir models exhibiting non-Gaussian characteristics | |
| CN117076931A (en) | Time sequence data prediction method and system based on conditional diffusion model | |
| WO2022241137A1 (en) | Physics-informed attention-based neural network | |
| CN115204531B (en) | Oil reservoir prediction method, equipment and medium based on Fourier neural operator | |
| Razak et al. | Embedding physical flow functions into deep learning predictive models for improved production forecasting | |
| CN112949944A (en) | Underground water level intelligent prediction method and system based on space-time characteristics | |
| CN110486009B (en) | Automatic parameter reverse solving method and system for infinite stratum | |
| CN115604131B (en) | Link flow prediction method, system, electronic device and medium | |
| CN112444850B (en) | Seismic data velocity modeling method, storage medium and computing device | |
| CN115204530B (en) | Oil reservoir prediction method based on Fourier neural operator and cyclic neural network | |
| Samuel et al. | Fast modelling of gas reservoir performance with proper orthogonal decomposition based autoencoder and radial basis function non-intrusive reduced order models | |
| CN116842375A (en) | Landslide displacement prediction method and system based on VMD and GA-Elman | |
| CN116011338A (en) | Full waveform inversion method based on self-encoder and deep neural network | |
| CN115375867A (en) | Method, system, equipment and medium for calculating geothermal resource quantity by using grid model | |
| Al-Shamma et al. | History matching of the Valhall field using a global optimization method and uncertainty assessment | |
| JP7512090B2 (en) | Earthquake motion evaluation model generation method, earthquake motion evaluation model generation device, earthquake motion evaluation method, and earthquake motion evaluation device | |
| Zhao et al. | Online generic diagnostic reservoir operation tools | |
| CN115169761A (en) | GNN and LSTM based complex well pattern oil field yield prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |