CN114510873A - Petroleum logging prediction method and device based on big data - Google Patents
Petroleum logging prediction method and device based on big data Download PDFInfo
- Publication number
- CN114510873A CN114510873A CN202210021650.9A CN202210021650A CN114510873A CN 114510873 A CN114510873 A CN 114510873A CN 202210021650 A CN202210021650 A CN 202210021650A CN 114510873 A CN114510873 A CN 114510873A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- logging
- prediction
- petroleum
- 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
- 239000003208 petroleum Substances 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 82
- 238000012544 monitoring process Methods 0.000 claims abstract description 38
- 238000007405 data analysis Methods 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 30
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 84
- 238000003062 neural network model Methods 0.000 claims description 69
- 239000011159 matrix material Substances 0.000 claims description 42
- 238000010801 machine learning Methods 0.000 claims description 41
- 238000004088 simulation Methods 0.000 claims description 21
- 230000008676 import Effects 0.000 claims description 20
- 230000004044 response Effects 0.000 claims description 18
- 239000003129 oil well Substances 0.000 claims description 14
- 239000004215 Carbon black (E152) Substances 0.000 claims description 8
- 229930195733 hydrocarbon Natural products 0.000 claims description 8
- 150000002430 hydrocarbons Chemical class 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 11
- 230000000694 effects Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 244000144972 livestock Species 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
Images
Classifications
-
- 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
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Marine Sciences & Fisheries (AREA)
- Primary Health Care (AREA)
- Mining & Mineral Resources (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a petroleum logging prediction method and a prediction device based on big data, which are applied to a petroleum logging prediction system, and the method comprises the following steps: acquiring petroleum logging big data; creating a prediction model based on the petroleum logging big data; acquiring oil gas monitoring metadata; and processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result. By adopting a distributed data acquisition and data analysis mode, the petroleum logging can be more accurately analyzed according to a large amount of data generated in the historical petroleum logging process, so that a more accurate petroleum logging prediction effect is realized, and the increasing prediction accuracy requirements of people are met; on the other hand, the physical model is combined with the machine model to predict and analyze the petroleum logging data, so that the data analysis and processing speed can be effectively improved, the prediction efficiency is improved, and meanwhile, the prediction accuracy can be effectively improved based on the training and analysis of big data.
Description
Technical Field
The invention relates to the technical field of petroleum logging, in particular to a petroleum logging prediction method based on big data and a petroleum logging prediction device based on the big data.
Background
Along with the increasing living demands of human beings, the demands of people on living matters are also increasing, in order to meet the increasing energy demands of people, the well logging technology is also continuously developed and perfected, and the well logging technology undergoes revolutionary reformation for a plurality of times since birth.
With the continuous development of science and technology and the continuous improvement of environmental protection concepts, people put forward higher requirements and more demands on the oil well logging technology. However, in the practical application of the existing oil well logging technology, the skilled person finds that the existing technology has at least the following technical problems:
on the one hand, the vast amount of data acquired and accumulated during the historical logging process is not fully mined and utilized due to the multi-format and multi-scale characteristics.
On the other hand, in the existing well logging technology, single-well logging data are analyzed to analyze the storage characteristics of a reservoir, and then the credibility of the analyzed data is determined and the logging data is further predicted and calculated according to the personal knowledge storage and the actual working experience of technicians, however, the analysis and calculation process based on expert experience has low interpretation precision, and particularly when lithology identification is carried out on a complex composition stratum, the detection error is large, and the exploration and development requirements are difficult to meet.
Disclosure of Invention
In order to overcome the technical problems in the prior art, the embodiment of the invention provides a petroleum logging prediction method based on big data, and by adopting a distributed data acquisition mode and a data analysis processing method based on a prediction model, the petroleum logging data is quickly and accurately analyzed and predicted, and the actual requirements of users are met.
In order to achieve the above object, an embodiment of the present invention provides an oil logging prediction method based on big data, which is applied to an oil logging prediction system, and the method includes: acquiring petroleum logging big data; creating a prediction model based on the petroleum logging big data; acquiring oil gas monitoring metadata; and processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
Preferably, the acquiring petroleum logging big data comprises: determining a data source and a data format corresponding to the data source; acquiring a data analysis tool corresponding to the data format, wherein the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode; generating a data import instruction; and controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and performing data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
Preferably, the creating a prediction model based on the petroleum logging big data comprises: respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data; creating a predictive model based on the physical algorithm model and the machine learning model.
Preferably, the creating a physical algorithm model based on the petroleum logging big data comprises: acquiring an initial neural network model and simulation logging data; calculating and determining characteristic information of the simulated logging data and the petroleum logging big data; and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
Preferably, the training the initial neural network model based on the feature information to generate the physical algorithm model includes: s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model; s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient; s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement; s2341) if yes, taking the intermediate model as the physical algorithm model; s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
Preferably, the method further comprises: determining a first neural network model and a second neural network model based on the physical algorithm model; performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model; performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model; acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder; and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
Preferably, the creating a machine learning model based on the petroleum logging big data comprises: generating a first training data set Xnew based on the petroleum logging big data; acquiring a preset number classification rule tree; generating a machine learning algorithm prediction based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm prediction is characterized in that: yenew ═ predict (tree, Xnew), where yenew is characterized as the prediction; or generating a corresponding data matrix X based on the simulated logging data; acquiring a response matrix Y corresponding to the data matrix X; generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as: ens ═ fixtenselble (X, Y, model, numbers, learners), where numbers are characterized as attribute information for data matrix X and learners are characterized as attribute information for response matrix Y.
Preferably, the method further comprises: generating corresponding simulation data based on the oil and gas monitoring metadata; optimizing the prediction model based on the simulation data to obtain an optimized model; and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
Preferably, the oil logging prediction system comprises a plurality of distributed computing nodes, and the processing of the hydrocarbon monitoring metadata based on the prediction model to generate corresponding oil logging prediction results comprises: generating at least one predictive task based on the hydrocarbon monitoring metadata; determining a matching distributed computing node corresponding to each prediction task; and executing a corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
Correspondingly, the embodiment of the invention also provides a petroleum logging prediction device based on big data, which comprises: the big data unit is used for acquiring petroleum logging big data; the model creating unit is used for creating a prediction model based on the petroleum logging big data; the metadata acquisition unit is used for acquiring oil gas monitoring metadata; and the prediction unit is used for processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
Preferably, the big data unit includes: the data source determining module is used for determining a data source and a data format corresponding to the data source; the analysis tool determining module is used for acquiring a data analysis tool corresponding to the data format, and the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode; the instruction generation module is used for generating a data import instruction; and the data import module is used for controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and performing data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
Preferably, the model creating unit includes: the middle model creating module is used for respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data; a predictive model creation module to create a predictive model based on the physical algorithm model and the machine learning model.
Preferably, the intermediate model creation module comprises a first model creation module configured to: acquiring an initial neural network model and simulation logging data; calculating and determining characteristic information of the simulated logging data and the petroleum logging big data; and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
Preferably, the training the initial neural network model based on the feature information to generate the physical algorithm model includes: s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model; s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient; s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement; s2341) if yes, taking the intermediate model as the physical algorithm model; s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
Preferably, the first model creation module is further configured to: determining a first neural network model and a second neural network model based on the physical algorithm model; performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model; performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model; acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder; and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
Preferably, the intermediate model creation module comprises a second model creation module configured to: generating a first training data set Xnew based on the petroleum logging big data; acquiring a preset number classification rule tree; generating a machine learning algorithm prediction based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm prediction is characterized in that: yenew ═ predict (tree, Xnew), where yenew is characterized as the prediction; or generating a corresponding data matrix X based on the simulated logging data; acquiring a response matrix Y corresponding to the data matrix X; generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as: ens ═ fixtenselble (X, Y, model, numbers, learners), where numbers are characterized as attribute information for data matrix X and learners are characterized as attribute information for response matrix Y.
Preferably, the apparatus further comprises a model optimization unit for: generating corresponding simulation data based on the oil and gas monitoring metadata; optimizing the prediction model based on the simulation data to obtain an optimized model; and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
Preferably, the oil logging prediction system comprises a plurality of distributed computing nodes, the prediction unit being configured to: generating at least one predictive task based on the hydrocarbon monitoring metadata; determining a matching distributed computing node corresponding to each prediction task; and executing a corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
In another aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the method provided by the embodiment of the present invention.
Through the technical scheme provided by the invention, the invention at least has the following technical effects:
by adopting a distributed data acquisition and data analysis mode, the petroleum logging can be more accurately analyzed according to a large amount of data generated in the historical petroleum logging process, so that a more accurate petroleum logging prediction effect is realized, and the increasing prediction accuracy requirements of people are met; on the other hand, the physical model is combined with the machine model to predict and analyze the petroleum logging data, so that the data analysis and processing speed can be effectively improved, the prediction efficiency is improved, and meanwhile, the prediction accuracy can be effectively improved based on the training and analysis of big data.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart of an implementation of a big data based oil well logging prediction method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a specific implementation of the big data acquisition method for petroleum logging based on big data according to the embodiment of the present invention;
FIG. 3 is a flow chart of a specific implementation of generating a physical algorithm model in a big data based oil well logging prediction method according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a big data-based oil logging prediction device provided by an embodiment of the invention.
Detailed Description
The following describes in detail embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
The terms "system" and "network" in embodiments of the present invention may be used interchangeably. The "plurality" means two or more, and in view of this, the "plurality" may also be understood as "at least two" in the embodiments of the present invention. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" generally indicates that the preceding and succeeding related objects are in an "or" relationship, unless otherwise specified. In addition, it should be understood that the terms first, second, etc. in the description of the embodiments of the invention are used for distinguishing between the descriptions and are not intended to indicate or imply relative importance or order to be construed.
Referring to fig. 1, an embodiment of the present invention provides a big data-based oil well logging prediction method applied to an oil well logging prediction system, where the method includes:
s10) acquiring petroleum logging big data;
s20) creating a prediction model based on the petroleum logging big data;
s30) acquiring oil and gas monitoring metadata;
s40) processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
In one possible embodiment, petroleum well log raw data is first acquired. In the prior art, petroleum logging data can be obtained through various ways, for example, large data of petroleum logging can be obtained through ways such as point logging hand-drawing, photoelectric recording pen, drawing and blueprinting, tape recording, and disk recording, so as to form a data base for petroleum logging prediction.
Referring to fig. 2, in an embodiment of the present invention, the acquiring petroleum logging big data includes:
s11) determining a data source and a data format corresponding to the data source;
s12) acquiring a data analysis tool corresponding to the data format, wherein the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode;
s13) generating a data import instruction;
s14) controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and carrying out data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
For example, in one possible embodiment, when the import operation of petroleum logging big data is performed, a data source and a data format corresponding to the data source are determined first, then a data analysis tool corresponding to the data format is obtained, for example, when the data recorded by a photoelectric recording table is imported, the corresponding data is imported into the petroleum logging prediction system in a specific format through the corresponding photoelectric conversion tool, and because the data acquisition modes used in different areas are different, in order to meet the data import requirements of the areas, the data analysis tool accesses the petroleum logging prediction system in a distributed mode and imports the parsed data into the petroleum logging prediction system. After the data analysis tool is determined, a corresponding data import instruction is generated, then the data analysis tool is controlled to acquire original data from a corresponding data source based on the data import instruction, and the original data is subjected to data analysis according to the data format so as to generate corresponding petroleum logging big data.
In the embodiment of the invention, by accessing various types of data analysis tools in a distributed manner, the petroleum logging data of each region can be accurately imported into the petroleum logging prediction system in real time, so that the petroleum logging prediction system can perform accurate prediction operation according to the acquired petroleum logging big data, and the petroleum logging prediction accuracy is improved.
However, after the petroleum logging big data is obtained, if a traditional manual processing mode is adopted for data analysis and processing, the requirement of people on data processing efficiency cannot be met, and meanwhile, the deviation of manual analysis and processing greatly reduces the data processing accuracy. Therefore, in order to solve the above technical problems, in the embodiment of the present invention, the petroleum logging data is automatically analyzed by further creating a prediction model, so that the prediction efficiency and the prediction accuracy can be effectively improved.
However, in the practical application process, due to the specific technical problems of strong diversity, fast data change, large data volume and the like in the technical field of oil logging, the traditional prediction model or the single prediction model cannot be used for quickly and accurately processing the oil logging data and predicting the subsequent oil logging data.
In order to solve the above technical problem, in an embodiment of the present invention, the creating a prediction model based on the petroleum logging big data includes: respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data; creating a predictive model based on the physical algorithm model and the machine learning model.
In the embodiment of the invention, the model based on the physical algorithm is combined with the machine learning model, and the prediction model of the petroleum logging is further generated, so that the petroleum logging can be simultaneously predicted from the aspects of physical characteristics and theoretical data (simulation data), the prediction speed is effectively improved, and the prediction efficiency and the prediction accuracy are improved.
In an embodiment of the present invention, the creating a physical algorithm model based on the petroleum logging big data includes: acquiring an initial neural network model and simulation logging data; calculating and determining characteristic information of the simulated logging data and the petroleum logging big data; and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
Referring to fig. 3, further, in the embodiment of the present invention, the training the initial neural network model based on the feature information to generate the physical algorithm model includes:
s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model;
s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient;
s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement;
s2341) if yes, taking the intermediate model as the physical algorithm model;
s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
In the prior art, various physical algorithms exist, and after comparative analysis, technicians determine that an algorithm model based on a neural network can perform optimal analysis on an oil well logging technology, so in a possible implementation manner, an initial neural network model and simulated well logging data are firstly obtained, for example, the initial neural network model is a universal neural network model, the technicians can preset certain simulated well logging data according to actual oil well logging data, and then characteristic information between the simulated well logging data and oil well logging big data is determined through calculation. For example, in the process of training through the initial neural network model, the simulated logging data and the petroleum logging data are respectively input into the initial neural network model, corresponding characteristic values are determined, and then the initial neural network model is trained according to the characteristic information, so that the physical algorithm model is generated.
For example, the initial neural network model is iteratively trained for a plurality of times through the characteristic information, and finally the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model are determined, and an intermediate model is determined based on the maximum correlation coefficient and the minimum correlation coefficient, at which point the input can be further refined, and obtaining an output result of the intermediate model by using the simulated logging data as input data, it may then be determined whether the deviation between the output and the big oil log data meets a predetermined deviation requirement, such as, in an embodiment of the present invention, the predetermined bias requirement may be a comparison of an error value of the cost function of the intermediate model with a predetermined error value (e.g. 0.001 ohm-meter), when the error value of the cost function is smaller than the preset error value, it may be determined that the deviation of the intermediate model satisfies a preset deviation requirement.
In one embodiment, it is determined that the deviation of the intermediate model does not meet a preset deviation requirement, so that the simulated logging data is adjusted to obtain adjusted data, the adjusted data is used as new input data to further train the intermediate model to obtain a new intermediate model, and then whether the output result of the new intermediate model meets the preset deviation requirement or not is continuously judged, iterative computation is continuously performed until the output result of the intermediate model meets the preset deviation requirement, and the finally determined intermediate model is used as a physical algorithm model.
In the embodiment of the invention, the physical algorithm model based on the neural network model is adopted to predict the petroleum logging data, so that the prediction efficiency and the prediction accuracy can be effectively improved. However, in the practical application process, technicians find that, because the maximum value and the minimum value of the petroleum logging data are often fixed or in the same interval range, a large number (e.g., millions of times) of forward calculations may be required to obtain a relatively accurate physical algorithm model in the process of training by using a single model, and thus, the actual requirements of people cannot be met.
In order to solve the above technical problem, in an embodiment of the present invention, the method further includes: determining a first neural network model and a second neural network model based on the physical algorithm model; performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model; performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model; acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder; and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
In a possible embodiment, after the physical algorithm model is preliminarily determined, a first neural network model and a second neural network model are further determined, for example, two identical physical algorithm models are copied as the first neural network model and the second neural network model, then the first neural network model and the second neural network model are respectively subjected to forward training and reverse training by using the petroleum logging big data, corresponding forward training models and reverse training models are obtained, then a preset encoder and a preset decoder are obtained, a corresponding automatic decoder is generated, and then the forward training models and the reverse training models are subjected to fusion processing by the automatic decoder to realize optimization of the physical algorithm models and generate the optimized physical algorithm models.
In the embodiment of the invention, the forward calculation model and the reverse calculation model are adopted for calculation, and the preliminarily obtained physical algorithm model is optimized, so that the convergence speed of the physical algorithm model in the actual application process can be further improved, the generation speed and the prediction accuracy of the physical algorithm model are effectively improved, and the actual requirements of users are met.
On the other hand, in order to realize further deep application of the logging technology, and realize accurate analysis and prediction of technologies such as lithology classification, logging data restoration, logging stratum parameter inversion, stratum shear wave extraction, virtual logging curves and the like, a machine learning model is further created.
In an embodiment of the present invention, the creating a machine learning model based on the petroleum logging big data comprises: generating a first training data set Xnew based on the petroleum logging big data; acquiring a preset number classification rule tree; generating a machine learning algorithm prediction based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm prediction is characterized in that: yenew ═ predict (tree, Xnew), where yenew is characterized as the prediction; or generating a corresponding data matrix X based on the simulated logging data; acquiring a response matrix Y corresponding to the data matrix X; generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as: ens, where number is characterized as the attribute information of the data matrix X and leaders is characterized as the attribute information of the response matrix Y.
In an embodiment of the present invention, the method further comprises: generating corresponding simulation data based on the oil and gas monitoring metadata; optimizing the prediction model based on the simulation data to obtain an optimized model; and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
In the embodiment of the invention, the simulation data is further generated based on the actually collected oil and gas monitoring metadata, and the generated prediction model is further verified and optimized through the simulation data, so that the prediction accuracy of the prediction model can be further improved.
However, in the practical application process, the data volume of the oil logging data from different data sources (for example, from a global data source) is huge, so if a centralized data processing mode is adopted, the processing load of the oil logging prediction system is greatly increased, the processing cost is increased, and the processing efficiency is reduced.
In order to solve the above technical problem, in an embodiment of the present invention, the oil well logging prediction system includes a plurality of distributed computing nodes, and the processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil well logging prediction result includes: generating at least one predictive task based on the hydrocarbon monitoring metadata; determining a matching distributed computing node corresponding to each prediction task; and executing a corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
In a possible implementation manner, the oil logging prediction system adopts a Hadoop-based distributed data storage and processing system, a plurality of distributed computing nodes are arranged in the system, after oil and gas monitoring metadata are obtained, corresponding at least one prediction task is generated, for example, a corresponding prediction task can be generated according to information such as an acquisition place and acquisition time of the oil and gas monitoring metadata, then a matching distributed computing node corresponding to each prediction task is determined, for example, the corresponding matching distributed computing node can be allocated to each prediction according to the node type of each distributed computing node and the task processing time required by each task, then each matching distributed computing node is controlled to execute the corresponding prediction task based on the prediction model of the livestock, and a corresponding oil logging prediction result is generated.
In the embodiment of the invention, the distributed system is adopted to analyze and process a large amount of collected petroleum logging metadata, so that the system load of the petroleum logging prediction system can be effectively balanced, the situations of data congestion and task queuing are avoided, the computing capacity of each distributed computing node can be effectively utilized, and the prediction efficiency is improved.
The petroleum logging prediction device based on big data provided by the embodiment of the invention is explained in the following with the accompanying drawings.
Referring to fig. 4, based on the same inventive concept, an embodiment of the present invention provides a big data based oil well logging prediction apparatus, including: the big data unit is used for acquiring petroleum logging big data; the model creating unit is used for creating a prediction model based on the petroleum logging big data; the metadata acquisition unit is used for acquiring oil gas monitoring metadata; and the prediction unit is used for processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
In an embodiment of the present invention, the big data unit includes: the data source determining module is used for determining a data source and a data format corresponding to the data source; the analysis tool determining module is used for acquiring a data analysis tool corresponding to the data format, and the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode; the instruction generation module is used for generating a data import instruction; and the data import module is used for controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and performing data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
In an embodiment of the present invention, the model creating unit includes: the middle model creating module is used for respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data; a predictive model creation module to create a predictive model based on the physical algorithm model and the machine learning model.
In an embodiment of the present invention, the intermediate model creating module includes a first model creating module, and the first model creating module is configured to: acquiring an initial neural network model and simulation logging data; calculating and determining characteristic information of the simulated logging data and the petroleum logging big data; and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
In this embodiment of the present invention, the training the initial neural network model based on the feature information to generate the physical algorithm model includes: s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model; s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient; s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement; s2341) if yes, taking the intermediate model as the physical algorithm model; s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
In an embodiment of the present invention, the first model creating module is further configured to: determining a first neural network model and a second neural network model based on the physical algorithm model; performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model; performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model; acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder; and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
In an embodiment of the present invention, the intermediate model creation module includes a second model creation module, and the second model creation module is configured to: generating a first training data set Xnew based on the petroleum logging big data; acquiring a preset number classification rule tree; generating a machine learning algorithm prediction based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm prediction is characterized in that: yenew ═ predict (tree, Xnew), where yenew is characterized as the prediction; or generating a corresponding data matrix X based on the simulated logging data; acquiring a response matrix Y corresponding to the data matrix X; generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as: ens ═ fixtenselble (X, Y, model, numbers, learners), where numbers are characterized as attribute information for data matrix X and learners are characterized as attribute information for response matrix Y.
In an embodiment of the present invention, the apparatus further includes a model optimization unit, and the model optimization unit is configured to: generating corresponding simulation data based on the oil and gas monitoring metadata; optimizing the prediction model based on the simulation data to obtain an optimized model; and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
In an embodiment of the invention, the oil logging prediction system comprises a plurality of distributed computing nodes, the prediction unit is configured to: generating at least one predictive task based on the hydrocarbon monitoring metadata; determining a matching distributed computing node corresponding to each prediction task; and executing a corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
Further, the embodiment of the present invention also provides a computer readable storage medium, on which a computer program is stored, and the program, when executed by a processor, implements the method described in the embodiment of the present invention.
Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solutions of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications all belong to the protection scope of the embodiments of the present invention.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, the embodiments of the present invention do not describe every possible combination.
Those skilled in the art will understand that all or part of the steps in the method according to the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In addition, any combination of various different implementation manners of the embodiments of the present invention is also possible, and the embodiments of the present invention should be considered as disclosed in the embodiments of the present invention as long as the combination does not depart from the spirit of the embodiments of the present invention.
Claims (19)
1. A petroleum logging prediction method based on big data is applied to a petroleum logging prediction system, and is characterized by comprising the following steps:
acquiring petroleum logging big data;
creating a prediction model based on the petroleum logging big data;
acquiring oil gas monitoring metadata;
and processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
2. The method of claim 1, wherein the acquiring petroleum well log big data comprises:
determining a data source and a data format corresponding to the data source;
acquiring a data analysis tool corresponding to the data format, wherein the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode;
generating a data import instruction;
and controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and performing data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
3. The method of claim 1, wherein creating a predictive model based on the petroleum well log big data comprises:
respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data;
creating a predictive model based on the physical algorithm model and the machine learning model.
4. The method of claim 3, wherein creating a physical algorithm model based on the petroleum well log big data comprises:
acquiring an initial neural network model and simulation logging data;
calculating and determining characteristic information of the simulated logging data and the petroleum logging big data;
and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
5. The method of claim 4, wherein training the initial neural network model based on the feature information to generate the physical algorithm model comprises:
s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model;
s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient;
s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement;
s2341) if yes, taking the intermediate model as the physical algorithm model;
s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
6. The method of claim 4, further comprising:
determining a first neural network model and a second neural network model based on the physical algorithm model;
performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model;
performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model;
acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder;
and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
7. The method of claim 3, wherein creating a machine learning model based on the petroleum well log big data comprises:
generating a first training data set Xnew based on the petroleum logging big data;
acquiring a preset number classification rule tree;
generating a machine learning algorithm predict based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm predict is characterized in that:
Ynew=predict(tree,Xnew),
wherein Ynew is characterized as a predicted outcome; or
Generating a corresponding data matrix X based on the simulated logging data;
acquiring a response matrix Y corresponding to the data matrix X;
generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as:
ens=fitensemble(X,Y,model,numberens,learners),
the number is represented as the attribute information of the data matrix X, and the learners is represented as the attribute information of the response matrix Y.
8. The method of claim 1, further comprising:
generating corresponding simulation data based on the oil and gas monitoring metadata;
optimizing the prediction model based on the simulation data to obtain an optimized model;
and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
9. The method of claim 1, wherein the oil logging prediction system comprises a plurality of distributed computing nodes, and wherein processing the hydrocarbon monitoring metadata based on the prediction model to generate corresponding oil logging predictions comprises:
generating at least one predictive task based on the hydrocarbon monitoring metadata;
determining a matching distributed computing node corresponding to each prediction task;
and executing a corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
10. An oil well logging prediction device based on big data, which is applied to an oil well logging prediction system, and is characterized in that the device comprises:
the big data unit is used for acquiring petroleum logging big data;
the model creating unit is used for creating a prediction model based on the petroleum logging big data;
the metadata acquisition unit is used for acquiring oil gas monitoring metadata;
and the prediction unit is used for processing the oil and gas monitoring metadata based on the prediction model to generate a corresponding oil logging prediction result.
11. The apparatus of claim 10, wherein the big data unit comprises:
the data source determining module is used for determining a data source and a data format corresponding to the data source;
the analysis tool determining module is used for acquiring a data analysis tool corresponding to the data format, and the data analysis tool is accessed to the petroleum logging prediction system in a distributed mode;
the instruction generation module is used for generating a data import instruction;
and the data import module is used for controlling the data analysis tool to acquire original data from the data source based on the data import instruction, and performing data analysis on the original data based on the data format to generate corresponding petroleum logging big data.
12. The apparatus of claim 10, wherein the model creation unit comprises:
the middle model creating module is used for respectively creating a physical algorithm model and a machine learning model based on the petroleum logging big data;
a predictive model creation module to create a predictive model based on the physical algorithm model and the machine learning model.
13. The apparatus of claim 12, wherein the intermediate model creation module comprises a first model creation module configured to:
acquiring an initial neural network model and simulation logging data;
calculating and determining characteristic information of the simulated logging data and the petroleum logging big data;
and training the initial neural network model based on the characteristic information to generate the physical algorithm model.
14. The apparatus of claim 13, wherein the training the initial neural network model based on the feature information to generate the physical algorithm model comprises:
s231) training the initial neural network model based on the characteristic information, and determining the maximum correlation coefficient and the minimum correlation coefficient of the initial neural network model;
s232) determining an intermediate model based on the maximum correlation coefficient and the minimum correlation coefficient;
s233) obtaining an output result of the intermediate model based on the simulated logging data, and judging whether the deviation between the output result and the petroleum logging big data meets a preset deviation requirement;
s2341) if yes, taking the intermediate model as the physical algorithm model;
s2342) if not, adjusting the simulated logging data to obtain adjusted data, training the intermediate model based on the adjusted data to obtain a new intermediate model, and continuing to execute the step S233).
15. The apparatus of claim 13, wherein the first model creation module is further configured to:
determining a first neural network model and a second neural network model based on the physical algorithm model;
performing forward training on the first neural network model based on the petroleum logging big data to obtain a forward training model;
performing reverse training on the second neural network model based on the petroleum logging big data to obtain a reverse training model;
acquiring a preset encoder and a preset decoder, and generating a corresponding automatic decoder based on the preset encoder and the preset decoder;
and optimizing the physical algorithm model based on the forward training model, the reverse training model and the automatic decoder to obtain the optimized physical algorithm model.
16. The apparatus of claim 12, wherein the intermediate model creation module comprises a second model creation module configured to:
generating a first training data set Xnew based on the petroleum logging big data;
acquiring a preset number classification rule tree;
generating a machine learning algorithm prediction based on the first training data set Xnew and the preset tree classification rule tree, wherein the machine learning algorithm prediction is characterized in that:
Ynew=predict(tree,Xnew),
wherein Ynew is characterized as a predicted outcome; or
Generating a corresponding data matrix X based on the simulated logging data;
acquiring a response matrix Y corresponding to the data matrix X;
generating ens a machine learning based on the data matrix X and the response matrix Y, the machine learning algorithm ens characterized as:
ens=fitensemble(X,Y,model,numberens,learners),
the number is represented as the attribute information of the data matrix X, and the learners is represented as the attribute information of the response matrix Y.
17. The apparatus of claim 10, further comprising a model optimization unit to:
generating corresponding simulation data based on the oil and gas monitoring metadata;
optimizing the prediction model based on the simulation data to obtain an optimized model;
and processing the oil and gas monitoring metadata based on the optimized model to generate a corresponding oil logging prediction result.
18. The apparatus of claim 10, wherein the oil logging prediction system comprises a plurality of distributed computing nodes, the prediction unit to:
generating at least one predictive task based on the hydrocarbon monitoring metadata;
determining a matching distributed computing node corresponding to each prediction task;
and executing the corresponding prediction task based on the prediction model through the matching distributed computing nodes to generate a corresponding petroleum logging prediction result.
19. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method of any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210021650.9A CN114510873A (en) | 2022-01-10 | 2022-01-10 | Petroleum logging prediction method and device based on big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210021650.9A CN114510873A (en) | 2022-01-10 | 2022-01-10 | Petroleum logging prediction method and device based on big data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114510873A true CN114510873A (en) | 2022-05-17 |
Family
ID=81550403
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210021650.9A Pending CN114510873A (en) | 2022-01-10 | 2022-01-10 | Petroleum logging prediction method and device based on big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114510873A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116244963A (en) * | 2023-03-22 | 2023-06-09 | 东营市睿建石油装备有限公司 | Petroleum logging prediction method and prediction device based on big data |
-
2022
- 2022-01-10 CN CN202210021650.9A patent/CN114510873A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116244963A (en) * | 2023-03-22 | 2023-06-09 | 东营市睿建石油装备有限公司 | Petroleum logging prediction method and prediction device based on big data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111191002B (en) | Neural code searching method and device based on hierarchical embedding | |
GB2577584A (en) | Utilizing a dynamic memory network to track digital dialog states and generate responses | |
CN112187554B (en) | Operation and maintenance system fault positioning method and system based on Monte Carlo tree search | |
US11556785B2 (en) | Generation of expanded training data contributing to machine learning for relationship data | |
US20210279643A1 (en) | Method and system for generating best performing data models for datasets in a computing environment | |
CN114510873A (en) | Petroleum logging prediction method and device based on big data | |
CN112487406B (en) | Network behavior analysis method based on machine learning | |
CN116522912B (en) | Training method, device, medium and equipment for package design language model | |
CN117521814A (en) | Question answering method and device based on multi-modal input and knowledge graph | |
KR20220032861A (en) | Neural architecture search method and attaratus considering performance in hardware | |
CN116719748A (en) | Scene generation method, device and medium of ship system | |
US20220404515A1 (en) | Systems and methods for mapping seismic data to reservoir properties for reservoir modeling | |
CN112884065B (en) | Deep learning model robust boundary evaluation method, device and application based on support vector machine | |
CN117999560A (en) | Hardware-aware progressive training of machine learning models | |
CN115858763A (en) | Urban management event analysis method based on multi-modal data fusion and application thereof | |
CN109829054A (en) | A kind of file classification method and system | |
KR20220097215A (en) | Method for embedding normalization and electronic device using the same | |
JP2022104911A (en) | Embedding normalization method, and electronic device using the same | |
US20230267277A1 (en) | Systems and methods for using document activity logs to train machine-learned models for determining document relevance | |
CN113762061A (en) | Quantitative perception training method and device for neural network and electronic equipment | |
CN114741583A (en) | Information recommendation method and device based on artificial intelligence and electronic equipment | |
CN115859768B (en) | Method and device for predicting work piece finishing time of dynamic assembly workshop | |
CN110837847A (en) | User classification method and device, storage medium and server | |
CN117933259B (en) | Named entity recognition method based on local text information | |
CN112836721B (en) | Image recognition method and device, computer equipment and readable storage medium |
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 |