Method and device for predicting productivity of tight oil reservoir
Through screening and weighting, geological, development and fracturing parameters of tight oil reservoirs are determined, combined with machine learning models, and a high-precision capacity prediction model is established, which solves the problem of lack of systematicity and integrity in the capacity prediction of tight oil wells in the existing technology, and realizes that High precision and suitable capacity prediction effects.
Patent Information
- Application Number
- CN202410329204.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-21
- Publication Date
- 2025-02-07
AI Technical Summary
The prior art lacks systematicity and integrity in the capacity prediction of tight oil wells. The main sensitive parameters are selected with multiple one-sided independence and are not correlated, resulting in the capacity prediction results being not applicable.
By obtaining dynamic and static data of wells in tight reservoirs, screening geological, development and fracturing parameters as main control parameters of production capacity, determining the correlation coefficient and weight of each parameter, combining the long-term and short-term memory and attention weight mechanisms, and using machine learning models to build high Accurate capacity prediction model.
The accuracy of the analysis of main control factors of production capacity and capacity prediction has been improved, and the problem of one-sided and unsystematic capacity evaluation index system in the existing technology has been solved, so as to achieve high accuracy and applicability of tight oil well production capacity prediction.
Smart Images
Figure CN119398199A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of oil and gas field exploration and development, and in particular to a method and device for predicting the productivity of a tight oil reservoir. Background Art
[0002] China has rich tight oil resources with a wide distribution range. At present, China's tight oil exploration has made progress in many basins, demonstrating my country's huge potential in tight oil exploration and development. How to efficiently develop my country's huge tight oil resources is an important issue at present.
[0003] Compared with conventional oil reservoirs, the economic and efficient development of tight oil reservoirs is more challenging. The main reasons for this phenomenon are internal and external factors. The internal factors are the geological characteristics of tight oil itself, including the structural characteristics of the formation, the wettability and driving energy of the reservoir, the viscosity of crude oil, etc., which are generally not easy to intervene directly; the external factors are mainly artificial development countermeasures and technical measures adopted, such as oilfield development methods, well density, oil production and water injection technology, fracturing and acidizing, etc., which are the key to development plan design and parameter adjustment and optimization. In the process of reservoir development, oil well productivity is a key part of the entire development plan, which determines the scale of oilfield construction and economic benefits. There are many studies on the productivity of conventional oil wells, and considerable research results have been achieved in both theory and practical application. However, the analysis of the main controlling factors of tight oil well productivity and the research on productivity prediction methods are still being explored.
[0004] The productivity of tight oil wells is affected by many factors. Existing technologies usually only study one of the factors. There are few studies on the combined application of the three factors, which leads to the lack of systematicness and integrity of the currently established tight oil well productivity evaluation system. Although domestic and foreign scholars have conducted research on the analysis of the main controlling factors of tight oil well productivity and productivity prediction, and have made certain progress, the selection of the main sensitive parameters is one-sided and independent, and has no correlation. The results of the main controlling factor analysis and productivity prediction can only be targeted at specific reservoirs and are not applicable. Therefore, it is urgent to establish an effective method for tight oil reservoir production control factor analysis and productivity prediction to promote the efficient development of tight oil reservoirs. Summary of the invention
[0005] In order to at least partially solve the technical problems existing in the prior art, the inventors have made the present invention, and through a specific implementation method, provide a method and device for predicting the production capacity of a tight oil reservoir, which can improve the accuracy of the analysis of the main controlling factors of production capacity and the production capacity prediction, and meet the development needs of tight oil reservoirs.
[0006] In a first aspect, an embodiment of the present invention provides a method for predicting the productivity of a tight oil reservoir, comprising:
[0007] Inputting the productivity main control parameter data of the well to be predicted in the target tight oil reservoir into the productivity prediction model, and predicting the productivity parameter data of the well to be predicted according to the output result of the productivity prediction model;
[0008] The production capacity master control parameters and the production capacity prediction model are obtained through the following steps:
[0009] Obtaining dynamic and static data of wells in tight oil reservoirs, obtaining data of multiple candidate production capacity master control parameters and production capacity parameters of the wells, determining the correlation coefficient between each candidate production capacity master control parameter and the production capacity parameter, and screening the production capacity master control parameter, wherein the multiple candidate production capacity master control parameters include geological, development and fracturing parameters;
[0010] The weight of the impact of each production capacity master control parameter on the production capacity parameter is determined, and input into the selected machine learning model. The production capacity prediction model is obtained based on the learning of the production capacity master control parameters and production capacity parameter data of the well by combining long-term and short-term memory and attention weight mechanisms.
[0011] In a second aspect, an embodiment of the present invention provides a tight oil reservoir productivity prediction device, the device comprising a productivity prediction module and a productivity prediction model establishment module;
[0012] The production capacity prediction module is used to input the production capacity main control parameter data of the well to be predicted in the target tight oil reservoir into the production capacity prediction model established by the production capacity prediction model establishment module, and predict the production capacity parameter data of the well to be predicted according to the output result of the production capacity prediction model;
[0013] The capacity prediction model establishment module is used to obtain dynamic and static data of wells in tight oil reservoirs, obtain data on multiple candidate capacity control parameters and capacity parameters of the wells, determine the correlation coefficient between each candidate capacity control parameter and the capacity parameter, and screen the capacity control parameters. The multiple candidate capacity control parameters include geological, development and fracturing parameters; determine the weight of the influence of each capacity control parameter on the capacity parameter, and input it into the selected machine learning model, adopt a combination of long-term and short-term memory and attention weight mechanism, and learn the capacity control parameters and capacity parameter data of the well to obtain a capacity prediction model.
[0014] In a third aspect, an embodiment of the present invention provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions, when executed by a processor, implement the above-mentioned method for predicting the productivity of tight oil reservoirs.
[0015] In a fourth aspect, an embodiment of the present disclosure provides a server, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for predicting the production capacity of tight oil reservoirs when executing the program.
[0016] The beneficial effects of the above technical solution provided by the embodiment of the present invention include at least:
[0017] The method for predicting the capacity of tight oil reservoirs provided in the embodiment of the present invention uses geological, development and fracturing parameters as candidate capacity main control parameters, screens the capacity main control parameters through correlation analysis, and determines the weight of each main control parameter, thereby solving the problem that the capacity evaluation index system in the prior art solution is one-sided and unsystematic, expanding and improving the technical means of the main control factors of the capacity of tight oil reservoirs, and being able to meet the actual production needs on site; based on the analysis results of the capacity main control parameters, a high-precision capacity prediction model is established with actual samples through artificial intelligence technology to realize the capacity prediction of tight oil wells, thereby avoiding the shortcomings of poor applicability and low precision of the capacity formula in the prior art solution, and the predicted capacity of tight oil wells is highly consistent with the actual data, which can meet the needs of the site.
[0018] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.
[0019] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0021] Figure 1 This is a flow chart of a method for predicting the productivity of a tight oil reservoir in Embodiment 1 of the present invention;
[0022] Figure 2 This is a specific implementation flow chart of the method for predicting the productivity of a tight oil reservoir in Embodiment 2 of the present invention;
[0023] Figure 3 This is a diagram of the production capacity evaluation index system of tight oil reservoirs in the xx region in the second embodiment of the present invention;
[0024] Figure 4 This is a weight ranking bar chart of tight oil reservoir production control factors in the xx region in the second embodiment of the present invention;
[0025] Figure 5 This is a comparison chart of the accuracy of various models of tight oil reservoir production capacity in the xx area in the second embodiment of the present invention;
[0026] Figure 6 This is a comparison chart of the predicted and actual production capacity of tight oil reservoirs in the xx region in the second embodiment of the present invention;
[0027] Figure 7 It is a schematic diagram of the structure of a tight oil reservoir productivity prediction device in an embodiment of the present invention. DETAILED DESCRIPTION
[0028] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0029] It should be understood that the terms described in the present invention are only for describing special embodiments and are not intended to limit the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which the present invention belongs. Although the present invention only describes preferred methods and materials, any methods and materials similar or equivalent to those described herein may also be used in the implementation or testing of the present invention. All documents mentioned in this specification are incorporated by reference to disclose and describe methods and / or materials related to the documents. In the event of conflict with any incorporated document, the content of this specification shall prevail.
[0030] Embodiment 1
[0031] Embodiment 1 of the present invention provides a method for establishing a tight oil reservoir productivity prediction model, and the process thereof is as follows Figure 1 As shown, the following steps are included:
[0032] Step S11: acquiring dynamic and static data of wells in tight oil reservoirs, obtaining data of multiple candidate productivity master control parameters and productivity parameters of the wells, determining the correlation coefficients between each candidate productivity master control parameter and the productivity parameter, and screening the productivity master control parameter.
[0033] The multiple candidate capacity control parameters include geological, development and fracturing parameters. The parameter characterizing the capacity can be one parameter or multiple parameters. For example, the actual one-year capacity, the one-year capacity after 50% water content and the peak capacity can be used as capacity parameters.
[0034] Specifically, the determination of the candidate capacity master control parameters may include:
[0035] (1) For the study of tight oil reservoirs, various static and dynamic data such as seismic data, geological results, well logging interpretation, perforation data, and production dynamics are collected to achieve classified input, classified database construction, and classified management of various types of data.
[0036] (2) Filter and remove redundant data, completely supplement missing data, correct or delete erroneous data, and finally organize the data into data that can be further processed and used.
[0037] (3) Standardize the data range, dimension, format, and type to facilitate subsequent calculations.
[0038] (4) Select various factors related to production capacity based on professional experience, make judgments through logical analysis, and preliminarily screen out candidate production capacity master control parameters.
[0039] The logical analysis method mainly analyzes the internal connection of the influencing factors in the reservoir development process based on logical relationships such as cause and effect, equivalence, definition, and process, removes the influencing factors with less information, and retains the influencing factors with more information, and preliminarily determines the candidate production capacity master control parameters.
[0040] Continue to eliminate repetitive parameters of the initially determined candidate capacity master control parameters through correlation analysis, so that the final candidate capacity master control parameters meet the requirement that the correlation coefficients between each pair are lower than the set correlation coefficient threshold.
[0041] After the final candidate capacity control parameters are screened out, each parameter is normalized to eliminate the influence of units and orders of magnitude.
[0042] An efficient, scientific and accurate data management database is established based on existing seismic, geological, well logging interpretation, production dynamics and other data. It can efficiently screen and remove redundant data, completely supplement lost data, correct or delete erroneous data, and solve a series of pre-processing problems such as data standardization. It greatly improves the quality of field data, reduces data dimensions, mines data value, greatly improves data processing efficiency, and reduces time costs.
[0043] In some embodiments, the correlation coefficient between each candidate capacity master control parameter and the capacity parameter may be determined using Pearson, Spearman or Kendall correlation coefficient determination method.
[0044] Optionally, the Pearson, Spearman and Kendall correlation coefficient determination methods may be used to determine the correlation coefficient between each candidate capacity master control parameter and the capacity parameter by weighted averaging.
[0045] Among them, the formula for determining the Pearson correlation coefficient is:
[0046]
[0047] In formula (1): X，Y is the correlation coefficient between influencing factor X and influencing factor Y. The correlation coefficient varies from -1 (negative correlation) to 1 (positive correlation). When the correlation coefficient is 0, it means that there is no correlation between the two factors; cov(X, Y) is the covariance; σ is the standard deviation.
[0048] The formula for determining the Spearman correlation coefficient is:
[0049]
[0050] In formula (2), p is the Spearman correlation coefficient, di represents the difference in the order of influencing factors, and n represents the number of influencing factors.
[0051] The formula for determining the Kendall correlation coefficient is:
[0052]
[0053] In formula (3), T au-a is the Kendall correlation coefficient, c represents the consistent logarithm, d represents the consistent logarithm, 1 / 2n(n-1) represents the number of combinations of all influencing factors. When there are no repeated values, the number of combinations is equal to c+d.
[0054] In some embodiments, the selected capacity control parameters are as follows:
[0055] The geological parameters include formation pressure, porosity, permeability, saturation, brittleness index, vertically developed oil layer thickness, oil layer drilling rate and the thickness ratio of Class I and Class II oil reservoirs; the development parameters include well spacing and horizontal section length, and the fracturing parameters include segment spacing, number of segments, arrowhead spacing, number of arrowheads, fracturing fluid volume and sand addition volume.
[0056] Step S12: Determine the weight of the impact of each production capacity master control parameter on the production capacity parameter, and input it into the selected machine learning model, adopt a combination of long-term and short-term memory and attention weight mechanism, and learn the production capacity master control parameters and production capacity parameter data of the well to obtain a production capacity prediction model.
[0057] In some embodiments, the coefficient of variation method, entropy method or CRITIC method may be used to determine the weight of the impact of each capacity master control parameter on the capacity parameter.
[0058] Optionally, the coefficient of variation method, entropy method and CRITIC method may be used to determine the weight of the influence of each capacity master control parameter on the capacity parameter through weighted averaging.
[0059] A. Coefficient of variation method:
[0060]
[0061] In formula (4), V i is the coefficient of variation of the i-th factor, W i is the weight of the i-th indicator determined by the coefficient of variation method.
[0062] B. Entropy method:
[0063]
[0064] In formula (5), gj is the coefficient of variation of the i-th factor, W j is the entropy weight of the i-th indicator determined by the coefficient of variation.
[0065] C.CRITIC method:
[0066]
[0067]
[0068] In formulas (6) and (7), W j is the weight of the influencing factor, V j is the coefficient of variation, R j is the degree of independence coefficient, C j To evaluate the comprehensive coefficient of influencing factors, W j is the weight of the influencing factors.
[0069] In some embodiments, the selected machine learning model may be a K-nearest neighbor learning model, a decision tree learning model, a random forest learning model, a Bayesian learning model, or a neural network MLP learning model.
[0070] Based on the inventive concept of the present invention, an embodiment of the present invention further provides a method for predicting the productivity of a tight oil reservoir, comprising:
[0071] The production capacity main control parameter data of the well to be predicted in the target tight oil reservoir is input into the production capacity prediction model, and the production capacity parameter data of the well to be predicted is predicted according to the output result of the production capacity prediction model.
[0072] The production capacity prediction model is a model established by the above method. The data source on which the model is established - tight oil reservoir, can be the same reservoir as the target tight oil reservoir, or a reservoir with a similar location and similar characteristics.
[0073] The method for predicting the capacity of tight oil reservoirs provided in the embodiment of the present invention uses geological, development and fracturing parameters as candidate capacity main control parameters, screens the capacity main control parameters through correlation analysis, and determines the weight of each main control parameter, thereby solving the problem that the capacity evaluation index system in the prior art solution is one-sided and unsystematic, expanding and improving the technical means of the main control factors of the capacity of tight oil reservoirs, and being able to meet the actual production needs on site; based on the analysis results of the capacity main control parameters, a high-precision capacity prediction model is established with actual samples through artificial intelligence technology to realize the capacity prediction of tight oil wells, thereby avoiding the shortcomings of poor applicability and low precision of the capacity formula in the prior art solution, and the predicted capacity of tight oil wells is highly consistent with the actual data, which can meet the needs of the site.
[0074] Embodiment 2
[0075] Embodiment 2 of the present invention provides a specific application of a method for predicting the production capacity of a tight oil reservoir. The xx study area is located in the northwest of the xx basin, with an exploration area of 7,300 square kilometers and an estimated petroleum geological resource of 4.6 billion tons. It is a rare oil-rich zone in China and a typical tight conglomerate reservoir. Fracturing horizontal wells are its main development technology. Since the block was put into production in 2016, it has exposed problems such as large differences in the development effect of fracturing horizontal wells, unclear main controlling factors leading to production capacity differences, and difficulty in predicting the production capacity of new wells.
[0076] The method provided by the embodiment of the present invention can quickly and accurately analyze the production control factors and predict the production capacity of the fractured horizontal wells in the study area, thereby providing a scientific basis for the efficient development of the block. Figure 2 As shown, the implementation steps of this embodiment are as follows:
[0077] Step S21: For the horizontal wells in the tight oil reservoirs in the study area that have been put into production and fractured, data are collected from three aspects: geology, development and fracturing parameters, and production capacity index data are collected at the same time.
[0078] There are three capacity indicators (actual one-year capacity, one-year capacity after 50% water content, and peak output), which are classified for input, database creation, and management.
[0079] Step S22: Screen and remove redundant data, supplement missing data, correct or delete erroneous data, and organize them into oil well productivity sample data that can be further processed and used.
[0080] Combined with the data actually collected in this block, it is necessary to properly clean and integrate the basic sample data to establish the original sample data of the factors affecting the productivity of fractured horizontal wells in this block.
[0081] Step S23: Standardize the data to eliminate the dimensional influence between different feature data, making it easier to perform subsequent calculations.
[0082] Sixteen parameters that may affect production capacity were collected, as shown in Table 1. The units of each parameter are meter, Darcy, MPa, cubic meter, etc. The Min-Max method was used for standardization to eliminate the impact of dimension.
[0083] Table 1 Parameters that may affect production capacity and their dimension statistics
[0084] Geological factors Development parameters Fracturing parameters Formation pressure, Mpa Well spacing, m Segment spacing, m Porosity, % Horizontal section length, m Segment number, level Permeability, mD Cluster spacing, m Saturation, % Number of clusters Brittleness Index Fracturing fluid volume, square Vertical producing oil layer thickness, m Add sand amount, square Oil layer drilling rate, % Thickness ratio of the first and second oil layers, %
[0085] Step S24: preliminarily screen out candidate capacity master control parameters through logical analysis.
[0086] There are many factors that affect production capacity. After analysis and judgment by reservoir engineers, eight geological factors were initially determined: formation pressure, porosity, permeability, saturation, brittleness index, vertical oil layer thickness, oil layer drilling rate, and the thickness ratio of Class I and II oil reservoirs; two development parameters: well spacing and horizontal section length; and six fracturing parameters: section spacing, number of sections, arrowhead spacing, number of arrowheads, fracturing fluid volume, and sand addition volume. Therefore, after deletion and integration, a set of tight oil reservoir production capacity evaluation index system in the xx region was finally established based on the above 16 influencing parameters, such as Figure 3 shown.
[0087] Step S25: Extract candidate master control parameter data according to the sample data of the research block, perform correlation analysis, calculate the Pearson, Spearman, and Kendall correlation coefficients respectively, and screen out the master control parameters by comparing the correlation coefficients.
[0088] Step S26: Use objective data analysis methods (coefficient of variation method, entropy method, random forest method) to determine the weights of each main control parameter.
[0089] The calculation weight results of the three algorithms can be sorted according to the results of any algorithm, or sorted according to the average value of the three algorithm weights to determine the primary and secondary relationship of the main control parameters, such as Figure 4 shown.
[0090] Step S27: Using deep learning technology, using the acquired production capacity sample data, and using deep learning algorithms such as multi-layer neural networks for training, a high-precision prediction model is obtained.
[0091] At the same time, four commonly used machine learning models (K nearest neighbor, decision tree, Bayesian, etc.) are used to build models, and the optimal prediction model is obtained by comparing the model accuracy.
[0092] The ratio of training set to test set is 0.9:0.1, and 5-fold cross validation is used to prevent overfitting. Among the four machine learning models, the training set accuracy is between 0.62-0.97, among which the random forest training set accuracy is the highest, and the test set accuracy is between 0.50-0.64. Deep learning-multi-layer neural network, set 50-200 neurons, 3-9 layers of neural network, the training set accuracy gradually improves (0.99), and the highest test set accuracy is 0.71, such as Figure 5 .
[0093] Step S28: Use the five trained prediction models to predict the productivity of new wells in tight oil reservoirs.
[0094] The new well data uses the parameter data of 7 new wells put into production in the past three years as the basic data. After data preprocessing, the prediction model is imported to predict the production capacity of new wells. The prediction results show that the deep learning-multi-layer neural network model has the highest prediction consistency rate (close to 86%), and the consistency rates of other prediction models are less than 80%, as shown in Table 2:
[0095] Table 2 Statistics of prediction results of each model
[0096]
[0097] like Figure 6 As shown, the capacity prediction result of this embodiment is basically consistent with the actual situation, with a small relative error and high prediction accuracy. Therefore, the technical solution of this embodiment can meet the production demand of tight oil reservoir capacity prediction.
[0098] Based on the inventive concept of the present invention, an embodiment of the present invention further provides a device for predicting the productivity of a tight oil reservoir. The structure of the device is as follows: Figure 7 As shown, it includes a capacity prediction module 71 and a capacity prediction model building module 72;
[0099] The capacity prediction module 71 is used to input the capacity master parameter data of the well to be predicted in the target tight oil reservoir into the capacity prediction model established by the capacity prediction model establishment module 72, and predict the capacity parameter data of the well to be predicted according to the output result of the capacity prediction model;
[0100] The capacity prediction model establishment module 72 is used to obtain dynamic and static data of wells in tight oil reservoirs, obtain data on multiple candidate capacity control parameters and capacity parameters of the wells, determine the correlation coefficient between each candidate capacity control parameter and the capacity parameter, and screen the capacity control parameters. The multiple candidate capacity control parameters include geological, development and fracturing parameters; determine the weight of the influence of each capacity control parameter on the capacity parameter, and input it into the selected machine learning model, adopt a combination of long-term and short-term memory and attention weight mechanism, and learn the capacity control parameters and capacity parameter data of the well to obtain a capacity prediction model.
[0101] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0102] Based on the inventive concept of the present invention, an embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and when the computer executable instructions are executed by a processor, the above-mentioned method for predicting the productivity of tight oil reservoirs is implemented.
[0103] Based on the inventive concept of the present invention, an embodiment of the present invention further provides a server, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for predicting the production capacity of tight oil reservoirs when executing the program.
[0104] Unless otherwise specifically stated, terms such as processing, computing, calculating, determining, displaying, etc. may refer to the actions and / or processes of one or more processing or computing systems, or similar devices, which operate and convert data represented as physical (e.g., electronic) quantities within registers or memories of a processing system into other data similarly represented as physical quantities within memories, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals may be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.
[0105] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of protection of the present disclosure. The attached method claims present the elements of the various steps in an exemplary order and are not intended to be limited to the specific order or hierarchy described.
[0106] In the above detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that an embodiment of the claimed subject matter requires more features than those set forth in each claim. On the contrary, as reflected in the appended claims, the invention is in a state of having less than all the features of the disclosed individual embodiments. Therefore, the appended claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.
[0107] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments herein can all be implemented as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the interchangeability between hardware and software, various illustrative components, blocks, modules, circuits, and steps are generally described above around their functions. Whether such functions are implemented as hardware or software depends on specific applications and the design constraints imposed on the entire system. A skilled person can implement the described functions in an alternative manner for each specific application, but such implementation decisions should not be interpreted as departing from the scope of protection of the present disclosure.
[0108] The steps of the method or algorithm described in conjunction with the embodiments herein may be directly embodied as hardware, a software module executed by a processor, or a combination thereof. The software module may be located in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a mobile disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The ASIC may be located in a user terminal. Of course, the processor and the storage medium may also be present in a user terminal as discrete components.
[0109] For software implementation, the techniques described in this application can be implemented with modules (e.g., procedures, functions, etc.) that perform the functions described in this application. These software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented within the processor or outside the processor. In the latter case, it is coupled to the processor in a communication manner via various means, which are well known in the art.
[0110] The above description includes examples of one or more embodiments. Of course, it is impossible to describe all possible combinations of components or methods for the purpose of describing the above embodiments, but it should be recognized by those skilled in the art that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications and variations that fall within the scope of protection of the appended claims. In addition, with respect to the term "comprising" used in the specification or claims, the word is covered in a manner similar to the term "including", just as "including," is explained as a transitional word in the claims. In addition, any term "or" used in the specification of the claims is intended to mean "non-exclusive or".
Claims
1. A method for predicting the productivity of a tight oil reservoir, characterized in that: include: Inputting the productivity main control parameter data of the well to be predicted in the target tight oil reservoir into the productivity prediction model, and predicting the productivity parameter data of the well to be predicted according to the output result of the productivity prediction model; The production capacity master control parameters and the production capacity prediction model are obtained through the following steps: Obtaining dynamic and static data of wells in tight oil reservoirs, obtaining data of multiple candidate production capacity master control parameters and production capacity parameters of the wells, determining the correlation coefficient between each candidate production capacity master control parameter and the production capacity parameter, and screening the production capacity master control parameter, wherein the multiple candidate production capacity master control parameters include geological, development and fracturing parameters; The weight of the impact of each production capacity master control parameter on the production capacity parameter is determined, and input into the selected machine learning model. The production capacity prediction model is obtained based on the learning of the production capacity master control parameters and production capacity parameter data of the well by combining long-term and short-term memory and attention weight mechanisms.
2. The method according to claim 1, characterized in that Before determining the correlation coefficient between each candidate capacity master control parameter and the capacity parameter, the method further includes: Through correlation analysis, repeated parameters of candidate capacity master control parameters are eliminated, so that the correlation coefficients between the final candidate capacity master control parameters are lower than the set correlation coefficient threshold.
3. The method according to claim 1, characterized in that The determining of the correlation coefficient between each candidate capacity master control parameter and the capacity parameter includes: Determine the correlation coefficient between each candidate capacity master control parameter and the capacity parameter using Pearson, Spearman or Kendall correlation coefficient determination method; or, The correlation coefficients of the candidate production capacity master control parameters and the production capacity parameters are determined by weighted averaging using the Pearson, Spearman and Kendall correlation coefficient determination methods.
4. The method according to claim 1, characterized in that: Determining the weight of each capacity master control parameter's influence on the capacity parameter includes: Determine the weight of each capacity master control parameter on the capacity parameter by using the coefficient of variation method, entropy method or CRITIC method; or, The coefficient of variation method, entropy method and CRITIC method are used to determine the weight of the impact of each capacity master control parameter on the capacity parameter through weighted average.
5. The method according to claim 1, characterized in that The selected machine learning model is a K nearest neighbor learning model, a decision tree learning model, a random forest learning model, a Bayesian learning model or a neural network MLP learning model.
6. The method according to claim 1, characterized in that The screening capacity master control parameters include: The main control parameters for screening capacity are as follows: The geological parameters include formation pressure, porosity, permeability, saturation, brittleness index, vertically developed oil layer thickness, oil layer drilling rate and the thickness ratio of Class I and Class II oil reservoirs; the development parameters include well spacing and horizontal section length, and the fracturing parameters include segment spacing, number of segments, arrowhead spacing, number of arrowheads, fracturing fluid volume and sand addition volume.
7. The method according to claim 1, characterized in that The capacity parameter includes at least one of the following parameters: One-year capacity, one-year capacity after 50% water content and peak capacity.
8. A device for predicting the productivity of a tight oil reservoir, characterized in that: The device includes a capacity prediction module and a capacity prediction model establishment module; The production capacity prediction module is used to input the production capacity main control parameter data of the well to be predicted in the target tight oil reservoir into the production capacity prediction model established by the production capacity prediction model establishment module, and predict the production capacity parameter data of the well to be predicted according to the output result of the production capacity prediction model; The production capacity prediction model establishment module is used to obtain dynamic and static data of wells in tight oil reservoirs, obtain data on multiple candidate production capacity main control parameters and production capacity parameters of the wells, determine the correlation coefficient between each candidate production capacity main control parameter and the production capacity parameter, and screen the production capacity main control parameters. The multiple candidate production capacity main control parameters include geological, development and fracturing parameters; determine the weight of the influence of each production capacity main control parameter on the production capacity parameter, and input it into the selected machine learning model, adopt a combination of long-term and short-term memory and attention weight mechanism, and learn the production capacity main control parameters and production capacity parameter data of the well to obtain a production capacity prediction model.
9. A computer storage medium, characterized in that The computer storage medium stores computer executable instructions, and when the computer executable instructions are executed by a processor, the method for predicting the productivity of a tight oil reservoir according to any one of claims 1 to 7 is implemented.
10. A server, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for predicting the productivity of a tight oil reservoir as claimed in any one of claims 1 to 7 is implemented.