System determination method based on LightGBM and RIME algorithms
By training gas well production models using the LightGBM and RIME algorithms, the optimal production regime was determined, which solved the problem of the lack of scientific basis in existing methods and improved gas well production and management efficiency.
Patent Information
- Application Number
- CN202411113293.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2026-03-03
AI Technical Summary
The existing methods for determining gas well production systems lack scientific basis, leading to errors and instability, and failing to effectively improve gas well recovery rates.
The LightGBM and RIME algorithms are used to preprocess production data, train an inter-row production prediction model, and find the optimal production system with the objective function of maximizing theoretical daily gas production.
It increased gas well production, reduced manual labor, achieved adaptive optimal production system switching, and improved gas well management efficiency and recovery rate.
Smart Images
Figure CN121599166A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of production optimization applications in the oil and gas industry, and specifically relates to a method, system, electronic device and storage medium for determining the system based on LightGBM and RIME algorithms. Background Technology
[0002] In the later stages of gas well production, due to capacity decay, bottomhole pressure increases, gas production decreases, and the well's fluid-carrying capacity becomes poor, leading to a continuous increase in fluid accumulation in the wellbore. This severely affects the normal production of the gas well, and some wells even experience production stoppage due to fluid accumulation. Continuous production of gas wells can no longer achieve the goal of continuous fluid carrying, therefore, intermittent production is necessary for gas wells with low production. Intermittent gas well production refers to a production method in which the gas well is shut down for a period of time to allow the gas well's energy to be replenished and restored, and then the well is reopened to carry out the accumulated fluid in the wellbore, thereby restoring the gas well's production capacity and improving the gas well recovery rate. Currently, most intermittent gas well production systems are formulated based on the experience of production personnel and historical data, combined with conventional production indicators and conditions.
[0003] Therefore, the existing methods for determining the system are simple and intuitive, and applicable to some routine situations, but they lack scientific basis and are prone to errors and instability. Summary of the Invention
[0004] To address the aforementioned problems in existing technologies—namely, while existing methods for determining regulations are simple and intuitive, applicable to some routine situations, they lack scientific basis and are prone to errors and instability—this invention provides a method for determining regulations based on the LightGBM and RIME algorithms. The method includes:
[0005] Production data from different production wells in different regions is obtained, and the production data is preprocessed to obtain preprocessed data.
[0006] The preprocessed data of each production well is divided into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well.
[0007] Different production regimes are input into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to find the optimal production regime with the maximization of theoretical daily gas production as the single objective function.
[0008] The production regime corresponding to the maximum theoretical daily gas production of each production well is the optimal production regime for that well.
[0009] In a preferred embodiment, the production data includes: well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, pipeline pressure, daily gas production, and daily fluid discharge.
[0010] In a preferred embodiment, the preprocessed data of each production well is divided into feature data and label data, specifically including:
[0011] The well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, and pipeline pressure of each production well are classified as characteristic data of that production well.
[0012] The daily gas production and daily liquid discharge of each production well are classified as the tag data for that production well.
[0013] Preprocessing the production data includes:
[0014] Empty sample values in the production data are filled by using the nearest values of the sample values.
[0015] Remove negative values from the production data;
[0016] Use box plots to remove abnormal data from the production data.
[0017] In a preferred embodiment, the single objective function is:
[0018] Where Obj is the theoretical daily gas production, m is the number of sampling points during well opening time; q i F represents the instantaneous gas production during the well opening time. s T1 is the sampling frequency; T2 is the well opening time; T2 is the well closing time.
[0019] In a preferred embodiment, using the RIME algorithm to find the optimal production regime includes:
[0020] For each intermittent production prediction model, different production regimes of the current production well are input into the intermittent production prediction model to predict the theoretical daily gas production of the current production well.
[0021] With the goal of maximizing the theoretical daily gas production as the single objective function and production time and number of intervals as constraints, the RIME algorithm is used to continuously update the production regime until the preset number of iterations is reached.
[0022] The production regime corresponding to the theoretical maximum daily gas production is the optimal production regime for the current production well.
[0023] A second aspect of the present invention provides a system for determining the regime of LightGBM and RIME algorithms, the system comprising:
[0024] The data acquisition module acquires production data from different production wells in different regions and preprocesses the production data to obtain preprocessed data.
[0025] The model training module is used to divide the preprocessed data of each production well into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well.
[0026] The system optimization module is used to input different production systems into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to optimize the optimal production system with the maximization of theoretical daily gas production as the single objective function.
[0027] The system is a module used to determine the optimal production system. The production system corresponding to the maximum theoretical daily gas production of each production well is the optimal production system for that well.
[0028] A third aspect of the present invention provides an electronic device comprising:
[0029] At least one processor; and
[0030] A memory communicatively connected to at least one of the processors; wherein,
[0031] The memory stores instructions that can be executed by the processor to implement the aforementioned method for determining the regime based on the LightGBM and RIME algorithms.
[0032] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer instructions for execution by a computer to implement the above-described method for determining the regime based on the LightGBM and RIME algorithms.
[0033] The beneficial effects of this invention are:
[0034] (1) Based on historical production dynamic data such as wellhead tubing pressure, casing pressure, pipeline pressure, instantaneous gas production, and liquid discharge of intermittent gas wells, this application uses the LightGBM_RIME algorithm to model the historical production dynamic data and customizes a method for determining the optimal production system suitable for intermittent wells.
[0035] (2) The method of this application can increase the production of intermittent gas wells while adaptively switching to the optimal production system, reducing manual workload and improving the management efficiency of intermittent gas wells.
[0036] (3) This application is of great guiding significance for the formulation and adjustment of intermittent production schemes for gas wells in the later stage of oil and gas field development, and can achieve the ultimate goal of improving recovery rate and saving energy and reducing consumption. Attached Figure Description
[0037] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0038] Figure 1 This is a flowchart of the method for determining the optimal regime for gas wells in an embodiment of the present invention;
[0039] Figure 2 This is a flowchart of a method for determining the system based on the LightGBM and RIME algorithms according to an embodiment of the present invention;
[0040] Figure 3 This is a schematic diagram of an intermittent production prediction model (NPM) generation scheme according to an embodiment of the invention;
[0041] Figure 4 This is a flowchart illustrating the generation process of an optimized generator according to an embodiment of the present invention;
[0042] Figure 5 This is a schematic diagram of a histogram algorithm diagram according to an embodiment of the present invention;
[0043] Figure 6 This is a diagram of a depth-limited leaf-wise leaf growth strategy according to an embodiment of the present invention;
[0044] Figure 7 This is a diagram of an EFB algorithm according to an embodiment of the present invention;
[0045] Figure 8 This is a box plot according to an embodiment of the present invention;
[0046] Figure 9 This is a schematic diagram of the structure of a computer system used to implement the methods, systems, and devices of this application. Detailed Implementation
[0047] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.
[0048] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.
[0049] This invention provides a method for determining the regime based on LightGBM and RIME algorithms, the method comprising:
[0050] Production data from different production wells in different regions is obtained, and the production data is preprocessed to obtain preprocessed data.
[0051] The preprocessed data of each production well is divided into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well.
[0052] Different production regimes are input into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to find the optimal production regime with the maximization of theoretical daily gas production as the single objective function.
[0053] The production regime corresponding to the maximum theoretical daily gas production of each production well is the optimal production regime for that well.
[0054] like Figure 1 As shown, the first part of this application uses the LightGBM algorithm to learn the relationship between different well opening and closing regimes and gas production, obtaining an intermittent well production prediction model (NPM). The second part of this application uses the intermittent well production prediction model (NPM) trained in the first part, combined with the RIME frost optimization algorithm, to obtain an intermittent well production optimization generator (AG). Given different well opening and closing regimes, it can quickly generate the optimal intermittent well opening and closing regime scheme. Here, the optimal intermittent well opening and closing regime scheme refers to the intermittent well opening and closing time scheme that maximizes gas production while satisfying the historical well opening and closing time range and the constraint of the number of well opening and closing times.
[0055] To more clearly explain the method for determining the regime based on the LightGBM and RIME algorithms of this invention, the following will be combined with... Figure 2 The steps in the embodiments of the present invention will be described in detail below.
[0056] The method for determining the system based on the LightGBM and RIME algorithms in the first embodiment of the present invention is described in detail below:
[0057] Production data from different production wells in different regions is obtained, and the production data is preprocessed to obtain preprocessed data.
[0058] In this embodiment, the production data includes: well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, pipeline pressure, daily gas production, and daily fluid discharge.
[0059] In this embodiment, the preprocessing of the production data includes: filling empty sample values in the production data by using neighboring values; removing negative data; and deleting abnormal data using a box plot method. The specific method for deletion using the box plot method is as follows: for characteristic data such as well depth, tubing inner diameter, casing inner diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, pipeline pressure, daily gas production, and daily fluid discharge, identify the upper edge, lower edge, median, and two quartiles for each characteristic data; connect the two quartiles to draw a box; then connect the upper and lower edges to the box, with the median in the middle of the box; data outside the upper and lower edges are considered abnormal data and are deleted.
[0060] The preprocessed data of each production well is divided into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well.
[0061] In this embodiment, the preprocessed data of each production well is divided into feature data and tag data. Specifically, the well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, and pipeline pressure of each production well are divided into feature data of that production well; and the daily gas production and daily liquid discharge of each production well are divided into tag data of that production well.
[0062] In this embodiment, the intermittent production prediction model (NPM) is trained. Since there are 50,000 existing data points, based on model training experience, the feature data and training data are split into training and test sets in a 7:3 ratio, respectively. The data are then input into the LightGBM algorithm to train production models for different wells and different production regimes, thus obtaining the intermittent production prediction model (NPM).
[0063] Different production regimes are input into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to find the optimal production regime with the maximization of theoretical daily gas production as the single objective function.
[0064] In this embodiment, the single objective function is:
[0065] Where Obj is the theoretical daily gas production, m is the number of sampling points during well opening time; q iF represents the instantaneous gas production during the well opening time. s T1 is the sampling frequency; T2 is the well opening time; T2 is the well closing time.
[0066] In this embodiment, the optimization of the optimal production regime using the RIME algorithm includes:
[0067] For each intermittent production prediction model, different production regimes of the current production well are input into the intermittent production prediction model to predict the theoretical daily gas production of the current production well. Taking the maximization of theoretical daily gas production as the single objective function and production time and intermittent production number as constraints, the RIME algorithm is used to continuously update the production regime until the preset number of iterations is reached. The production regime corresponding to the maximum theoretical daily gas production is the optimal production regime of the current production well.
[0068] In this embodiment, the optimal production regime is sought by inputting different production regimes into the intermittent production prediction model (NPM) to predict the production of the gas well and calculate the objective function value obj. The maximum value of the objective function is recorded, and the different production regimes are updated again according to the four formulas of the RIME algorithm. The next iteration is carried out until the maximum number of iterations is met. The production regime corresponding to the maximum recorded objective function value is the optimal production regime for the well. This is the process of the intermittent production optimization generator (AG) finding the optimal production regime.
[0069] The production regime corresponding to the maximum theoretical daily gas production of each production well is the optimal production regime for that well.
[0070] In this embodiment, the optimal production regime for intermittent gas wells is output, the algorithm is deployed, and field implementation is guided; the optimal production regime is automatically executed, and the optimal production regime is sent to the automatic control device for automated execution.
[0071] like Figure 3 As shown, when establishing the intermittent well production relationship simulator, the input parameters are different intermittent well regimes. The well opening time and shut-in time in the intermittent well regime are within the range of historical intermittent well regimes. The output (target) is the gas production of the gas well under the intermittent well regime. An intermittent well production prediction model is constructed. The intermittent well regime and production relationship simulator is trained by production dynamic data such as oil pressure, casing pressure, and instantaneous gas production under historical intermittent well regimes. The intermittent well production prediction model (NPM) is generated. By inputting the intermittent well regime that needs to be predicted for the gas well through the intermittent well production prediction model (NPM), the theoretical gas production of the gas well under the intermittent well regime can be obtained.
[0072] like Figure 4As shown, the optimization generator is actually an intelligent optimization algorithm. The optimization process is as follows: Based on the historical well opening and closing system data of gas wells, N intermittent opening systems are randomly generated. The opening and closing times of the initial N intermittent opening systems are within the reasonable range of the historical intermittent opening systems. The gas production of the gas well under each intermittent opening scheme is calculated through the intermittent production prediction model (NPM). The intermittent opening scheme with the largest gas production is the optimal intermittent opening scheme. The optimal intermittent opening scheme is recorded and saved. The N intermittent opening schemes are updated according to the update method of the optimization algorithm. The process is iterated and looped, recording and updating the optimal intermittent opening scheme until the stopping condition is met. The optimal intermittent opening scheme and the theoretical gas production of the gas well under the optimal intermittent injection scheme are output.
[0073] The algorithm used in this invention is described as follows:
[0074] To address the issues of increased computational time and memory consumption in the GBDT algorithm when dealing with large datasets or high-dimensional datasets, and the costly calculation of split gain at each split point in the XGBoost algorithm, especially with large datasets, Microsoft proposed the LightGBM (Light Gradient Boosting Machine) algorithm in 2017. This algorithm, also an improvement on GBDT, effectively solves the problem of handling massive datasets compared to GBDT and XGBoost, achieving excellent results in practical applications. The LightGBM algorithm features include: histogram algorithm (finding the optimal split point, histogram difference acceleration), leaf-wise tree growth strategy, GOSS, EFB, support for categorical features, high-efficiency parallelism, and cache hit rate optimization.
[0075] like Figure 4 As shown in the figure, after obtaining the histogram of a leaf, LightGBM can obtain the histogram of its sibling leaf at a very low cost by subtracting the histograms. As shown in the figure below, once the histogram of a certain leaf and the histogram of its parent node are obtained, the histogram of another sibling leaf can also be obtained quickly. By using this method, the speed of the LightGBM algorithm is further improved.
[0076] like Figure 5As shown, the LightGBM algorithm employs a depth-constrained leaf-wise growth strategy (reducing computational burden and avoiding overfitting). This strategy uses a leaf-wise splitting method, where each split occurs only on the leaf node with the largest splitting gain, leaving other leaf nodes unsplit. This method results in smaller errors and faster learning compared to layer-wise splitting. However, the lack of splitting on other leaves leads to insufficient refinement, and repeatedly splitting only one leaf per layer increases tree depth, potentially causing overfitting. Therefore, the LightGBM algorithm constrains tree depth during leaf-wise growth to prevent overfitting.
[0077] The LightGBM algorithm introduces Gradient-based One-Side Sampling (GOSS), which aims to reduce the number of samples by using the gradient magnitude of the samples as a measure of their importance. It retains all samples with large gradients ("retaining all difficult problems") and randomly samples samples with small gradients ("easy problems") proportionally. This allows the model to learn information from both small and large gradient samples ("practicing difficult problems and doing some easy problems in practice, which helps maintain stable performance or even exceed expectations in real exams"). This reduces the number of samples and improves the training speed of the model without changing the original data distribution.
[0078] like Figure 6 As shown, mutually exclusive feature bundling (reducing feature dimensionality) is crucial for high-dimensional data. High-dimensional data is typically very sparse, and features are mutually exclusive (e.g., features generated through one-hot encoding will not simultaneously be zero). This type of data negatively impacts model performance and speed. The Exclusive Feature Bundling (EFB) algorithm can address the sparsity problem of high-dimensional data. As shown in the diagram, features 1, 2, and 3 are mutually exclusive sparse features. The EFB algorithm bundles these three features into a single dense new feature, which then replaces the original three features. This reduces feature dimensionality without loss of information, avoids unnecessary zero-value calculations, and improves the speed of gradient boosting algorithms.
[0079] The advantages of the LightGBM algorithm are as follows:
[0080] 1) High-efficiency parallelism on a single machine: LightGBM is designed with distributed GBDT, which enables a single machine to use as much data as possible without sacrificing speed.
[0081] 2) Low communication cost in multi-machine parallel training: LightGBM achieves linear acceleration by optimizing gradient calculation and communication strategies during multi-machine parallel training.
[0082] 3) Reduced memory usage: LightGBM uses a one-sided gradient sampling algorithm. (Gradient-based One-Side Sampling) excludes most of the samples with small weights, calculates the information gain using only the remaining samples, and reduces changes in the data distribution.
[0083] 4) Processing high-dimensional data: LightGBM designs a lossless method to reduce feature dimensionality, especially for high-dimensional sparse data. By bundling mutually exclusive features, the integrity of information can be maintained while reducing feature dimensionality.
[0084] 5) Histogram-based decision tree algorithm: LightGBM uses a histogram-based decision tree algorithm, which discretizes continuous feature values and constructs a histogram. When traversing the data, it finds the optimal split point based on the histogram statistics.
[0085] 6) Leaf-wise splitting strategy: LightGBM adopts a leaf-wise splitting strategy, which selects only the node with the largest splitting gain for splitting, thus avoiding unnecessary overhead.
[0086] 7) Reduced memory usage and faster training: LightGBM reduces memory usage and speeds up model training by saving only the discretized values of features.
[0087] Introduction to the Frost Optimization Algorithm (RIME) used in this invention:
[0088] The Rime optimization algorithm was proposed by Hang Su in February 2023. Inspired by the growth mechanism of haze ice, the algorithm proposes a frost ice search strategy by simulating the movement of soft frost ice particles.
[0089] The RIME algorithm simulates the growth process of soft and hard frost, constructing a soft frost search strategy and a hard frost puncture mechanism to realize the exploration and exploitation behaviors in optimization methods. It also improves the greedy selection mechanism in the algorithm by updating the population during the optimal solution selection phase, thus enhancing RIME's exploitation capabilities.
[0090] The core formula for particle update in the frost-ice optimization algorithm is as follows:
[0091] Among them, forst newThis is the updated frost value; forst old α is the frost value from the previous round; best is the best solution within the current search range; α and β are parameters that control the frost update speed; rand() is a random term that introduces randomness to increase the diversity of the algorithm.
[0092] Introduction to box plot algorithm, such as Figure 7 As shown, a box plot is a graphical representation of data distribution. It describes a five-fold summary of a set of data—the lower bound, the first quartile (Q1), the median, the third quartile (Q3), the upper bound, and any potential outliers. It's like a box, with the median at the center and Q1 and Q3 representing the upper and lower edges, respectively.
[0093] A quartile is a value in statistics that corresponds to one of the three division points when all values are arranged in ascending order and divided into four equal parts. There are three quartiles: the lower quartile (Q1), the median (Q2), and the upper quartile (Q3).
[0094] Interquartile Range (IQR): This is Q3-Q1, which is the difference between the upper and lower quartiles.
[0095] Upper and lower limits: The upper and lower limits are not the maximum and minimum values of the entire data sample, but rather the upper limit = the maximum value after removing outliers (removing Q3 + 1.5IQR) and the lower limit = the minimum value after removing outliers (removing Q1 - 1.5IQR). Two line segments are drawn at the upper and lower limits to serve as the dividing points of outliers.
[0096] Box plots have the following functions:
[0097] Displaying data distribution: Box plots clearly show the central tendency and dispersion of data. Identifying outliers: Outliers on a box plot may reveal abnormal data or data requiring special attention. Comparing data across different groups: By plotting box plots on different groups, we can visually compare the central location and distribution of these groups.
[0098] Although the steps in the above embodiments are described in the above order, those skilled in the art will understand that in order to achieve the effect of this embodiment, different steps do not need to be executed in such an order. They can be executed simultaneously (in parallel) or in a reverse order. These simple variations are all within the protection scope of this invention.
[0099] The second embodiment of the present invention provides a system for determining the regime based on LightGBM and RIME algorithms, the system comprising:
[0100] The data acquisition module acquires production data from different production wells in different regions and preprocesses the production data to obtain preprocessed data.
[0101] The model training module is used to divide the preprocessed data of each production well into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well.
[0102] The system optimization module is used to input different production systems into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to optimize the optimal production system with the maximization of theoretical daily gas production as the single objective function.
[0103] The system is a module used to determine the optimal production system. The production system corresponding to the maximum theoretical daily gas production of each production well is the optimal production system for that well.
[0104] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0105] It should be noted that the system determination system based on LightGBM and RIME algorithms provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.
[0106] An electronic device according to a third embodiment of the present invention includes: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to implement the above-described method for determining the regime based on the LightGBM and RIME algorithms.
[0107] A computer-readable storage medium according to a fourth embodiment of the present invention stores computer instructions, which are executed by the computer to implement the above-described method for determining the regime based on the LightGBM and RIME algorithms.
[0108] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the electronic devices and storage media described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0109] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. The programs corresponding to the software modules and method steps can be placed in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the invention.
[0110] The following is for reference. Figure 9 It shows a schematic diagram of the structure of a computer system for implementing the methods, systems, and devices of this application. Figure 9 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.
[0111] like Figure 9 As shown, the computer system includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 602 or programs loaded from storage section 608 into Random Access Memory (RAM) 603. The RAM 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An Input / Output (I / O) interface 605 is also connected to the bus 604.
[0112] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.
[0113] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.
[0114] Computer program code for performing the operations of this application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, and C++, and conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0116] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.
[0117] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.
[0118] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.
Claims
1. A method for determining the system based on LightGBM and RIME algorithms, characterized in that, The method includes: Production data from different production wells in different regions is obtained, and the production data is preprocessed to obtain preprocessed data. The preprocessed data of each production well is divided into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well. Different production regimes are input into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to find the optimal production regime with the maximization of theoretical daily gas production as the single objective function. The production regime corresponding to the maximum theoretical daily gas production of each production well is the optimal production regime for that well.
2. The method for determining the regime based on LightGBM and RIME algorithms according to claim 1, characterized in that, The production data includes: well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, pipeline pressure, daily gas production, and daily fluid discharge.
3. The method for determining the regime based on LightGBM and RIME algorithms according to claim 2, characterized in that, The preprocessed data for each production well is divided into feature data and label data, specifically including: The well depth, well diameter, well opening time, well shut-in time, well opening oil pressure, well opening casing pressure, well shut-in oil pressure, well shut-in casing pressure, and pipeline pressure of each production well are classified as characteristic data of that production well. The daily gas production and daily liquid discharge of each production well are classified as the tag data for that production well.
4. The method for determining the regime based on LightGBM and RIME algorithms according to claim 3, characterized in that, Preprocessing the production data includes: Empty sample values in the production data are filled by using the nearest values of the sample values. Remove negative values from the production data; Use box plots to remove abnormal data from the production data.
5. The method for determining the regime based on LightGBM and RIME algorithms according to claim 4, characterized in that, The single-objective function is: Where Obj is the theoretical daily gas production, m is the number of sampling points during well opening time; q i F represents the instantaneous gas production during the well opening time. s T1 is the sampling frequency; T2 is the well opening time; T2 is the well closing time.
6. The method for determining the regime based on LightGBM and RIME algorithms according to claim 5, characterized in that, The RIME algorithm is used to find the optimal production regime, including: For each intermittent production prediction model, different production regimes of the current production well are input into the intermittent production prediction model to predict the theoretical daily gas production of the current production well. With the goal of maximizing the theoretical daily gas production as the single objective function and production time and number of intervals as constraints, the RIME algorithm is used to continuously update the production regime until the preset number of iterations is reached. The production regime corresponding to the theoretical maximum daily gas production is the optimal production regime for the current production well.
7. A system for determining a system based on LightGBM and RIME algorithms, characterized in that, The system includes: The data acquisition module acquires production data from different production wells in different regions and preprocesses the production data to obtain preprocessed data. The model training module is used to divide the preprocessed data of each production well into feature data and label data; based on the feature data and label data, the LightGBM algorithm is used to train the intermittent production prediction model for each production well. The system optimization module is used to input different production systems into the intermittent production prediction model of each production well; for each production well's intermittent production prediction model, the RIME algorithm is used to optimize the optimal production system with the maximization of theoretical daily gas production as the single objective function. The system is a module used to determine the optimal production system. The production system corresponding to the maximum theoretical daily gas production of each production well is the optimal production system for that well.
8. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by the processor to implement the system determination method based on the LightGBM and RIME algorithms as described in any one of claims 1-7.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are executed by the computer to implement the system determination method based on the LightGBM and RIME algorithms as described in any one of claims 1-7.