A Machine Learning-Based Method and Apparatus for Determining the Boundary of Macrocontrol Charts for Screw Pumps

Abstract

This application provides a method and apparatus for determining the boundary of a macroscopic control chart of a screw pump based on machine learning. The method includes: calculating the pump head parameter of each screw pump in a preset area, and representing the pump head parameter and corresponding pump efficiency parameter of the screw pumps in the form of a data distribution density heat map to determine the working distribution point of the screw pumps in the preset area; using the pump head parameter and pump efficiency parameter of the screw pumps to train a preset neural network training model to obtain a pump efficiency change curve representing the mapping relationship between the pump head parameter and the pump efficiency parameter; and combining the data distribution density heat map and the pump efficiency change curve to determine the boundary of the macroscopic control chart of the screw pump.

Description

Technical Field

[0001] This application relates to the field of petroleum engineering technology, and in particular to a method and apparatus for determining the boundary of a macroscopic control chart of a screw pump based on machine learning. Background Technology

[0002] The relationship between pump efficiency and head of a screw pump is a crucial indicator of whether the pump is operating normally. Analyzing the relationship between pump efficiency and head of all screw pumps in a specific region provides a macroscopic understanding of the operational status of all screw pump wells in that oilfield. The macroscopic control chart for screw pump wells is created by dividing the chart into normal and abnormal operating condition zones based on the relationship between pump head and efficiency. By comparing the actual operating conditions of individual or multiple screw pumps with this macroscopic control chart, the operational status of the screw pumps can be evaluated.

[0003] Determining a reasonable operating range for screw pumps, i.e., the normal operating range, is crucial for accurately judging the working status of screw pumps. Currently, most research and applications of macro control charts, both domestically and internationally, focus on oil pump wells, with limited research on macro control charts for screw pump wells. Furthermore, much of this research leans towards empirical or theoretical calculations, resulting in significant discrepancies between the obtained expressions for the relationship between pump efficiency and pump head and actual applications. Summary of the Invention

[0004] This application provides a method and apparatus for determining the boundary of a macroscopic control chart for a screw pump based on machine learning. It can accurately determine the relationship between pump efficiency and pump head of a screw pump well within a preset area, and determine a reasonable operating range through a heat map of the working point distribution, providing guidance for the determination and adjustment of the production system of the screw pump well.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] Firstly, a method for determining the boundary of a macroscopic control chart for a screw pump based on machine learning is provided, the method including:

[0007] The pump head parameters of each screw pump in a preset area are calculated and obtained respectively. The pump head parameters and corresponding pump efficiency parameters of the screw pumps are represented in the form of a data distribution density heat map to determine the working distribution points of the screw pumps in the preset area.

[0008] The pump head parameters and pump efficiency parameters of several screw pumps are used to train a preset neural network training model to obtain a pump efficiency change curve that represents the mapping relationship between pump head parameters and pump efficiency parameters.

[0009] By combining the data distribution density heatmap and the pump efficiency variation curve, the boundary of the macroscopic control diagram of the screw pump is determined.

[0010] As can be seen from the method described in the first aspect, since the pump head parameters and corresponding pump efficiency parameters in the preset area are trained and learned by using a neural network model, the relationship between pump efficiency and pump head is closer to the actual situation on site. Combined with the working point distribution density map, it can intuitively reflect the working point distribution status in the preset area. Based on this, the macro control chart boundary is more in line with the actual situation and has a smaller deviation from the actual situation, thus providing a more reasonable theoretical basis and guidance for mastering and adjusting the working system of the screw pump on site.

[0011] In one possible design scheme, the steps for calculating the pump head parameters of the screw pump include:

[0012] Calculate the pump outlet pressure P based on the first expression. outlet :

[0013] P outlet =TP+ρgh pump

[0014] In the formula, TP is the oil pressure, and h is the hydraulic pressure. pump This refers to the depth of the pump.

[0015] Calculate the pump inlet pressure P according to the second expression. inlet :

[0016] P inlet =CP+ρgh Submergence

[0017] In the formula, CP represents the sleeve pressure, and h represents the pressure. Submergence Submersion degree;

[0018] The pump head parameter P of the screw pump is calculated based on the third expression. diff :

[0019] P diff =P outlet -P inlet

[0020] In the formula, P outlet P is the pump outlet pressure. inlet This is the pump inlet pressure.

[0021] In one possible design scheme, the pump head parameters and corresponding pump efficiency parameters of several screw pumps are represented in the form of a data distribution density heat map to determine the operating distribution points of several screw pumps within a preset area, including:

[0022] The pump efficiency parameters of each screw pump are normalized according to the fourth expression:

[0023]

[0024] In the formula, PE 100The pump efficiency is the pump efficiency at a speed of 100 rpm, where rpm is any speed. PE rpm This represents the pump efficiency at that speed.

[0025] In one possible design scheme, the pump head parameters and corresponding pump efficiency parameters of several screw pumps are represented in the form of a data distribution density heat map to determine the operating distribution points of several screw pumps within a preset area, and the scheme further includes:

[0026] The pump head parameter of each screw pump is normalized according to the fifth expression:

[0027]

[0028] In the formula, X norm For the normalized pump head, X max With X min These are the maximum and minimum rated pump heads corresponding to the current type of pump, respectively, and X is the pump head obtained from the current calculation.

[0029] The normalized pump head and pump efficiency parameters are displayed in the form of a data distribution density heatmap.

[0030] In one possible design scheme, the pump head parameters and pump efficiency parameters of several screw pumps are used to train a pre-defined neural network training model, including:

[0031] The pump head parameter of the screw pump is determined as the input of the preset neural network training model, and the pump efficiency parameter corresponding to the pump head parameter is determined as the output of the preset neural network training model. The preset neural network training model is trained using the pump head parameters and corresponding pump efficiency parameters of several screw pumps to obtain the pump efficiency change curve, which represents the mapping relationship between the pump head parameter and the pump efficiency parameter.

[0032] In one possible design, the preset neural network training model is a BP neural network model.

[0033] In one possible design scheme, the boundaries of the macroscopic control diagram of the screw pump are determined by combining the data distribution density heat map and the pump efficiency variation curve, including:

[0034] By combining the data distribution density heat map, the pump efficiency change curve is shifted upward and downward in parallel, so that a preset number of distribution working points within the preset area are located within the range of the two parallel trend lines, thus determining the boundary of the macro control chart of the screw pump.

[0035] One possible design approach also includes:

[0036] Obtain the production parameters of the screw pump within a preset area, wherein the production parameters include: oil pressure, sleeve pressure, submersion degree, pump depth, screw pump speed, pump efficiency, and fluid density.

[0037] Secondly, embodiments of this application also provide a machine learning-based device for determining the boundary of a screw pump's macroscopic control chart, the device comprising:

[0038] The module is used to calculate the pump head parameters of each screw pump in a preset area, and to represent the pump head parameters and corresponding pump efficiency parameters of the screw pumps in the form of a data distribution density heat map to determine the working distribution points of the screw pumps in the preset area.

[0039] The training module is used to train a preset neural network training model using the pump head parameters and pump efficiency parameters of several screw pumps to obtain a pump efficiency change curve that represents the mapping relationship between the pump head parameters and the pump efficiency parameters.

[0040] The determination module is used to determine the boundaries of the macroscopic control diagram of the screw pump by combining the data distribution density heat map and the pump efficiency change curve.

[0041] Thirdly, embodiments of this application provide a storage medium storing a computer program, which is executed by a computer during runtime to perform the machine learning-based macro control chart boundary determination method for screw pumps as described in any possible implementation of the first aspect.

[0042] Fourthly, a computer program product is provided, including a computer program or instructions that, when run on a computer, cause the computer to execute the machine learning-based screw pump macrocontrol chart boundary determination method described in any possible implementation of the first aspect. Attached Figure Description

[0043] Figure 1 A schematic diagram illustrating the implementation process of the machine learning-based macroscopic control chart boundary determination method for screw pumps provided in this application embodiment;

[0044] Figure 2 A normalized heat map of pump head and pump efficiency distribution provided for the application embodiments;

[0045] Figure 3 A schematic diagram showing the combination and comparison of pump efficiency variation curves and distribution heat maps provided in the application embodiments;

[0046] Figure 4 Macroscopic control diagram of screw pump provided for the application embodiment;

[0047] Figure 5 A schematic diagram of the structure of a machine learning-based screw pump macro control map boundary determination device provided in the application embodiment;

[0048] Figure 6A schematic diagram of the structure of the electronic device provided in the application embodiment. Detailed Implementation

[0049] The technical solution in this application will now be described with reference to the accompanying drawings.

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0052] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0054] To address the problems existing in related technologies, this application provides a method for determining the boundary of a screw pump's macroscopic control chart based on machine learning. This method is applied to electronic devices, which may specifically include mobile phones, tablets, televisions (also known as smart screens, large-screen devices, etc.), laptops, ultra-mobile personal computers (UMPCs), handheld computers, netbooks, personal digital assistants (PDAs), wearable electronic devices, in-vehicle devices (also known as vehicle systems), virtual reality devices, etc. This application does not impose any limitations on these. The functions implemented by the desktop component interaction method provided in this application can be achieved by the electronic device's processor calling program code, whereby the program code can be stored in a computer storage medium.

[0055] This application provides a method for determining the boundary of a macroscopic control chart for a screw pump based on machine learning. Figure 1 This application provides a schematic diagram of the implementation process of a machine learning-based method for determining the boundary of a screw pump's macroscopic control chart, as illustrated in the embodiments of this application. Figure 1 As shown, it includes:

[0056] Step S1: Calculate the pump head parameters of each screw pump in the preset area, and represent the pump head parameters and corresponding pump efficiency parameters of the screw pumps in the form of a data distribution density heat map to determine the working distribution points of the screw pumps in the preset area.

[0057] Step S2: Use the pump head parameters and pump efficiency parameters of several screw pumps to learn and train the preset neural network training model to obtain the pump efficiency change curve that represents the mapping relationship between the pump head parameters and the pump efficiency parameters.

[0058] Step S3: Combine the data distribution density heat map and pump efficiency change curve to determine the boundary of the macroscopic control diagram of the screw pump.

[0059] The following section provides a detailed explanation of the specific process of determining the boundary of the macroscopic control chart for a screw pump based on machine learning.

[0060] Step S1: Calculate the pump head parameter of each screw pump in the preset area, and represent the pump head parameter and corresponding pump efficiency parameter of the screw pump in the form of a data distribution density heat map to determine the working distribution point of the screw pump in the preset area.

[0061] In this embodiment of the application, before step S1, the method further includes: obtaining the production parameters of the screw pump within a preset area, wherein the production parameters include: oil pressure, casing pressure, submersion degree, pump depth, screw pump speed, pump efficiency, and fluid density.

[0062] Specifically, the production parameters of all screw pumps within a preset area can be obtained through manual input or automatic data acquisition and transmission from downhole sensors; and all production parameters are recorded in a relational database for storage so that they can be viewed and used in real time.

[0063] As one possible implementation, the step of calculating the pump head parameter of each screw pump in a preset area is as follows: calculate the pump outlet pressure P according to the first expression. outlet :

[0064] P outlet =TP+ρgh pump

[0065] In the formula, TP is the oil pressure, and h is the hydraulic pressure. pumpWhere ρ is the depth of the pump and ρ is the fluid density;

[0066] Calculate the pump inlet pressure P according to the second expression. inlet :

[0067] P inlet =CP+ρgh Submergence

[0068] In the formula, CP represents the sleeve pressure, and h represents the pressure. Submergence Submersion degree;

[0069] The pump head parameter P of the screw pump is calculated based on the third expression. diff :

[0070] P diff =P outlet -P inlet

[0071] In the formula, P outlet P is the pump outlet pressure. inlet This is the pump inlet pressure.

[0072] After calculating the pump head of each screw pump, the pump head and corresponding pump efficiency data are displayed in the form of a data distribution density heatmap, which provides a clear view of the distribution of screw pump operating conditions within the preset area. It is important to note that because the screw pump speed has a certain impact on pump efficiency, in order to map the pump efficiency at different speeds to the same interval for comparison, the pump efficiency needs to be normalized to the pump efficiency at the same preset speed. In this embodiment, the preset speed is 100 rpm. It should be noted that the preset speed can be any value; 100 rpm is merely an exemplary implementation value and is not intended to limit the calculation.

[0073] As one possible implementation, the pump efficiency parameters of each screw pump are normalized according to the fourth expression:

[0074]

[0075] In the formula, PE 100 The pump efficiency is the pump efficiency at a speed of 100 rpm, where rpm is any speed. PE rpm This represents the pump efficiency at that speed.

[0076] Since different pump types have different rated heads, in order to compare pump types with different rated heads under the same standard, it is necessary to normalize the pump head and map the head of different types of pumps to the range of [0,1].

[0077] Specifically, the pump head parameter of each screw pump is normalized according to the fifth expression:

[0078]

[0079] In the formula, X norm For the normalized pump head, X max With X min These are the maximum and minimum rated pump heads corresponding to the current type of pump, respectively, and X is the pump head obtained from the current calculation.

[0080] Please see Figure 2 The normalized pump head and pump efficiency parameters are displayed in the form of a data distribution density heatmap.

[0081] By using the method steps in step S1, the pump head parameters and corresponding pump efficiency parameters of each screw pump in a preset area are calculated and displayed using a data distribution density heat map, making it easy to intuitively understand the working position relationship of the screw pumps.

[0082] Step S2: Use the pump head parameters and pump efficiency parameters of several screw pumps to learn and train the preset neural network training model to obtain the pump efficiency change curve that represents the mapping relationship between the pump head parameters and the pump efficiency parameters.

[0083] Specifically, after obtaining the pump head parameters and the corresponding pump efficiency parameters, a preset neural network training model is used to learn the distribution of the screw pump's operating points, thereby obtaining the variation law of the pump efficiency parameters with the pump head parameters.

[0084] In this embodiment, the preset neural network training model adopts a BP (BackPropagation) neural network model. BP neural networks are the most commonly used networks in machine learning methods; they are multi-layered feedforward neural networks. In a BP neural network, data is propagated forward, and errors are propagated backward. A BP neural network consists of a data input layer, a parameter hidden layer, and a result output layer. The signal enters the hidden layer from the input layer through a non-linear activation function, and finally reaches the output layer. The error between the output layer result and the expected result is propagated back to the hidden layer, and finally to the input layer, adjusting the weight coefficients and bias vectors of each layer in sequence. Gradient descent is used to gradually approximate the weight and coefficient bias vector values ​​that satisfy the error accuracy, enabling the network to achieve optimal performance.

[0085] Using the screw pump head as the input of the neural network and the pump efficiency as the output of the neural network, the preset neural network training model is trained using the data obtained in step S1 to obtain the mapping relationship between pump head and pump efficiency; and using this mapping relationship, the variation trend of normalized pump head in the range of [0,1] is plotted.

[0086] Compared to using physical relationships to determine the relationship between pump efficiency and head, using neural networks to learn from real data of screw pumps within a preset area results in a relationship between pump efficiency and head that is closer to the actual situation on site, leading to smaller deviations.

[0087] Step S3: Combine the data distribution density heat map and pump efficiency change curve to determine the boundary of the macroscopic control diagram of the screw pump.

[0088] Please see Figure 3 In detail, by combining the data distribution density heat map, the pump efficiency change curve is shifted upward and downward in parallel, so that a preset number of distribution working points within the preset area are located within the range of the two parallel trend lines, thus determining the boundary of the macro control chart of the screw pump.

[0089] In this embodiment, the pump efficiency variation curve between pump head and pump efficiency can be obtained according to the method in step S2. This pump efficiency variation curve is then shifted horizontally upwards and downwards. Optionally, the upward shift is 20%, and the downward shift is 20%. The area between the curve corresponding to the upward shift and the downward shift is defined as the target area. It should be noted that the data for the upward and downward shifts can be adjusted according to actual conditions and are not limited here. The screw pump well macro-control chart is essentially a simplified diagram of the relationship between pump head and pump efficiency. By analyzing the working condition of the screw pump well using this chart, a comparable pump efficiency model can be established for different blocks and under different conditions.

[0090] Assuming the region between the curve corresponding to the parallel upward shift of the pump efficiency change curve and the curve corresponding to the parallel downward shift of the pump efficiency change curve is defined as the first region, and combining the first region with the data distribution density heatmap, a predetermined number of screw pump operating points within the predetermined region are located within the first region, thus determining the boundary of the screw pump macroscopic control chart. It should be noted that the predetermined number corresponds to more than half of the total number of screw pump operating points within the predetermined region. Optionally, the predetermined number can be 70% of the total number of screw pump operating points within the predetermined region; the specific number is not limited.

[0091] Please see Figure 4 Based on the data distribution density heat map and the pump efficiency change curves of parallel upward and downward movement, the boundaries of the macroscopic control diagram of the screw pump can be divided into three areas: small parameter deviation, reasonable parameter deviation, and large parameter deviation.

[0092] In this embodiment, the pump head during the screw pump well production process is first calculated, then normalized to the range of [0,1], and the pump efficiency is normalized to the pump efficiency at a rotational speed of 100 rpm. A heat map of the pump operating point distribution is generated based on the processed pump efficiency and head data. Then, a BP neural network method is used to learn the variation law of pump efficiency with pump head, obtaining the mapping relationship between pump head and pump efficiency. Finally, the heat map of the pump operating point distribution is combined with the pump efficiency variation curve to determine the boundaries of the screw pump macroscopic control chart.

[0093] One possible implementation involves first calculating the pump head data, then mapping the pump head and pump efficiency distribution to a fixed range for easy comparison. Next, a heat map of the screw pump's operating point distribution is created using the processed pump head and efficiency data. Then, a BP neural network is used to determine the pump efficiency variation curve, and combined with the heat map of the screw pump's operating point distribution, reasonable limits for the macroscopic control chart are determined. The limits of the screw pump's macroscopic control chart help distinguish the operating state of the screw pump, aid in understanding the overall operating status of screw pumps within an oilfield or block, and provide a theoretical basis and guidance for adjusting and optimizing the screw pump production system.

[0094] Please see Figure 5 This application embodiment also provides a machine learning-based macroscopic control chart boundary determination device 20 for screw pumps, the device 20 comprising:

[0095] The module 210 is used to calculate the pump head parameters of each screw pump in a preset area, and to represent the pump head parameters and corresponding pump efficiency parameters of the screw pumps in the form of a data distribution density heat map to determine the working distribution points of the screw pumps in the preset area.

[0096] Training module 220 is used to train a preset neural network training model using the pump head parameters and pump efficiency parameters of several screw pumps to obtain a pump efficiency change curve that represents the mapping relationship between pump head parameters and pump efficiency parameters.

[0097] Module 230 is used to determine the boundaries of the macroscopic control diagram of the screw pump by combining the data distribution density heat map and the pump efficiency change curve.

[0098] In summary, this application provides a method and apparatus for determining the boundary of a macroscopic control chart of a screw pump based on machine learning. The method includes: calculating the pump head parameter of each screw pump in a preset area, and representing the pump head parameter and corresponding pump efficiency parameter of the screw pumps in the form of a data distribution density heat map to determine the working distribution point of the screw pumps in the preset area; using the pump head parameter and pump efficiency parameter of the screw pumps to train a preset neural network training model to obtain a pump efficiency change curve representing the mapping relationship between the pump head parameter and the pump efficiency parameter; and combining the data distribution density heat map and the pump efficiency change curve to determine the boundary of the macroscopic control chart of the screw pump.

[0099] It should be noted that, in the embodiments of this application, if the above-described method for determining the macroscopic control chart boundary of a screw pump based on machine learning is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0100] Accordingly, this application provides a storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps in determining the degree of interest provided in the above embodiments.

[0101] This application provides an electronic device; Figure 6 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application, such as... Figure 6 As shown, the electronic device 500 includes: a processor 501, at least one communication bus 502, a user interface 503, at least one external communication interface 504, and a memory 505. The communication bus 502 is configured to enable communication between these components. The user interface 503 may include a display screen, and the external communication interface 504 may include standard wired and wireless interfaces. The processor 501 is configured to execute a program stored in the memory for determining the degree of interest, to implement the steps in determining the degree of interest provided in the above embodiment.

[0102] The descriptions of the above embodiments of the electronic devices and storage media are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the embodiments of the computer devices and storage media of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0103] It should be noted that the descriptions of the storage medium, electronic device, and trash can embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0104] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0105] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0107] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0108] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0109] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0110] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0111] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining the boundary of a macroscopic control chart for a screw pump based on machine learning, characterized in that, The method includes: The pump head parameters of each screw pump in a preset area are calculated and obtained respectively. The pump head parameters and corresponding pump efficiency parameters of the screw pumps are represented in the form of a data distribution density heat map to determine the working distribution points of the screw pumps in the preset area. The pump head parameters and pump efficiency parameters of several screw pumps are used to train a preset neural network training model to obtain a pump efficiency change curve that represents the mapping relationship between pump head parameters and pump efficiency parameters. By combining the data distribution density heat map and the pump efficiency variation curve, the boundaries of the macroscopic control diagram of the screw pump are determined; The steps for calculating the pump head parameters of a screw pump include: Calculate the pump outlet pressure Poutlet based on the first expression: In the formula, TP is the oil pressure and hpump is the pump depth; Calculate the pump inlet pressure Pinlet according to the second expression: In the formula, CP is the casing pressure, and hSubmergence is the submergence degree; The pump head parameter Pdiff of the screw pump is calculated based on the third expression: In the formula, Poutlet is the pump outlet pressure and Pinlet is the pump inlet pressure; The step of representing the pump head parameters and corresponding pump efficiency parameters of several screw pumps in the form of a data distribution density heat map to determine the working distribution points of several screw pumps within a preset area includes: The pump efficiency parameters of each screw pump are normalized according to the fourth expression: In the formula, PE100 is the pump efficiency at a speed of 100, rpm is any speed, and PErpm is the pump efficiency at that speed. The method of representing the pump head parameters and corresponding pump efficiency parameters of several screw pumps in the form of a data distribution density heat map to determine the working distribution points of several screw pumps within a preset area also includes: The pump head parameter of each screw pump is normalized according to the fifth expression: In the formula, Xnorm is the normalized pump head, Xmax and Xmin are the maximum and minimum rated pump heads corresponding to the current type of pump, respectively, and X is the pump head obtained by calculation. The normalized pump head and pump efficiency parameters are displayed in the form of a data distribution density heatmap. The process of using the pump head parameters and pump efficiency parameters of several screw pumps to learn and train a preset neural network training model includes: The pump head parameter of the screw pump is determined as the input of the preset neural network training model, and the pump efficiency parameter corresponding to the pump head parameter is determined as the output of the preset neural network training model. The preset neural network training model is trained using the pump head parameters and corresponding pump efficiency parameters of several screw pumps to obtain the pump efficiency change curve that represents the mapping relationship between the pump head parameter and the pump efficiency parameter. The determination of the macroscopic control diagram boundary of the screw pump by combining the data distribution density heat map and the pump efficiency variation curve includes: By combining the data distribution density heat map, the pump efficiency change curve is shifted upward and downward in parallel, so that a preset number of distribution working points within the preset area are located within the range of the two parallel trend lines, thus determining the boundary of the macro control chart of the screw pump.

2. The method for determining the boundary of a screw pump macroscopic control chart based on machine learning according to claim 1, characterized in that, The preset neural network training model is a BP neural network model.

3. The method for determining the boundary of a screw pump macroscopic control chart based on machine learning according to claim 1, characterized in that, The method further includes: Obtain the production parameters of the screw pump within a preset area, wherein the production parameters include: oil pressure, sleeve pressure, submersion degree, pump depth, screw pump speed, pump efficiency, and fluid density.

4. A machine learning-based device for determining the boundary of a screw pump's macroscopic control chart, characterized in that, The apparatus includes a module for performing the machine learning-based method for determining the boundary of a macroscopic control chart of a screw pump, as described in any one of claims 1-3.

5. A storage medium, characterized in that, The storage medium stores a computer program, which is executed by a computer as described in any one of claims 1-3, to determine the boundary of the macroscopic control chart of a screw pump based on machine learning.

Images