New well yield prediction method based on time sequence alignment and related equipment

Abstract

The invention discloses a new well yield prediction method based on time sequence alignment and related equipment, and belongs to the technical field of oilfield development. Historical data of all single wells are acquired, the historical data are classified according to areas, production years and well types, and a new well yield prediction result is obtained by using a time sequence alignment method. Establishing a single-well daily oil production reference table according to regions, well types and years, and finally predicting the yield of the new well in combination with the production plan data of the new well. By adopting the method, the low-efficiency repeated labor of data collection and arrangement can be reduced, the working efficiency is improved, an existing contribution rate prediction method is replaced, the prediction precision is improved, the average prediction error rate is smaller than 5%, technical support is provided for annual production allocation, monthly production allocation, planning scheme compiling, yield monitoring and early warning services, and the method is suitable for popularization and application. And meanwhile, the method can be extended to the fields of previous-year well yield prediction and measure effect prediction, and has a wide application prospect.

Description

Technical Field

[0001] The invention belongs to the technical field of oilfield development, and in particular relates to a new well production prediction method based on time series alignment and related equipment. Background Art

[0002] Production forecasting is the core of oilfield production allocation and planning. At present, each oilfield mainly grasps the law of reservoir changes and predicts future production through the "three rates" (decline rate, availability rate, and contribution rate). However, as the development objects become more and more complex, the development methods and development processes continue to improve, and there are more and more factors affecting oil well production. The difficulty of predicting availability rate and contribution rate continues to increase. How to accurately and precisely predict the production of new wells is an important research direction.

[0003] At present, the disclosed methods mainly focus on using machine learning to predict production. For example, CN114925623A discloses a method and system for predicting oil and gas reservoir production, which uses a long short-term memory network static model to predict oil and gas production, and then adjusts the predicted production through a Kalman filter dynamic adjustment model. The production prediction object is an oil and gas reservoir. However, this method can only predict the production of an oil and gas reservoir, and is not refined to a single well. CN110400006A discloses a method for predicting oil well production based on a deep learning algorithm, which comprehensively considers the dynamic and static parameters of the oil well and applies a deep learning method to construct a model. Construction and prediction, because the input data set is the dynamic and static data of the oil well, this method is only applicable to the production prediction of the wells that have been put into production; CN114575802A discloses a method for predicting the production of oil wells in high water-cut oil reservoirs based on machine learning. Through the vector autoregression algorithm, the oil production of the oil well and the injection volume of the water injection well are used as prediction influencing factors, and the time series model is constructed using the liquid volume change curves of the water injection well and the oil production well. The time step of the fitting curve is determined by selecting the lag order, and the oil well production at different time steps in the future is iteratively calculated. However, this method is only applicable to the production prediction of old wells in water-drive oil reservoirs. Summary of the invention

[0004] In order to overcome the shortcomings of the above-mentioned technology, the present invention provides a new well production prediction method based on time series alignment and related equipment, which can solve the technical problem that the existing prediction method cannot make high-precision predictions on the production of new wells.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A new well production prediction method based on time series alignment, comprising:

[0007] Obtain historical data of all single wells and classify the historical data by region, production year, and well type;

[0008] Based on the time series alignment method and combined with the classified historical data, a single well daily oil production reference table is established;

[0009] Combining the production plan data of the new well and the single well daily oil production reference table, the predicted production of the new well is obtained.

[0010] Furthermore, the historical data includes basic information and monthly production data of all single wells.

[0011] Furthermore, the basic information includes well number, well type, well type, deployment year, commissioning date, target coordinates, affiliated unit, operating area, oil field, development unit and geographical location; the production data includes production date, daily liquid production, daily oil production, daily water production, verified daily liquid production, verified daily oil production, verified daily water production, production time and water content.

[0012] Furthermore, the specific steps of establishing a single well daily oil production reference table are as follows:

[0013] The classified historical data are aligned according to the time step, and the average daily oil production of all single wells corresponding to each time point is calculated. All average daily oil productions are screened according to the region, production year and well type. Based on the screened average daily oil production, a single well daily oil production reference table is obtained.

[0014] Furthermore, the average daily oil production of all single wells corresponding to each time point is calculated by weighted average.

[0015] Furthermore, the production plan data is obtained by the production plan input device through batch analysis and import or manual entry.

[0016] Furthermore, the predicted production of the new well is obtained based on the predicted oil production of the directional well, the predicted oil production of the high-angle well and the predicted oil production of the horizontal well corresponding to the new well; the predicted oil production of the directional well, the predicted oil production of the high-angle well and the predicted oil production of the horizontal well are all calculated through the production plan data and the single well daily oil production reference table.

[0017] A new well production prediction system based on time series alignment, used to implement the steps of the new well production prediction method based on time series alignment, comprising:

[0018] The data classification module is used to obtain the historical data of all single wells and classify the historical data according to region, production year, and well type;

[0019] The time series alignment module is used to establish a single well daily oil production reference table based on the time series alignment method and combined with the classified historical data;

[0020] The production prediction module is used to combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

[0021] A device comprising:

[0022] Memory for storing computer programs;

[0023] A processor is used to implement the steps of the above-mentioned new well production prediction method based on time series alignment when executing the computer program.

[0024] A computer-readable storage medium stores a computer program, which is used to implement the steps of the above-mentioned new well production prediction method based on time series alignment when executed by a processor.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention also provides a new well production prediction method based on time series alignment. The method obtains historical data of all single wells, classifies the historical data according to region, production year, and well type, and then uses the time series alignment method to establish a single well daily oil production reference table by region, well type, and year, and finally predicts the new well production in combination with the production plan data of the new well. The use of this method can reduce the inefficient and repetitive work of data collection and organization, improve work efficiency, replace the existing contribution rate prediction method, and improve prediction accuracy. The average prediction error rate is less than 5%, providing technical support for annual production allocation, monthly production allocation, planning scheme preparation, production monitoring and early warning services, and can be extended to the fields of well production prediction and measure effect prediction in the previous year, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flow chart of a new well production prediction method based on time series alignment provided by an embodiment of the present invention;

[0028] Figure 2 A comparison chart of the monthly predicted production and actual production of new wells in 2020 at the First Oil Production Plant using the new well production prediction method based on time series alignment provided in an embodiment of the present invention;

[0029] Figure 3 A flow chart of a new well production prediction method based on time series alignment provided by the present invention;

[0030] Figure 4 A schematic diagram of the structure of a new well production prediction system based on time series alignment provided by the present invention. DETAILED DESCRIPTION

[0031] The present invention provides a new well production prediction method based on time series alignment, such as Figure 3 As shown, the following steps are included:

[0032] S1: Obtain the historical data of all single wells and classify the historical data according to region, production year, and well type.

[0033] The above historical data include basic information and monthly production data of all individual wells.

[0034] Specifically, the basic information includes well number, well type, well type, deployment year, commissioning date, target coordinates, affiliated unit, operating area, oil field, development unit and geographical location; production data includes production date, daily liquid production, daily oil production, daily water production, verified daily liquid production, verified daily oil production, verified daily water production, production time and water content.

[0035] S2: Based on the time series alignment method and combined with the classified historical data, a single well daily oil production reference table is established.

[0036] The specific steps for establishing a single well daily oil production reference table are as follows:

[0037] The classified historical data are aligned according to the time step, and the average daily oil production of all single wells corresponding to each time point is calculated. All average daily oil productions are screened according to the region, production year and well type. Based on the screened average daily oil production, a single well daily oil production reference table is obtained.

[0038] In particular, the average daily oil production of all single wells corresponding to each time point (ie, each natural month) is calculated by weighted average.

[0039] S3: Combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

[0040] Here, the production plan data is obtained by the production plan input device through batch analysis and import or manual entry.

[0041] Specifically, the predicted production of a new well is based on the predicted oil production of the directional well, the predicted oil production of the high-angle well and the predicted oil production of the horizontal well corresponding to the new well; the predicted oil production of the directional well, the predicted oil production of the high-angle well and the predicted oil production of the horizontal well are all calculated through the production plan data and the single well daily oil production reference table.

[0042] like Figure 4 As shown, the present invention also provides a new well production prediction system based on time series alignment, including: a data classification module, used to obtain the historical data of all single wells, and classify the historical data according to region, production year, and well type; a time series alignment module, used to establish a single well daily oil production reference table based on the time series alignment method and the classified historical data; a production prediction module, used to combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

[0043] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the new well production prediction method based on time series alignment when executing the computer program.

[0044] When the processor executes the computer program, the above-mentioned steps of new well production prediction based on time series alignment are implemented, for example: historical data of all single wells are obtained, and the historical data are classified according to region, production year, and well type; based on the time series alignment method, a single well daily oil production reference table is established in combination with the classified historical data; and the predicted production of the new well is obtained by combining the production plan data of the new well and the single well daily oil production reference table.

[0045] Alternatively, when the processor executes the computer program, the functions of each module in the above system are implemented, for example: a data classification module is used to obtain the historical data of all single wells and classify the historical data according to region, production year, and well type; a time series alignment module is used to establish a single well daily oil production reference table based on the time series alignment method and the classified historical data; a production prediction module is used to combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

[0046] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete preset functions, and the instruction segments are used to describe the execution process of the computer program in the new well production prediction device based on time series alignment. For example, the computer program can be divided into a data classification module, a time series alignment module and a production prediction module; the specific functions of each module are as follows: a data classification module, which is used to obtain the historical data of all single wells and classify the historical data according to the region, production year, and well type; a time series alignment module, which is used to establish a single well daily oil production reference table based on the time series alignment method and the classified historical data; a production prediction module, which is used to combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

[0047] The new well production prediction device based on time series alignment can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The new well production prediction device based on time series alignment may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above is an example of a new well production prediction device based on time series alignment, and does not constitute a limitation on the new well production prediction device based on time series alignment, and may include more components than the above, or a combination of certain components, or different components. For example, the new well production prediction device based on time series alignment may also include input and output devices, network access devices, buses, etc.

[0048] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The processor is the control center of the new well production prediction based on time series alignment, and uses various interfaces and lines to connect various parts of the whole new well production prediction device based on time series alignment.

[0049] The memory can be used to store the computer program and / or module, and the processor implements various functions of the new well production prediction device based on time series alignment by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0050] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0051] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the new well production prediction method based on time series alignment are implemented.

[0052] If the modules / units integrated in the new well production prediction system based on time series alignment are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0053] Based on such understanding, the present invention implements all or part of the processes in the above-mentioned new well production prediction method based on time series alignment, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned new well production prediction method based on time series alignment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or preset intermediate form, etc.

[0054] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0055] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0056] The present invention will be further described below in conjunction with the embodiments and drawings:

[0057] Example

[0058] In order to solve the problem mentioned in the background technology: the currently disclosed methods mainly focus on the application of machine learning to predict production, such as CN114925623A discloses a method and system for predicting oil and gas reservoir production, which uses a long short-term memory network static model to predict oil and gas production, and then adjusts the predicted production through a Kalman filter dynamic adjustment model. The production prediction object is an oil and gas reservoir. However, this method can only predict the production of an oil and gas reservoir, and is not refined to a single well; CN110400006A discloses a method for predicting oil well production based on a deep learning algorithm, which comprehensively considers the dynamic and static parameters of the oil well and applies a deep learning method The method is used to build and predict the model. Since the input data set is the dynamic and static data of the oil well, this method is only applicable to the production prediction of the wells that have been put into production. CN114575802A discloses a method for predicting the production of oil wells in high water-cut oil reservoirs based on machine learning. Through the vector autoregression algorithm, the oil production of the oil well and the injection volume of the water injection well are used as prediction influencing factors. The time series model is constructed using the liquid volume change curves of the water injection well and the oil production well. The time step of the fitting curve is determined by selecting the lag order, and the oil well production at different time steps in the future is iteratively calculated. However, this method is only applicable to the production prediction of old wells in water-drive oil reservoirs.

[0059] like Figure 1 As shown, this embodiment provides a new well production prediction method based on time series alignment, which automatically extracts production data over the years, applies the time series alignment method to establish a single well daily oil production level reference table by region, well type, and year, and combines the annual deployment and production plan to make a detailed monthly prediction of the new well production. This method is simple to operate and can replace the contribution rate prediction method. The reference table can be dynamically corrected as time goes by. It has the advantages of being more precise, more accurate, and more convenient. It is suitable for the production prediction of newly commissioned wells and provides intelligent technical support for oilfield production allocation, planning scheme preparation, production monitoring, and early warning.

[0060] This method includes three stages: data preparation, time series alignment statistics, and yield prediction; the specific steps are as follows:

[0061] Data preparation stage:

[0062] The system automatically extracts the basic information and production data of each oil well from the oil and gas production database (A2) (the basic information and production data are collectively referred to as historical data), and classifies and stores the data according to region, production year, and well type to provide data support for subsequent calculations.

[0063] Among them, the basic information of the well includes well number, well type, well type, deployment year, production date, target coordinates, affiliated unit, operation area, oil field, development unit, geographical location, etc. The monthly production data includes production date, daily liquid production, daily oil production, daily water production, verified daily liquid production, verified daily oil production, verified daily water production, production time, water content, etc. The two types of information are associated through the well number, and the production data is classified and stored according to the combination of three conditions: region, production year, and well type.

[0064] Time series alignment statistics stage:

[0065] The historical data of all single wells in each classification are aligned according to the time step, and then the average daily oil production (average daily oil production level) of all single wells at each time point is calculated. The data of the previous 12 months (i.e., the average daily oil production) are intercepted to establish a reference table of daily oil production of single wells by region and well type.

[0066] In this stage, the weighted average of the wells put into production in the past three years is taken to obtain the reference value of the daily oil production level in the area, as shown in formula (1):

[0067] A ij =A (i-1)j *0.7+A (i-2)j *0.2+A (i-3)j *0.1 (1)

[0068] In formula (1), A ij is the average daily oil production level of a single well in the jth month of the i-th year, A (i-1)j It is the average daily oil production of all wells put into production in the (i-1)th year in the jth month.

[0069] Production forecast stage:

[0070] The production plan data for the next year is collected through the production plan input device, combined with the single well daily oil production reference table, to accurately predict the production of new wells in each month of the next year, and the summary of 12 months is the annual allocated production.

[0071] Among them, the production plan input device supports batch analysis and import or manual entry.

[0072] The production forecast calculation is shown in formulas (2) and (3):

[0073] Q n =A n +B n +C n (2)

[0074]

[0075] J i is the value of the directional well in the i-th month in the single well daily oil production reference table, Xn-i+1 is the number of directional wells put into production in month n-i+1. Assume n=3, A3=J1*X3+J2*X2+J3*X1.

[0076]

[0077] K i is the value of the high-angle well in the i-th month in the single-well daily oil production reference table, Y n-i+1 is the number of directional wells put into production in month n-i+1. Assume n=3, B3=K1*Y3+K2*Y2+K3*Y1.

[0078]

[0079] L i is the value of the horizontal well in the i-th month in the single well daily oil production reference table, Z n-i+1 is the number of directional wells put into production in month n-i+1. Assume n=3, C3=L1*Z3+L2*Z2+L3*Z1.

[0080] The following is a set of examples to further explain and illustrate the method used in this embodiment:

[0081] The forecast method for the new well production of the first oil production plant in a certain region in 2020 is as follows:

[0082] (1) Data preparation

[0083] a. The single well monthly production data and basic well information of the wells put into production in 2019, 2018 and 2017 of the First Oil Production Plant are automatically extracted from the oil and water well production database (A2) through the data extractor.

[0084] b. The production data is classified and stored according to the well type through the data classifier, such as 2019 horizontal wells, 2019 directional wells, 2019 high-angle wells, 2018 qualitative wells, etc.

[0085] (2) Time series alignment statistics

[0086] a. Use the month labeler to label all the classified wells. For example, for a new well put into production in March 2018, the natural month March is its first month, April is its second month, and so on. February 2019 is its twelfth month, which means that the data for the first 12 months are captured.

[0087] b. Horizontally, the alignment calculator is used to calculate the average daily oil production level for all single wells in the same category with the same number of months. Vertically, the daily oil production of the same type in 2019, 2018, and 2017 is weighted averaged to calculate the daily oil production level of a single well of this type in 2020.

[0088] c. After repeating step b for many times, the reference value of daily oil production of a single well of each type in each month is obtained, as shown in Table 1, and is stored in the result memory.

[0089] Table 1 Reference table of daily oil production per well of the First Oil Production Plant in 2020

[0090]

[0091] (3) Production forecast

[0092] a. Import the monthly production plan for 2020 in batches through the plan input tool.

[0093] b. The production of each type and each natural month is calculated through the production predictor, and the sum of each type and month is the predicted total production of new wells in the first oil production plant.

[0094] Figure 2 As shown, Figure 2 This is a comparison curve chart of the monthly predicted production and actual production of new wells in 2020 at the First Oil Production Plant. It can be seen that by adopting the new well production prediction method based on time series alignment provided in this embodiment, the predicted value of the new well production is very close to the actual production value in each natural month. For example, in March, April and September, the error can almost be zero. After actual testing, the average prediction error rate is less than 5%. The use of this prediction method not only improves work efficiency, replaces the traditional contribution rate prediction method, but also improves the prediction accuracy.

[0095] In summary, the present invention provides a new well production prediction method based on time series alignment, which has the following advantages over the existing prediction methods:

[0096] This method obtains the historical data of all single wells, classifies the historical data according to region, production year, and well type, and then uses the time series alignment method to establish a single well daily oil production reference table by region, well type, and year, and finally predicts the production of new wells in combination with the production plan data of new wells; the use of this method can reduce the inefficient and repetitive work of data collection and collation, improve work efficiency, replace the existing contribution rate prediction method, improve prediction accuracy, and provide technical support for annual production allocation, monthly production allocation, planning scheme preparation, production monitoring and early warning services. At the same time, it can be extended to the fields of well production prediction and measure effect prediction in the previous year, and has broad application prospects.

[0097] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. A new well production prediction method based on time series alignment, characterized in that: include: Obtain historical data of all single wells and classify the historical data by region, production year, and well type; Based on the time series alignment method and combined with the classified historical data, a single well daily oil production reference table is established; Combining the production plan data of the new well and the single well daily oil production reference table, the predicted production of the new well is obtained.

2. A new well production prediction method based on time series alignment according to claim 1, characterized in that: in, The historical data includes basic information and monthly production data of all individual wells.

3. A new well production prediction method based on time series alignment according to claim 2, characterized in that: The basic information includes well number, well type, well type, deployment year, production date, target coordinates, affiliated unit, operating area, oil field, development unit and geographical location; the production data includes production date, daily liquid production, daily oil production, daily water production, verified daily liquid production, verified daily oil production, verified daily water production, production time and water content.

4. The method for predicting new well production based on time series alignment according to claim 1, characterized in that: in, The specific steps to establish a single well daily oil production reference table are as follows: The classified historical data are aligned according to the time step, and the average daily oil production of all single wells corresponding to each time point is calculated. All average daily oil productions are screened according to the region, production year and well type. Based on the screened average daily oil production, a single well daily oil production reference table is obtained.

5. A new well production prediction method based on time series alignment according to claim 4, characterized in that: in, The average daily oil production of all single wells corresponding to each time point is calculated by weighted average.

6. A new well production prediction method based on time series alignment according to claim 1, characterized in that: in, The production plan data is obtained by the production plan input device through batch analysis and import or manual entry.

7. The method for predicting new well production based on time series alignment according to claim 1, characterized in that: in, The predicted production of a new well is based on the predicted oil production of the directional well, the predicted oil production of the highly deviated well and the predicted oil production of the horizontal well corresponding to the new well; the predicted oil production of the directional well, the predicted oil production of the highly deviated well and the predicted oil production of the horizontal well are all calculated through the production plan data and the single well daily oil production reference table.

8. A new well production prediction system based on time series alignment, used to implement the steps of the new well production prediction method based on time series alignment according to any one of claims 1 to 7, characterized in that: include: The data classification module is used to obtain the historical data of all single wells and classify the historical data according to region, production year, and well type; The time series alignment module is used to establish a single well daily oil production reference table based on the time series alignment method and combined with the classified historical data; The production prediction module is used to combine the production plan data of the new well and the single well daily oil production reference table to obtain the predicted production of the new well.

9. A device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the new well production prediction method based on time series alignment as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the new well production prediction method based on time series alignment as described in any one of claims 1 to 7.

Images