Gas field production prediction method, device and computer equipment

Abstract

The application discloses a kind of gas field production prediction method, device and computer equipment.The method comprises: obtaining original gas field production sequence;Based on prediction parameter and original gas field production sequence, obtain predicted gas field production sequence;In response to the current predicted gas field production sequence and original gas field production sequence do not satisfy similar condition, update prediction parameter, obtain new prediction parameter;Based on new prediction parameter and original gas field production sequence, obtain new predicted gas field production sequence;Such cycle, until the current predicted gas field production sequence and original gas field production sequence satisfy similar condition, based on prediction parameter obtains target prediction parameter;Based on target prediction parameter and original gas field production sequence, the gas field production corresponding to reference time is predicted.This kind of mode, target prediction parameter can obtain the predicted gas field production sequence close to original gas field production sequence, and the gas field production corresponding to reference time predicted based on target prediction parameter is relatively accurate.

Description

Technical Field

[0001] This application relates to the field of gas field development and planning technology, and in particular to a method, apparatus and computer equipment for predicting gas field production. Background Technology

[0002] In the field of gas field development and planning, gas field production forecasting is a crucial step. A reasonable forecast of gas field production is beneficial for guiding future work deployments. Gas field production forecasting is one of the fundamental tasks for the energy industry to conduct medium- and long-term planning, and the accuracy of gas field production forecasting determines the accuracy and rationality of medium- and long-term planning.

[0003] Currently, the prediction of gas field production at a reference time is based directly on prediction parameters set based on experience and the original gas field production sequence. However, the reliability of the prediction parameters considered in the process of predicting gas field production is poor, resulting in poor accuracy of the predicted gas field production and a large deviation from the actual situation. This has a significant adverse impact on the accurate planning of gas field production in the medium and long term. Summary of the Invention

[0004] This application provides a method, apparatus, and computer device for predicting gas field production, which can be used to improve the accuracy of the predicted gas field production. The technical solution is as follows:

[0005] On one hand, embodiments of this application provide a method for predicting gas field production, the method comprising:

[0006] Obtain the original gas field production sequence, which is composed of multiple original gas field productions arranged in chronological order according to their corresponding time periods;

[0007] Based on the prediction parameters and the original gas field production sequence, a predicted gas field production sequence is obtained, wherein the predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence.

[0008] In response to the fact that the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, the prediction parameters on which the current predicted gas field production sequence is obtained are updated to obtain new prediction parameters; based on the new prediction parameters and the original gas field production sequence, a new predicted gas field production sequence is obtained; this process is repeated until the current predicted gas field production sequence meets the similarity condition with the original gas field production sequence, and then the target prediction parameters are obtained based on the prediction parameters on which the current predicted gas field production sequence is obtained;

[0009] Based on the target prediction parameters and the original gas field production sequence, predict the gas field production corresponding to the reference time.

[0010] In one possible implementation, the prediction parameters include model setting parameters; obtaining the predicted gas field production sequence based on the prediction parameters and the original gas field production sequence includes:

[0011] The original gas field production sequence is accumulated to obtain the original accumulated gas field production sequence, which is an equal-time interval sequence;

[0012] Based on the parameters set in the model, a gas field production prediction model matching the original cumulative gas field production sequence is constructed.

[0013] Based on the gas field production prediction model, a predicted cumulative gas field production sequence is obtained, and the predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence.

[0014] The predicted cumulative gas field production sequence is subjected to cumulative subtraction processing to obtain the predicted gas field production sequence.

[0015] In one possible implementation, the prediction parameters further include time interval adjustment parameters; the step of accumulating the original gas field production sequence to obtain the original accumulated gas field production sequence includes:

[0016] In response to the fact that the original gas field production sequence is a non-equidistant time interval sequence, the original gas field production sequence is accumulated based on the time interval adjustment parameter to obtain the original accumulated gas field production sequence corresponding to the target time sequence, wherein the target time sequence is an equidistant time interval sequence that matches the time sequence corresponding to the original gas field production sequence.

[0017] In one possible implementation, obtaining the predicted cumulative gas field production sequence based on the gas field production prediction model includes:

[0018] The first original cumulative gas field production in the original cumulative gas field production sequence is taken as the predicted cumulative gas field production corresponding to the first time in the target time series;

[0019] Based on the gas field production prediction model, predict the cumulative gas field production corresponding to each time in the target time series except for the first time.

[0020] The predicted cumulative gas field production corresponding to the first time and the predicted cumulative gas field production corresponding to each time are arranged in chronological order to obtain the predicted cumulative gas field production sequence.

[0021] In one possible implementation, updating the prediction parameters on which the current predicted gas field production sequence is based to obtain new prediction parameters includes:

[0022] To obtain the variation pattern of predicted gas field production with prediction parameters;

[0023] Based on the variation of the predicted gas field production with the predicted parameters and the difference between the current predicted gas field production sequence and the original gas field production sequence, the predicted parameters on which the current predicted gas field production sequence is obtained are updated to obtain new predicted parameters.

[0024] In one possible implementation, obtaining the raw gas field production sequence includes:

[0025] Obtain the measured gas field production sequence;

[0026] The measured gas field production sequence is preprocessed to obtain the original gas field production sequence.

[0027] In one possible implementation, the preprocessing of the measured gas field production sequence includes:

[0028] The measured gas field production sequence is preprocessed using at least one of outlier removal, smoothing, and standardization.

[0029] In one possible implementation, the prediction parameters are determined based on at least one of reserves factors, economic level factors, and policy guidance factors.

[0030] On the other hand, embodiments of this application provide a gas field production prediction device, the device comprising:

[0031] The first acquisition unit is used to acquire the original gas field production sequence, which is composed of multiple original gas field productions arranged in chronological order according to their corresponding time periods.

[0032] The second acquisition unit is used to acquire a predicted gas field production sequence based on the prediction parameters and the original gas field production sequence, wherein the predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence.

[0033] The update unit is used to update the prediction parameters on which the current predicted gas field production sequence is based in response to the fact that the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, so as to obtain new prediction parameters.

[0034] The third acquisition unit is used to acquire a new predicted gas field production sequence based on the new prediction parameters and the original gas field production sequence; this process is repeated until the current predicted gas field production sequence and the original gas field production sequence meet similarity conditions, and the target prediction parameters are acquired based on the prediction parameters on which the current predicted gas field production sequence is acquired.

[0035] The prediction unit is used to predict the gas field production at a reference time based on the target prediction parameters and the original gas field production sequence.

[0036] In one possible implementation, the prediction parameters include model setting parameters; the second acquisition unit is used to perform cumulative processing on the original gas field production sequence to obtain an original cumulative gas field production sequence, wherein the original cumulative gas field production sequence is an equal-time interval sequence; based on the model setting parameters, a gas field production prediction model matching the original cumulative gas field production sequence is constructed; based on the gas field production prediction model, a predicted cumulative gas field production sequence is obtained, wherein the predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence; and the predicted cumulative gas field production sequence is subjected to cumulative subtraction processing to obtain the predicted gas field production sequence.

[0037] In one possible implementation, the prediction parameters further include time interval adjustment parameters; the second acquisition unit is further configured to, in response to the original gas field production sequence being a non-equidistant time interval sequence, perform cumulative processing on the original gas field production sequence based on the time interval adjustment parameters to obtain the original cumulative gas field production sequence corresponding to the target time sequence, wherein the target time sequence is an equidistant time interval sequence that matches the time sequence corresponding to the original gas field production sequence.

[0038] In one possible implementation, the second acquisition unit is further configured to use the first original cumulative gas field production in the original cumulative gas field production sequence as the predicted cumulative gas field production corresponding to the first time in the target time series; predict the predicted cumulative gas field production corresponding to each time in the target time series other than the first time based on the gas field production prediction model; and arrange the predicted cumulative gas field production corresponding to the first time and the predicted cumulative gas field production corresponding to each time in chronological order to obtain the predicted cumulative gas field production sequence.

[0039] In one possible implementation, the updating unit is used to obtain the variation law of predicted gas field production with prediction parameters; based on the variation law of predicted gas field production with prediction parameters and the difference between the current predicted gas field production sequence and the original gas field production sequence, the prediction parameters on which the current predicted gas field production sequence is based are updated to obtain new prediction parameters.

[0040] In one possible implementation, the first acquisition unit is used to acquire a measured gas field production sequence; and to preprocess the measured gas field production sequence to obtain the original gas field production sequence.

[0041] In one possible implementation, the first acquisition unit is further configured to preprocess the measured gas field production sequence based on at least one of outlier removal, smoothing, and standardization.

[0042] In one possible implementation, the prediction parameters are determined based on at least one of reserves factors, economic level factors, and policy guidance factors.

[0043] On the other hand, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor to enable the computer device to implement any of the above-described gas field production prediction methods.

[0044] On the other hand, a computer-readable storage medium is also provided, wherein at least one computer program is stored in the computer-readable storage medium, the at least one computer program being loaded and executed by a processor to enable a computer to implement any of the above-described methods for predicting gas field production.

[0045] On the other hand, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the gas field production prediction methods described above.

[0046] The technical solution provided in this application has at least the following beneficial effects:

[0047] In this embodiment, the gas field production corresponding to the reference time is predicted based on target prediction parameters and the original gas field production sequence. The target prediction parameters are obtained by using prediction parameters that can yield a predicted gas field production sequence that meets similar conditions to the original gas field production sequence. In other words, the target prediction parameters can yield a predicted gas field production sequence that is relatively close to the original gas field production sequence, thus making the gas field production corresponding to the reference time predicted based on the target prediction parameters more accurate. This is beneficial for ensuring the accuracy and rationality of long-term gas field production planning. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the implementation environment of a gas field production prediction method provided in an embodiment of this application;

[0050] Figure 2 This is a flowchart of a gas field production prediction method provided in an embodiment of this application;

[0051] Figure 3 This is a schematic diagram illustrating a process for obtaining a predicted gas field production sequence according to an embodiment of this application;

[0052] Figure 4 This is a schematic diagram illustrating an embodiment of this application for obtaining a target predicted gas field production sequence;

[0053] Figure 5 This is a graph constructed based on the original gas field production sequence provided in an embodiment of this application;

[0054] Figure 6 This is a comparison chart of a curve constructed based on the gas field production corresponding to a reference time and a curve constructed based on the original gas field production sequence, provided in an embodiment of this application.

[0055] Figure 7 This is a schematic diagram of a gas field production prediction device provided in an embodiment of this application;

[0056] Figure 8 This is a schematic diagram of a computer software system provided in an embodiment of this application;

[0057] Figure 9 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0059] Figure 1 A schematic diagram of the implementation environment for the gas field production prediction method provided in this application embodiment is shown. This implementation environment may include: a terminal 11 and a server 12.

[0060] The gas field production prediction method provided in this application embodiment can be executed by terminal 11, server 12, or jointly by terminal 11 and server 12; this application embodiment does not limit this. In the case where the gas field production prediction method provided in this application embodiment is jointly executed by terminal 11 and server 12, server 12 undertakes the main calculation work, and terminal 11 undertakes the secondary calculation work; or, server 12 undertakes the secondary calculation work, and terminal 11 undertakes the main calculation work; or, server 12 and terminal 11 perform collaborative calculations using a distributed computing architecture.

[0061] In one possible implementation, terminal 11 can be any electronic product capable of human-computer interaction with a user through one or more methods such as a keyboard, touchpad, touchscreen, remote control, voice interaction, or handwriting device, such as a PC (Personal Computer), mobile phone, PDA (Personal Digital Assistant), wearable device, PPC (Pocket PC), tablet computer, etc. Server 12 can be a single server, a server cluster consisting of multiple servers, or a cloud computing service center. Terminal 11 and server 12 establish a communication connection via wired or wireless network.

[0062] Those skilled in the art should understand that the above-described terminal 11 and server 12 are merely examples. Other existing or future terminals or servers that are applicable to this application should also be included within the scope of protection of this application, and are hereby incorporated by reference.

[0063] Based on the above Figure 1 The implementation environment shown in this application embodiment provides a method for predicting gas field production. This method is executed by a computer device, which can be a server 12 or a terminal 11; this application embodiment does not limit the specific computer device used. Figure 2 As shown, the gas field production prediction method provided in this application includes the following steps 201 to 204:

[0064] In step 201, the original gas field production sequence is obtained. The original gas field production sequence is composed of multiple original gas field productions arranged in chronological order according to their corresponding time periods.

[0065] This application does not limit the number of original gas field productions constituting the original gas field production sequence. For example, the number of original gas field productions constituting the original gas field production sequence may be 5 or 10, etc. Each original gas field production corresponds to a time period, and the original gas field productions are arranged sequentially according to their corresponding time periods to obtain the original gas field production sequence. The time corresponding to the original gas field production can refer to a year, a month, or a day, etc., and this application does not limit this. For example, if the original gas field production corresponds to a year, the original gas field production is used to indicate the gas field production generated throughout that year; if the original gas field production corresponds to a month, the original gas field production is used to indicate the gas field production generated throughout that month; if the original gas field production corresponds to a day, the original gas field production is used to indicate the gas field production generated throughout that day. This application does not limit the type of gas field measured by the gas field production. For example, the gas field measured by the gas field production is a natural gas field; or, the gas field measured by the gas field production is an oil and gas field.

[0066] For example, arranging the times corresponding to the production of each primary gas field in chronological order can form a time series corresponding to the primary gas field production sequence. The l-th primary gas field production (l is an integer not less than 1) in the primary gas field production sequence corresponds to the l-th time in the time series corresponding to the primary gas field production sequence. That is, the number of times in the time series corresponding to the primary gas field production sequence is the same as the number of primary gas field productions in the primary gas field production sequence, and there is a one-to-one correspondence between the times in the time series corresponding to the primary gas field production sequence and the primary gas field productions in the primary gas field production sequence.

[0067] In one possible implementation, the original gas field production sequence is the measured gas field production sequence, which is a highly efficient way to obtain the original gas field production sequence.

[0068] In another possible implementation, the raw gas field production sequence is obtained by: acquiring the measured gas field production sequence; preprocessing the measured gas field production sequence to obtain the raw gas field production sequence. This method yields a more reliable raw gas field production sequence. The measured gas field production sequence refers to a sequence composed of multiple measured gas field productions arranged sequentially according to their corresponding time periods. This application does not limit the number of measured gas field productions included in the measured gas field production sequence; this can be flexibly adjusted according to actual circumstances.

[0069] After obtaining the measured gas field production sequence, the measured gas field production sequence is preprocessed to obtain a reliable raw gas field production sequence. In one possible implementation, the preprocessing of the measured gas field production sequence involves preprocessing the measured gas field production sequence based on at least one of the following methods: outlier removal, smoothing, and standardization. That is, the measured gas field production sequence can be preprocessed based on only one of these methods; it can be preprocessed based on two of these methods; or it can be preprocessed based on all three methods. This application embodiment does not limit this approach. Regardless of the method, the gas field production sequence obtained after preprocessing is referred to as the raw gas field production sequence. Of course, in some embodiments, the measured gas field production sequence can also be preprocessed based on other methods.

[0070] For example, in the case of preprocessing the measured gas field production sequence using multiple methods such as outlier removal, smoothing, and standardization, the preprocessing process is as follows: first, the measured gas field production sequence is processed using the former method; then, the gas field production sequence obtained by the former method is processed using the latter method; until the gas field production sequence processed by the last method is obtained. The execution order of the multiple methods can be set based on experience, and this embodiment does not limit this.

[0071] For example, when preprocessing the measured gas field production sequence based on two methods—outlier removal and smoothing—one can first process the measured gas field production sequence by removing outliers to obtain the first gas field production sequence, and then process the first gas field production sequence by smoothing to obtain the original gas field production sequence; alternatively, one can first process the measured gas field production sequence by smoothing to obtain the first gas field production sequence, and then process the first gas field production sequence by removing outliers to obtain the original gas field production sequence.

[0072] This application embodiment illustrates the preprocessing of a measured gas field production sequence using three methods: outlier removal, smoothing, and standardization. Exemplarily, the execution order of these three methods is outlier removal, smoothing, and standardization. In this case, the process of preprocessing the measured gas field production sequence to obtain the original gas field production sequence includes the following three steps:

[0073] Step 1: Process the measured gas field production sequence by removing outliers to obtain the first gas field production sequence.

[0074] Because the actual measurement of gas field production is affected by various factors, such as weather and testing instrument effects, the measured gas field production series may contain production figures that deviate from the normal pattern. These deviations can be considered outliers. Therefore, processing the measured gas field production series using outlier removal methods can eliminate production figures that deviate from the normal pattern.

[0075] In one possible implementation, the process of processing the measured gas field production sequence by removing outliers is as follows: for any measured gas field production in the measured gas field production sequence, if the difference between the measured gas field production and the average measured gas field production is greater than a reference threshold, the measured gas field production is removed; if the difference between the measured gas field production and the average measured gas field production is not greater than the reference threshold, the measured gas field production is retained.

[0076] The average measured gas field production refers to the average production of each measured gas field in the measured gas field production sequence. The reference threshold is set based on experience. For example, the reference threshold is the product of the standard deviation and the deviation interval coefficient. The standard deviation is the standard deviation of the production of each measured gas field in the measured gas field production sequence. The deviation interval coefficient is set based on experience or can be flexibly adjusted according to the actual application scenario. This application embodiment does not limit this. For example, the deviation interval coefficient is determined based on a preset confidence level.

[0077] For example, for the production of the i-th measured gas field x in the measured gas field production sequence i If the actual measured gas field production x i If the following formula 1 is satisfied, then the measured gas field production x i Remove.

[0078]

[0079] Where i is a positive integer not greater than q, and q is the total number of measured gas field productions in the measured gas field production sequence; This represents the average measured gas field production. x represents the measured production of the i-th gas field. i Compared with average measured gas field production The difference between them; N represents the deviation interval coefficient; S represents the standard deviation. NS represents the reference threshold.

[0080] After processing the measured gas field production sequence by removing outliers, the sequence consisting of the retained measured gas field production is called the first gas field production sequence.

[0081] Step 2: Process the production sequence of the first gas field using a smoothing method to obtain the production sequence of the second gas field.

[0082] In one possible implementation, the smoothing process for processing the first gas field production sequence is as follows: for any first gas field production that satisfies the first condition in the first gas field production sequence, calculate the average value of the number of reference first gas field productions before and after the first gas field production, centered on the first gas field production, and replace the first gas field production with the calculated average value.

[0083] For example, for any first gas field production in the first gas field production sequence, if there are a number of first gas field productions before and after the first gas field production in the first gas field production sequence centered on the first gas field production sequence, then the first gas field production is considered to satisfy the first condition.

[0084] The reference number is set based on experience or can be flexibly adjusted according to the application scenario. This application embodiment does not limit this. For example, the reference number can be the confidence interval coefficient. Assuming the reference number is p (p is an integer not less than 1), the smoothing process used in this application embodiment can be regarded as a (2p+1)-point normal distribution function filtering process.

[0085] For example, suppose the reference quantity is p, and suppose the production of the j-th gas field in the production sequence of the first gas field is y. j If (jp) is an integer not less than 1 and (j+p) is an integer not greater than n, then it means that there exists a sequence of production outputs of the first gas field with the production output y of that first gas field as the starting point. j The production of each of the p gas fields before and after the center, that is, the production y of the first gas field. j The first condition is met. Here, n represents the total output of the first gas field in the first gas field production sequence. In this case, the first gas field production y is taken from the first gas field production sequence. j Given the production of p gas fields before and after the center (a total of 2p gas fields), calculate the production of these 2p gas fields and the total production y of the gas field. j The average value y j ', y is obtained by calculation j Replace y j In other words, y j ′ is for y j-p ,…,y j-1 ,y j ,y j+1 …y j+p The value y is obtained by averaging the production of the (2p+1)th gas field. j The calculation of ′ can be achieved based on the following formula 2:

[0086]

[0087] In an exemplary embodiment, the first gas field production sequence also includes first gas field production that does not meet the first condition. For any first gas field production, if there are no reference number of first gas field productions preceding and following that first gas field production as centered on that first gas field production, then that first gas field production is considered not to meet the first condition. The absence of reference number of first gas field productions preceding and following that first gas field production in the first gas field production sequence indicates that the number of first gas field productions preceding that first gas field production is less than the reference number, or the number of first gas field productions following that first gas field production is less than the reference number.

[0088] For example, if the production sequence of the first gas field is located at the first gas field production y j If the previous production from the first gas field was less than the reference quantity p, then y calculated based on Formula 3 is used. j Replace y j :

[0089]

[0090] For example, if the production sequence of the first gas field is located at the first gas field production y j If the production of the first gas field is less than the reference quantity p, then y calculated based on Formula 4 is used. j Replace y j :

[0091]

[0092] After smoothing the production sequence of the first gas field, the production sequence of the second gas field is obtained. For example, assume the production sequence of the first gas field is represented as (y1, y2, y3, ..., y...). n Then, the production sequence of the second gas field obtained after processing the first gas field production sequence using a smoothing method can be represented as (y′1, y′2, y′3, ..., y′). n Because the smoothing process does not remove data, the total output of the second gas field in the second gas field production sequence is the same as the total output of the first gas field in the first gas field production sequence. Furthermore, the smoothing process only changes the numerical value of the gas field output, not the corresponding time period.

[0093] Step 3: Process the production sequence of the second gas field using a standardization method to obtain the original gas field production sequence.

[0094] In one possible implementation, the standardization process for processing the second gas field production sequence is as follows: determine the minimum and maximum gas field production in the second gas field production sequence; for any second gas field production in the second gas field production sequence, standardize the production of any second gas field based on the minimum and maximum gas field production, and replace the production of any second gas field with the gas field production obtained after the standardization process.

[0095] In one possible implementation, the process of standardizing the production of any second gas field based on the minimum and maximum gas field production is carried out according to the following formula 5:

[0096]

[0097] Where, y′ v z′ represents the v-th processed data in the production sequence of the second gas field, where v is a positive integer not greater than n, and n is the total production of the second gas field in the production sequence; v This represents the production y′ of the second gas field based on the minimum and maximum gas field production. v The gas field production obtained after standardization; minA represents the minimum gas field production in the second gas field production sequence; maxA represents the maximum gas field production in the second gas field production sequence.

[0098] After processing the production sequence of the second gas field using a standardization method, the original gas field production sequence is obtained. For example, assume the production sequence of the second gas field is represented as (y′1, y′2, y′3, ..., y′...). n If the production sequence of the second gas field is processed using standardization, the original gas field production sequence can be represented as (z′1, z′2, z′3, ... z′). n Because the standardization process does not remove data, the total output of the second gas field in the second gas field production sequence is the same as the total output of the original gas field in the original gas field production sequence. Furthermore, the standardization process only changes the numerical value of the gas field output, not the corresponding time period.

[0099] It should be noted that this application embodiment only illustrates the preprocessing of the measured gas field production sequence based on three methods: outlier removal, smoothing, and standardization, with the execution order being outlier removal, smoothing, and standardization. This application embodiment is not limited to this. Regardless of the method, after preprocessing the measured gas field production sequence, the original gas field production sequence is obtained, and then step 202 is executed.

[0100] In step 202, based on the prediction parameters and the original gas field production sequence, a predicted gas field production sequence is obtained, and the predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence.

[0101] Prediction parameters refer to the parameters used in predicting gas field production. For example, prediction parameters are set based on experience. In setting prediction parameters, various factors that may affect gas field production can be considered to ensure that the predicted gas field production based on the set prediction parameters is closer to the actual situation. This application does not limit the factors considered in setting prediction parameters. For example, the factors considered in setting prediction parameters include at least one of reserve factors, economic level factors, and policy guidance factors. That is, prediction parameters are determined based on at least one of reserve factors, economic level factors, and policy guidance factors. Reserve factors, economic level factors, and policy guidance factors are all factors that can affect gas field production. Reserve factors refer to information about the gas field's reserves, economic level factors refer to information about the gas field's economic level, and policy guidance information refers to information about the gas field's policy guidance.

[0102] The prediction parameters may include one or more aspects of parameters, which influence different sub-processes in the process of predicting gas field production. For example, the process of predicting gas field production includes sub-processes of obtaining the accumulated gas field production sequence with equal time intervals required to construct the gas field production prediction model, sub-processes of constructing a gas field production prediction model matching the accumulated gas field production sequence, and sub-processes of predicting gas field production based on the gas field production prediction model. For example, the prediction parameters include model setting parameters. These model setting parameters influence the sub-processes of constructing a gas field production prediction model matching the accumulated gas field production sequence. For example, the model setting parameters include one or more model setting sub-parameters, which are not limited in this embodiment.

[0103] Exemplarily, the prediction parameters include model setting parameters and time interval adjustment parameters. The time interval adjustment parameters influence the sub-process of obtaining the accumulated gas field production sequence with equal time intervals required to construct the gas field production prediction model. Exemplarily, the time interval adjustment parameters include one or more time interval adjustment sub-parameters, which is not limited in this embodiment. Of course, in exemplary embodiments, the prediction parameters may also include other parameters.

[0104] After obtaining the prediction parameters and the original gas field production sequence, a predicted gas field production sequence is acquired. This sequence consists of multiple predicted gas field production figures arranged sequentially according to their corresponding time periods. Each predicted gas field production figure in the predicted production sequence corresponds one-to-one with the original gas field production figures in the original gas field production sequence. In other words, the number of predicted gas field production figures in the predicted production sequence is the same as the number of original gas field production figures in the original gas field production sequence. Furthermore, the time corresponding to a predicted gas field production figure at a specific position in the predicted production sequence is the same as the time corresponding to the original gas field production figure at the same position in the original gas field production sequence.

[0105] In one possible implementation, the prediction parameters include model setting parameters. Based on the prediction parameters and the original gas field production sequence, the process of obtaining the predicted gas field production sequence includes the following steps 1 to 4:

[0106] Step 1: Accumulate the original gas field production sequence to obtain the original accumulated gas field production sequence, which is an equal-time interval sequence.

[0107] The original gas field production sequence may have significant volatility and randomness. By accumulating the original gas field production sequence, it is beneficial to weaken the volatility and randomness between data, thereby obtaining an original cumulative gas field production sequence with more obvious regularity. Then, by constructing a gas field production prediction model that matches the original cumulative gas field production data sequence, the cumulative gas field production can be predicted more accurately.

[0108] For example, the gas field production prediction model used in this application embodiment is a grey model (GM). Specifically, for the case where there is only one original gas field production sequence, the grey model used is the GM(1,1) model. The GM(1,1) model represents the establishment of a model for a variable using a first-order differential equation.

[0109] The original cumulative gas field production sequence required for constructing the gas field production prediction model is an equal-time interval sequence to ensure the reliability of the constructed gas field production prediction model. An equal-time interval sequence means that the time interval between any two adjacent original cumulative gas field production values ​​in the original cumulative gas field production sequence is the same.

[0110] The original gas field production sequence may be an equal-time interval sequence or a non-equal-time interval sequence, and this application embodiment does not limit this. The method of accumulating the original gas field production sequence to obtain the original accumulated gas field production sequence differs depending on the different cases of the original gas field production sequence. That is, before performing step 1, it is necessary to first determine whether the original gas field production sequence is an equal-time interval sequence.

[0111] For example, the original gas field production sequence can be represented as X (0) =(X (0) (t1),…,X (0) (t a ),…,X (0) (t n Among them, X (0) Represents the original gas field production sequence; X (0) (t a ) represents time t in the original gas field production sequence. a The corresponding gas field production. The time series corresponding to the original gas field production sequence can be represented as T. (0) =(t1,…,t a ,…,t n ).

[0112] In a raw gas field production sequence, there is a time interval between every two adjacent raw gas field production values. After obtaining the raw gas field production sequence, it is possible to determine whether the sequence is an isochronous sequence based on the time interval between every two adjacent raw gas field production values. If the time interval between every two adjacent raw gas field production values ​​is equal to the same constant, then the raw gas field production sequence is determined to be an isochronous sequence; if the time interval between every two adjacent raw gas field production values ​​is not equal to the same constant, then the raw gas field production sequence is determined to be a non-isochronous sequence.

[0113] In one possible implementation, if the original gas field production sequence is an equal-time interval sequence, the original gas field production sequence is accumulated to obtain the original accumulated gas field production sequence by directly accumulating the production of each original gas field in the original gas field production sequence.

[0114] For example, if the original gas field production sequence is an isochronous sequence, then the original gas field production sequence can be represented as X. (0) =(X (0) (t1),X (0) (t2),…,X (0) (t n The time interval Δt between the production of two adjacent primary gas fields. k =t k -t k-1 All values ​​are equal to the same constant, where k = 2, 3, 4, ..., n. The original accumulated gas field production sequence is obtained by directly summing the production of each original gas field in the original gas field production sequence, denoted as X. (1) =(X (1) (t1),X (1)(t2),…,X (1) (t n In other words, the time corresponding to the production output of each original accumulated gas field in the original accumulated gas field production sequence is the same as the time corresponding to the production output of each original gas field in the original gas field production sequence. Since the original gas field production sequence is an isochronous sequence, the original accumulated gas field production sequence obtained in this way is also an isochronous sequence. Original accumulated gas field production sequence X (1) The production X of any original cumulative gas field in the data (1) (t k Based on formula 6, the following calculations were performed:

[0115]

[0116] If the original gas field production sequence is not a non-equal time interval sequence, then the time interval between any two adjacent original gas field production outputs in the original gas field production sequence is not equal to the same constant. This indicates that the original gas field production sequence is missing original gas field production outputs corresponding to certain times. If the production outputs of each original gas field in the original gas field production sequence are directly summed, an equal time interval summed gas field production sequence cannot be obtained.

[0117] In one possible implementation, the prediction parameters include not only model setting parameters but also time interval adjustment parameters. When the original gas field production sequence is not an isochronous sequence, the method for accumulating the original gas field production sequence to obtain the original accumulated gas field production sequence is as follows: In response to the original gas field production sequence being an isochronous sequence, based on the time interval adjustment parameters, the original gas field production sequence is accumulated to obtain the original accumulated gas field production sequence corresponding to the target time series. The target time series is an isochronous sequence that matches the time series corresponding to the original gas field production sequence.

[0118] The target time series has equal intervals between every two adjacent times. The target time series is an isochronous sequence that matches the time series corresponding to the original gas field production series. For example, the target time series is an isochronous sequence that fills in missing times in the time series corresponding to the original gas field production series. For example, the interval between any two times in the target time series is a preset prediction time step. For instance, if the time series corresponding to the original gas field production series is (2000, 2001, 2002, 2004), and the preset prediction time step is 1 year, then the target time series is (2000, 2001, 2002, 2003, 2004).

[0119] For example, the target time series is T (1) =(t1,t2,…,t m The interval Δt between any two adjacent time points k =tk -t k-1 This is equal to the same constant, where k = 2, 3, 4, ..., m (m is an integer not less than 2). For example, based on the time interval adjustment parameter, the original gas field production sequence X is... (0) =(X (0) (t1),…,X (0) (t a ),…,X (0) (t n The accumulation process is performed to obtain the original accumulated gas field production sequence X corresponding to the target time series. (1) =(X (1) (t1),X (1) (t2),…,X (1) (t m The process is based on Formula 7:

[0120]

[0121] Among them, t k Δt represents any time in the target time series. k =t k -t k-1 G and g are time interval adjustment parameters, both integer variables, with g less than G. The original cumulative gas field production sequence X is obtained based on Formula 7. (1) =(X (1) (t1),X (1) (t2),…,X (1) (t m In the process, first use the first formula in Formula 7 to obtain X. (1) (t1), and then use the second formula in Formula 7 to iterate and obtain X. (1) (t2),…,X (1) (t m ).

[0122] X is obtained by iteratively using the second formula in Formula 7. (1) (t2),…,X (1) (t m During the process, if the original gas field production sequence contains a time t k-1 The corresponding original gas field production, but time t is not available. k The corresponding original gas field production will be located at time t in the original gas field production sequence. k-1 The original gas field production value following the corresponding original gas field production value is used as the provisional X. (0) (t k ). Based on temporary X (0) (t k X was calculated. (1) (tk After that, X (1) (t k ) and X (1) (t k-1 The difference between the two values ​​is used as the estimated X. (0) (t k Add ) to the original gas field production sequence to ensure that time t exists in the original gas field production sequence. k-1 The corresponding original gas field production. In this way, not only can the original cumulative gas field production sequence X be obtained, but also... (1) =(X (1) (t1),X (1) (t2),…,X (1) (t m Furthermore, it can also obtain an isochronous sequence of raw gas field production X, which includes the raw gas field production at each time point in the target time series. (0)′ =(X (0) (t1),X (0) (t2),…,X (0) (t m The original gas field production sequence X of these time series (0)′ It can provide data support for constructing a gas field production prediction model that matches the original cumulative gas field production sequence.

[0123] Next, an example will be given to illustrate the process of obtaining the original cumulative gas field production sequence based on Formula 7, to facilitate understanding. Assume the original gas field production sequence is X. (0) =(X (0) (2000),X (0) (2001),X (0) (2002),X (0) (2004)), the target time series is T (1) = (2000, 2001, 2002, 2003, 2004), the original cumulative gas field production sequence to be obtained is X. (1) =(X (1) (2000),X (1) (2001),X (1) (2002),X (1) (2003),X (1) (2004)). The process of obtaining the original cumulative gas field production sequence is to sequentially obtain X. (1) (2000), X (1) (2001), X (1) (2002), X (1) (2003) and X (1) The process of (2004):

[0124] X (1)(2000)=X (0) (2000)

[0125]

[0126]

[0127]

[0128]

[0129] Among them, X (0) (2003) is X (1) (2003) and X (1) The difference from (2002). Through the above calculation process, the original cumulative gas field production sequence X can be obtained. (1) =(X (1) (2000),X (1) (2001),X (1) (2002),X (1) (2003),X (1) (2004)) can also obtain the original gas field production sequence X at equal time intervals. (0)′ =(X (0) (2000),X (0) (2001),X (0) (2002),X (0) (2003),X (0) (2004)).

[0130] In the process of obtaining the first-order accumulation formula required for the GM(1,1) model for non-equal time interval sequences, the related technology will reduce the time interval Δt. k All weights are assigned to X 0 (t k Compared to this approach, the embodiments of this application dynamically correct the original gas field production sequence with non-equidistant time intervals by considering time interval adjustment parameters (G and g), and adjust the time interval Δt. k The weights g / G are assigned to X (0) (t k-1 On the time interval Δt k The weight (Gg) / G is assigned to X (0) (t k This approach facilitates the acquisition of highly reliable raw cumulative gas field production sequences by updating G and g, thereby enabling the development of gas field production prediction models that can make more accurate predictions, and ultimately improving the accuracy of prediction results.

[0131] Step 2: Based on the model setting parameters, construct a gas field production prediction model that matches the original cumulative gas field production sequence.

[0132] After obtaining the original cumulative gas field production sequence, a gas field production prediction model matching the original cumulative gas field production sequence is constructed based on the model setting parameters. For example, the gas field production prediction model is a GM(1,1) model. The process of constructing the gas field production prediction model matching the original cumulative gas field production sequence based on the model setting parameters is implemented according to Equation 8:

[0133]

[0134] Among them, NewX (1) (t k+1 () represents the gas field production prediction model that matches the original cumulative gas field production sequence. Formula 8 is the implementation formula for the gas field production prediction model that matches the original cumulative gas field production sequence; time t k+1 The subscript k+1 represents time t. k+1 The gas field is located at position k+1 in the target time series. The gas field production prediction model is used to predict the time t at position k+1 in the target time series. k+1 In the process of collecting the corresponding cumulative gas field production data, the value of k is substituted into Formula 8 for prediction. That is, according to Formula 8, the predicted cumulative gas field production for each time point in the target time series, excluding the first time point, can be obtained. B1 and B2 represent the model setting parameters; X (0) (t1) represents the first raw output in the equal-time-interval raw output sequence; a and b are parameters calculated according to Formula 9. It should be noted that Formula 8 above is obtained by adding B1 and B2 to the implementation formula of the GM(1,1) prediction model. The derivation process of the implementation formula of the GM(1,1) prediction model will not be elaborated in this embodiment.

[0135]

[0136] Step 3: Obtain the predicted cumulative gas field production sequence based on the gas field production prediction model. The predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence.

[0137] The predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence. That is, the number of predicted cumulative gas field productions in the predicted cumulative gas field production sequence is the same as the number of original cumulative gas field productions in the original cumulative gas field production sequence, and the time corresponding to the predicted cumulative gas field production at a certain position in the predicted cumulative gas field production sequence is the same as the time corresponding to the original cumulative gas field production at the same position in the original cumulative gas field production sequence.

[0138] In one possible implementation, the process of obtaining the predicted cumulative gas field production sequence based on the gas field production prediction model is as follows: The first original cumulative gas field production in the original cumulative gas field production sequence is used as the predicted cumulative gas field production corresponding to the first time in the target time series; the predicted cumulative gas field production corresponding to each time in the target time series, excluding the first time, is predicted based on the gas field production prediction model; the predicted cumulative gas field production corresponding to the first time and the predicted cumulative gas field production corresponding to each time are arranged sequentially according to time to obtain the predicted cumulative gas field production sequence. For example, the gas field production prediction model predicts any time t in the target time series, excluding the first time. k+1 The corresponding method for predicting cumulative gas field production is as follows: Substitute the value of k into Formula 8 to calculate the production at any given time t. k+1 The corresponding predicted cumulative gas field production.

[0139] Step 4: Perform cumulative subtraction processing on the predicted cumulative gas field production sequence to obtain the predicted gas field production sequence.

[0140] After obtaining the predicted cumulative gas field production sequence, a cumulative subtraction process is performed on the predicted cumulative gas field production sequence to obtain the predicted gas field production sequence. For example, if the original gas field production sequence is an equal-time interval sequence, the gas field production sequence obtained after the cumulative subtraction process on the predicted cumulative gas field production sequence is directly used as the predicted gas field production sequence.

[0141] If the original gas field production sequence is not a time interval sequence, since the predicted cumulative gas field production sequence is a time interval sequence, after the cumulative subtraction processing of the predicted cumulative gas field production time series, the result is a time interval predicted gas field production sequence. It is necessary to extract the predicted gas field production that corresponds one-to-one with the original gas field production in the original gas field production sequence from the time interval predicted gas field production sequence to form a predicted gas field production sequence.

[0142] For example, the predicted cumulative gas field production sequence is denoted as NewX. (1) NewX (1) =(Newx (1) (t1),Newx (1) (t2),…,Newx (1) (tm Regarding NewX (1) After cumulative reduction processing, the NewX gas field production sequence with equal time intervals is obtained. (0) NewX (0) =(Newx (0) (t1),Newx (0) (t2),…,Newx (0) (t m If we set NewX (1) It's NewX (0) The cumulative generation sequence, then NewX (0) It's NewX (1) The restored sequence.

[0143] For example, if the original gas field production sequence is an isochronous sequence, then the NewX gas field production can be directly processed based on Equation 10. (1) Cumulative reduction processing:

[0144] NewX (0) (t k ) = NewX (1) (t k )-NewX (0) (t k-1 ) (Formula 10)

[0145] For example, if the original gas field production sequence is a non-uniform time interval sequence, then the NewX gas field production sequence can be processed based on Equation 11. (1) Cumulative reduction processing:

[0146]

[0147] It should be noted that steps 1 to 4 above are illustrated using a set of prediction parameters. This application does not limit the number of prediction parameters; that is, there can be multiple sets of prediction parameters. For cases with multiple sets of prediction parameters, a predicted gas field production sequence can be obtained based on each set of prediction parameters and the original gas field production sequence. In other words, multiple predicted gas field production sequences can be obtained simultaneously based on multiple sets of prediction parameters.

[0148] For example, the process of obtaining a predicted gas field production sequence is as follows: Figure 3 As shown. The measured gas field production data were collected as x1, x2, x3, x4, ..., x. q A measured gas field production sequence is constructed. The measured gas field production sequence is preprocessed using at least one of outlier removal, smoothing, and standardization to obtain the original gas field production sequence. Based on the target time series and the time interval adjustment parameters (G and g) in the prediction parameters, the original gas field production sequence with non-equidistant time intervals is corrected to obtain the original cumulative gas field production sequence X.(1) Based on the model setting parameters (B1 and B2) in the prediction parameters, a gas field production prediction model matching the original gas field production sequence is constructed. Based on the gas field production prediction model, the predicted gas field production sequence is obtained.

[0149] In step 203, in response to the fact that the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, the prediction parameters on which the current predicted gas field production sequence is obtained are updated to obtain new prediction parameters; based on the new prediction parameters and the original gas field production sequence, a new predicted gas field production sequence is obtained; this process is repeated until the current predicted gas field production sequence meets the similarity condition with the original gas field production sequence, and the target prediction parameters are obtained based on the prediction parameters on which the current predicted gas field production sequence is obtained.

[0150] The predicted gas field production sequence is the sequence of predicted gas field production. This predicted sequence may be very close to the original gas field production sequence, or it may differ significantly, depending on the effectiveness of the prediction parameters. The process of obtaining the target prediction parameters is a dynamic adjustment process, which is a cyclical process. The current predicted gas field production sequence is obtained based on the latest prediction parameters. After each iteration of obtaining the predicted gas field production sequence based on the prediction parameters and the original gas field production sequence, it is determined whether the current predicted gas field production sequence and the original gas field production sequence meet the similarity criteria. If they do, the prediction parameters used to obtain the current predicted gas field production sequence are updated to obtain new prediction parameters. Then, based on the new prediction parameters and the original gas field production sequence, a new predicted gas field production sequence is obtained.

[0151] If the current predicted gas field production sequence and the original gas field production sequence meet the similarity condition, the loop process ends, the target prediction parameters are obtained based on the prediction parameters on which the current predicted gas field production sequence is obtained, and then step 204 is executed.

[0152] It should be noted that, for the case where multiple predicted gas field production sequences are obtained based on multiple sets of prediction parameters in each loop, the current number of predicted gas field production sequences is multiple. The similarity condition between the current predicted gas field production sequence and the original gas field production sequence means that at least one of the multiple predicted gas field production sequences satisfies the similarity condition with the original gas field production sequence.

[0153] The similarity criteria are set based on experience or can be flexibly adjusted according to the application scenario; this application does not limit this. For example, two gas field production sequences satisfying the similarity criteria means that the similarity between the two gas field production sequences is not less than a similarity threshold. The similarity threshold is set based on experience; for example, the similarity threshold is 90%. This application does not limit the method for calculating the similarity between two gas field production sequences. For example, the similarity between the two gas field production sequences is obtained by calculating the Euclidean distance between them.

[0154] If the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, it indicates a significant difference between the two, meaning the reliability of the prediction parameters used to obtain the current predicted gas field production is poor. In this case, the prediction parameters used to obtain the current predicted gas field production are updated to obtain new prediction parameters. Then, based on the new prediction parameters and the original gas field production sequence, a new predicted gas field production sequence is obtained.

[0155] In an exemplary embodiment, if the number of prediction parameters used to obtain the current predicted gas field production is a set, the set of prediction parameters is directly updated to obtain the updated prediction parameters.

[0156] In an exemplary embodiment, when there are multiple sets of prediction parameters used to obtain the current predicted gas field production, the process of updating these prediction parameters to obtain new prediction parameters involves updating each set of prediction parameters individually to obtain new prediction parameters. In this approach, the number of new prediction parameters is the same as the number of original prediction parameters.

[0157] In an exemplary embodiment, when there are multiple sets of prediction parameters used to obtain the current predicted gas field production, the process of updating these prediction parameters to obtain new prediction parameters involves updating the prediction parameters that meet the selection criteria from the multiple sets of prediction parameters to obtain new prediction parameters. In this approach, the number of new prediction parameters may be the same as or different from the original prediction parameters. For example, prediction parameters that meet the selection criteria refer to prediction parameters whose similarity between the obtained predicted gas field production sequence and the original gas field production sequence is not less than a first threshold. The first threshold is set empirically and is different from the similarity threshold; for example, the first threshold is 70%.

[0158] The embodiments of this application do not limit the method of updating the prediction parameters, as long as it enables the acquisition of a predicted gas field production sequence that is closer to the original gas field production sequence based on the updated prediction parameters.

[0159] This application embodiment uses the example of a single predicted gas field production sequence, that is, a set of prediction parameters used to obtain the current predicted gas field production sequence. In the exemplary embodiment, the prediction parameters used to obtain the current predicted gas field production sequence are updated to obtain new prediction parameters by: obtaining the update magnitude of the prediction parameters; and updating the prediction parameters based on the update magnitude. The update magnitude is determined manually or by the computer equipment itself, and this application embodiment does not limit this.

[0160] In an exemplary embodiment, the method for updating the prediction parameters used to obtain the current predicted gas field production sequence to obtain new prediction parameters is as follows: The variation pattern of predicted gas field production with the prediction parameters is obtained; based on this variation pattern and the difference between the current predicted gas field production sequence and the original gas field production sequence, the prediction parameters used to obtain the current predicted gas field production sequence are updated to obtain new prediction parameters. The variation pattern of predicted gas field production with the prediction parameters is used to indicate how the prediction parameters specifically affect the change in predicted gas field production. This variation pattern can be determined manually or automatically by computer equipment; this embodiment does not limit this approach.

[0161] For example, the variation of predicted gas field production with prediction parameters can refer to the variation of predicted gas field production with prediction parameters for all times in the time series corresponding to the original gas field production sequence, or it can refer to the variation of predicted gas field production with prediction parameters for a portion of the time series corresponding to the original gas field production sequence. This application does not limit this. For example, the variation of predicted gas field production with prediction parameters for any given time can indicate the variation of predicted gas field production with respect to all parameters in the prediction parameters for that given time, or it can refer to the variation of predicted gas field production with respect to a portion of the prediction parameters for that given time. This application does not limit this.

[0162] For example, the method for obtaining the variation pattern of predicted gas field production at any given time with any two parameters in the prediction parameters is as follows: Based on the predicted gas field production at any given time in the historically obtained predicted gas field production sequence, and the two parameters in the prediction parameters on which the predicted gas field production sequence is based, a three-dimensional surface is constructed; the variation pattern of predicted gas field production at any given time with any two parameters in the prediction parameters is analyzed based on the three-dimensional surface. For example, taking time t1 as an example, the two parameters in the prediction parameters are B1 and B2. Based on the predicted gas field production at time t1 and the B1 and B2 on which the predicted gas field production at t1 is based, a three-dimensional surface can be obtained. Based on this three-dimensional surface, it can be analyzed that the predicted gas field production at time t1 increases with the increase of B1 and decreases with the increase of B2.

[0163] After obtaining the variation pattern of predicted gas field production with prediction parameters, the prediction parameters used to obtain the current predicted gas field production sequence are updated based on this variation pattern and the difference between the current predicted gas field production sequence and the original gas field production sequence, resulting in new prediction parameters. Based on the difference between the current and original gas field production sequences, it becomes clear how to modify the predicted gas field production in the current sequence to narrow the gap with the original sequence. Furthermore, based on the variation pattern of predicted gas field production with prediction parameters, it becomes clear how to update the prediction parameters to achieve the necessary changes in the predicted gas field production in the current sequence. This approach allows the predicted gas field production sequence obtained based on the new prediction parameters to more closely approximate the original gas field production sequence.

[0164] For example, if the difference between the current predicted gas field production sequence and the original gas field production sequence indicates that the predicted gas field production at time t1 in the current predicted gas field production sequence is less than the original gas field production at time t1 in the original gas field production sequence, and the predicted gas field production at time t1 changes with the prediction parameters B1 and B2 as follows: the predicted gas field production at time t1 increases with increasing B1 and decreases with increasing B2. In this case, it is necessary to increase B1 and decrease B2.

[0165] After obtaining the new prediction parameters, a new predicted gas field production sequence is obtained based on the new prediction parameters and the original gas field production sequence. The implementation principle of this process is the same as that of step 202, and will not be repeated here.

[0166] If the current predicted gas field production sequence and the original gas field production sequence meet the similarity condition, it indicates that the reliability of the prediction parameters on which the current predicted gas field production sequence is obtained is high. In this case, the target prediction parameters are obtained based on the prediction parameters on which the current predicted gas field production sequence is obtained.

[0167] For example, if the number of prediction parameters used to obtain the current predicted gas field production sequence is one set, then the prediction parameters used to obtain the current predicted gas field production sequence are directly used as the target prediction parameters; if the number of prediction parameters used to obtain the current predicted gas field production sequence is multiple sets, then the prediction parameters that satisfy the target conditions among the prediction parameters used to obtain the current predicted gas field production sequence are used as the target prediction parameters. Here, satisfying the target conditions means obtaining the prediction parameters used in the current predicted gas field production sequence that have the highest similarity to the original gas field production sequence.

[0168] For example, after obtaining the target prediction parameters, the predicted gas field production sequence obtained based on the target prediction parameters and the original gas field production sequence is used as the target predicted gas field production sequence, which has a high degree of agreement with the original gas field production sequence.

[0169] For example, the process of obtaining the target predicted gas field production sequence is as follows: Figure 4 As shown, the process involves: preprocessing the measured gas field production sequence to obtain the original gas field production sequence; acquiring prediction parameters and a target time series; constructing a gas field production prediction model based on the preset prediction parameters, the target time series obtained through preset time steps, and the original gas field production sequence; obtaining the predicted gas field production sequence based on the gas field production prediction model; determining whether the predicted gas field production sequence and the original gas field production sequence meet similarity conditions; if the predicted gas field production sequence and the original gas field production sequence do not meet similarity conditions, updating the prediction parameters to obtain new prediction parameters; then returning to the step of constructing the gas field production prediction model; if the predicted gas field production sequence and the original gas field production sequence meet similarity conditions, acquiring the target prediction parameters and using the predicted gas field production sequence obtained based on the target prediction parameters and the original gas field production sequence as the target predicted gas field production sequence.

[0170] In step 204, the gas field production corresponding to the reference time is predicted based on the target prediction parameters and the original gas field production sequence.

[0171] The reference times are set based on experience or can be flexibly adjusted according to application output. There can be one or more reference times. For example, the reference times include, but are not limited to, the missing times in the time series corresponding to the original gas field production sequence, and a specified number of times following the last time in the time series corresponding to the original gas field production sequence. The specified number is set based on experience or can be flexibly adjusted according to the application scenario; for example, the specified number may be 5.

[0172] In one possible implementation, taking a reference time as an example, the process of predicting the gas field production corresponding to the reference time based on the target prediction parameters and the original gas field production sequence is as follows: The original gas field production sequence is accumulated to obtain an original accumulated gas field production sequence, which is an isochronous sequence; based on the target model setting parameters in the target prediction parameters, a target gas field production prediction model matching the original accumulated gas field production sequence is constructed; the first predicted accumulated gas field production corresponding to the reference time is predicted based on the target gas field production prediction model; the second predicted accumulated gas field production corresponding to a time preceding the reference time and having a specified time interval from the reference time is predicted based on the target gas field production prediction model; the difference between the first predicted accumulated gas field production and the second predicted gas field production is taken as the gas field production corresponding to the reference time. The specified time interval refers to the time interval between any two times corresponding to the original accumulated gas field production in the original accumulated gas field production sequence.

[0173] For example, suppose there are 12 original gas field production sequences, with the sequence {400, 580, 650, 730, 990, 1200, 1400, 1600, 1700, 2300, 2700, 4000}. The corresponding time series is {0, 3, 5, 8, 11, 15, 20, 24, 30, 37, 41, 47}, and the time interval (time interval) Δt between adjacent times in the time series is {0, 3, 2, 3, 3, 4, 5, 4, 6, 7, 4, 6}. In other words, the original gas field production sequence is a non-equidistant time interval sequence. The curve constructed based on the original gas field production sequence is shown below. Figure 5 As shown.

[0174] Using the gas field production prediction method provided in this application, the gas field production at each time point in the time series corresponding to the original gas field production sequence, as well as at times 50, 54, and 58 (i.e., reference times), were predicted. A comparison chart of the curves constructed based on the gas field production at the reference times and the curves constructed based on the original gas field production sequence is shown below. Figure 6 As shown. In Figure 6 In the diagram, the rhombus represents the original gas field production in the original gas field production sequence, and the square represents the gas field production at the reference time. According to... Figure 6It can be seen that the gas field production at each time point in the predicted original gas field production sequence is in good agreement with the original gas field production, with a small error probability of over 95%. Furthermore, the predicted gas field production at times 50, 54, and 58 is consistent with the changing trend of gas field production, indicating a high accuracy in the predicted gas field production.

[0175] In this embodiment, the gas field production corresponding to the reference time is predicted based on target prediction parameters and the original gas field production sequence. The target prediction parameters are obtained by using prediction parameters that can yield a predicted gas field production sequence that meets similar conditions to the original gas field production sequence. In other words, the target prediction parameters can yield a predicted gas field production sequence that is relatively close to the original gas field production sequence, thus making the gas field production corresponding to the reference time predicted based on the target prediction parameters more accurate. This is beneficial for ensuring the accuracy and rationality of long-term gas field production planning.

[0176] See Figure 7 This application provides a gas field production prediction device, which includes:

[0177] The first acquisition unit 701 is used to acquire the original gas field production sequence, which is composed of multiple original gas field productions arranged in chronological order according to their corresponding time.

[0178] The second acquisition unit 702 is used to acquire a predicted gas field production sequence based on the prediction parameters and the original gas field production sequence, wherein the predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence.

[0179] The update unit 703 is used to update the prediction parameters on which the current predicted gas field production sequence is based when the original gas field production sequence does not meet the similarity condition, so as to obtain new prediction parameters.

[0180] The third acquisition unit 704 is used to acquire a new predicted gas field production sequence based on the new prediction parameters and the original gas field production sequence; this process is repeated until the current predicted gas field production sequence and the original gas field production sequence meet the similarity condition, and the target prediction parameters are acquired based on the prediction parameters on which the current predicted gas field production sequence is acquired.

[0181] Prediction unit 705 is used to predict the gas field production at a reference time based on the target prediction parameters and the original gas field production sequence.

[0182] In one possible implementation, the prediction parameters include model setting parameters; the second acquisition unit 702 is used to perform cumulative processing on the original gas field production sequence to obtain an original cumulative gas field production sequence, which is an equal-time interval sequence; based on the model setting parameters, a gas field production prediction model matching the original cumulative gas field production sequence is constructed; based on the gas field production prediction model, a predicted cumulative gas field production sequence is obtained, where the predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence; and the predicted cumulative gas field production sequence is subjected to cumulative subtraction processing to obtain a predicted gas field production sequence.

[0183] In one possible implementation, the prediction parameters also include time interval adjustment parameters; the second acquisition unit 702 is further configured to, in response to the original gas field production sequence being a non-equidistant time interval sequence, perform cumulative processing on the original gas field production sequence based on the time interval adjustment parameters to obtain the original cumulative gas field production sequence corresponding to the target time sequence, wherein the target time sequence is an equidistant time interval sequence that matches the time sequence corresponding to the original gas field production sequence.

[0184] In one possible implementation, the second acquisition unit 702 is further configured to take the first original cumulative gas field production in the original cumulative gas field production sequence as the predicted cumulative gas field production corresponding to the first time in the target time series; predict the predicted cumulative gas field production corresponding to each time in the target time series except for the first time based on the gas field production prediction model; and arrange the predicted cumulative gas field production corresponding to the first time and the predicted cumulative gas field production corresponding to each time in chronological order to obtain the predicted cumulative gas field production sequence.

[0185] In one possible implementation, the update unit 703 is used to obtain the variation law of the predicted gas field production with the prediction parameters; based on the variation law of the predicted gas field production with the prediction parameters and the difference between the current predicted gas field production sequence and the original gas field production sequence, the prediction parameters on which the current predicted gas field production sequence is based are updated to obtain new prediction parameters.

[0186] In one possible implementation, the first acquisition unit 701 is used to acquire the measured gas field production sequence; and to preprocess the measured gas field production sequence to obtain the original gas field production sequence.

[0187] In one possible implementation, the first acquisition unit 701 is further configured to preprocess the measured gas field production sequence based on at least one of outlier removal, smoothing, and standardization.

[0188] In one possible implementation, the forecasting parameters are determined based on at least one of reserves, economic level, and policy orientation factors.

[0189] In this embodiment, the gas field production corresponding to the reference time is predicted based on target prediction parameters and the original gas field production sequence. The target prediction parameters are obtained by using prediction parameters that can yield a predicted gas field production sequence that meets similar conditions to the original gas field production sequence. In other words, the target prediction parameters can yield a predicted gas field production sequence that is relatively close to the original gas field production sequence, thus making the gas field production corresponding to the reference time predicted based on the target prediction parameters more accurate. This is beneficial for ensuring the accuracy and rationality of long-term gas field production planning.

[0190] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional units when implementing its functions. In practical applications, the above functions can be assigned to different functional units as needed, that is, the internal structure of the apparatus can be divided into different functional units to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0191] Exemplary, this application provides a computer software system for implementing the above-described gas field production prediction method. For example... Figure 8 As shown, the computer software system includes a data acquisition module, a preprocessing module, a dynamic correction module, a prediction calculation module, and a calling module.

[0192] The data acquisition module is used to collect historical production data from the gas field to be predicted, forming a measured gas field production sequence. The module supports importing tables (e.g., Excel spreadsheets), data (e.g., DAT data), and various database interfaces, allowing the measured gas field production data to be imported into the system's built-in database. After importing the data, a measured gas field production data curve can be generated to verify the accuracy of the imported data. For example, the data acquisition module also has a data export function, allowing data to be exported from the software's built-in database, for example, exporting data from the software's built-in database in tabular form. For example, the data acquisition module also has a real-time monitoring data input interface, which can be used to revise the gas field production prediction model using new measured data.

[0193] The preprocessing module is used to preprocess the acquired measured gas field production sequence. The preprocessing module's functions include, but are not limited to, outlier removal, smoothing, and standardization. For example, confidence intervals can be set within the preprocessing module. For example, the preprocessed raw gas field production sequence is stored in a built-in database and has an independent curve display function, allowing for comparison with the measured gas field production sequence.

[0194] The dynamic correction module is used to perform unequal time interval correction on the raw gas field production sequence generated by the data processing module, generating the X required for the GM(1,1) model. (1) The system collects a series of data and stores them in a built-in database. The dynamic correction module has the function of adjusting the time interval and setting the time interval adjustment parameters G and g. It also has an independent curve display function and can be compared with the original gas field production sequence.

[0195] The prediction calculation module is used to calculate the aforementioned X. (1) The system performs predictive calculations and analysis on a series of data to generate a predicted gas field production sequence, which is then stored in a built-in database. The prediction calculation module has a parameter scanning function. It sets the value range of constants for parameters such as the prediction time step and model setting parameters B1 and B2, performs a parameterized scan, compares the prediction results with the original data, and determines the most ideal prediction parameters. This module also has a function scanning function, which uses e... x The system scans functions such as normal distribution and Gaussian distribution to determine the predicted gas field production sequence that matches the original gas field production sequence.

[0196] The module employs a multi-parameter coupled partial differential equation algorithm. It sets prediction parameters influenced by factors such as reserves, economic levels, and policy guidance, and programs the gas field production prediction model to generate a multi-parameter prediction 3D surface. By taking a longitudinal profile in any given direction, the resulting curve is the prediction curve under a specific parameter combination, thus realizing multi-parameter scanning back analysis functionality. The user interface includes functions such as raw data retrieval, parameter identification, prediction step size setting, prediction curve generation, and data download and transmission.

[0197] For example, the aforementioned computer software system is deployed on a cloud-based computer device. For example, the cloud-based computer device includes a big data cloud storage and a cloud processor. Access to the cloud platform's gas field production dynamic correction prediction system is achieved via remote login. The program runs calculations on the cloud processor to implement the gas field production prediction method described above. Remote printing of prediction reports and data downloading are also possible. For example, an SMS alert function is set up on the cloud platform; unauthorized users will be promptly reported to monitoring personnel via SMS.

[0198] In an exemplary embodiment, a computer device is also provided, see [link to example]. Figure 9 The computer device includes a processor 901 and a memory 902, the memory 902 storing at least one computer program. The at least one computer program is loaded and executed by one or more processors 901 to enable the computer device to implement any of the aforementioned gas field production prediction methods. The computer device can be a terminal or a server.

[0199] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores at least one computer program that is loaded and executed by a processor of a computer device to enable the computer to implement any of the above-described methods for predicting gas field production.

[0200] In one possible implementation, the aforementioned computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0201] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the gas field production prediction methods described above.

[0202] It should be noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the above exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0203] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0204] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for predicting gas field production, characterized in that, The method includes: Obtain the original gas field production sequence, which is composed of multiple original gas field productions arranged in chronological order according to their corresponding time periods; In response to the fact that the original gas field production sequence is a non-equal time interval sequence, the original gas field production sequence is accumulated based on the time interval adjustment parameter to obtain the original accumulated gas field production sequence corresponding to the target time sequence. The target time sequence is an equal time interval sequence that fills in the missing time in the time sequence corresponding to the original gas field production sequence. The original accumulated gas field production sequence is an equal time interval sequence. Based on the model setting parameters, a gas field production prediction model matching the original cumulative gas field production sequence is constructed. Based on the gas field production prediction model, a predicted cumulative gas field production sequence is obtained, and the predicted cumulative gas field production in the predicted cumulative gas field production sequence corresponds one-to-one with the original cumulative gas field production in the original cumulative gas field production sequence. The predicted cumulative gas field production sequence is subjected to cumulative subtraction processing to obtain a predicted gas field production sequence. The predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence. In response to the fact that the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, the prediction parameters on which the current predicted gas field production sequence is obtained are updated to obtain new prediction parameters; based on the new prediction parameters and the original gas field production sequence, a new predicted gas field production sequence is obtained; this process is repeated until the current predicted gas field production sequence meets the similarity condition with the original gas field production sequence, and then the target prediction parameters are obtained based on the prediction parameters on which the current predicted gas field production sequence is obtained; Based on the target prediction parameters and the original gas field production sequence, predict the gas field production corresponding to the reference time. The process of accumulating the original gas field production sequence based on the time interval adjustment parameter to obtain the original accumulated gas field production sequence corresponding to the target time sequence is implemented based on the following formula 7: (Official 7) The original gas field production sequence is as follows: The original cumulative gas field production sequence is The For any time in the target time series; G and g are time interval adjustment parameters, and G and g are integer variables, with g being less than G; The process of constructing a gas field production prediction model that matches the original cumulative gas field production sequence based on model setting parameters is implemented according to the following formula 8: (Official 8) The This represents a gas field production prediction model that matches the original cumulative gas field production sequence. When the original gas field production sequence is a non-equal time interval sequence, the cumulative gas field production sequence is predicted based on the following formula 11. Perform cumulative reduction processing: (Official 11).

2. The method according to claim 1, characterized in that, The step of obtaining the predicted cumulative gas field production sequence based on the gas field production prediction model includes: The first original cumulative gas field production in the original cumulative gas field production sequence is taken as the predicted cumulative gas field production corresponding to the first time in the target time series; Based on the gas field production prediction model, predict the cumulative gas field production corresponding to each time in the target time series except for the first time. The predicted cumulative gas field production corresponding to the first time and the predicted cumulative gas field production corresponding to each time are arranged in chronological order to obtain the predicted cumulative gas field production sequence.

3. The method according to claim 1 or 2, characterized in that, The step of updating the prediction parameters on which the current predicted gas field production sequence is based to obtain new prediction parameters includes: To obtain the variation pattern of predicted gas field production with prediction parameters; Based on the variation of the predicted gas field production with the predicted parameters and the difference between the current predicted gas field production sequence and the original gas field production sequence, the predicted parameters on which the current predicted gas field production sequence is obtained are updated to obtain new predicted parameters.

4. The method according to claim 1 or 2, characterized in that, The acquisition of the original gas field production sequence includes: Obtain the measured gas field production sequence; The measured gas field production sequence is preprocessed to obtain the original gas field production sequence.

5. The method according to claim 4, characterized in that, The preprocessing of the measured gas field production sequence includes: The measured gas field production sequence is preprocessed using at least one of outlier removal, smoothing, and standardization.

6. The method according to claim 1 or 2, characterized in that, The prediction parameters are determined based on at least one of the following factors: reserves, economic level, and policy guidance.

7. A gas field production prediction device, characterized in that, The device includes: The first acquisition unit is used to acquire the original gas field production sequence, which is composed of multiple original gas field productions arranged in chronological order according to their corresponding time periods. The second acquisition unit is configured to, in response to the fact that the original gas field production sequence is a non-equal-time sequence, perform accumulation processing on the original gas field production sequence based on time interval adjustment parameters to obtain the original accumulated gas field production sequence corresponding to the target time sequence, wherein the target time sequence is an equal-time sequence with missing times filled in the time sequence corresponding to the original gas field production sequence, and the original accumulated gas field production sequence is an equal-time sequence; construct a gas field production prediction model matching the original accumulated gas field production sequence based on model setting parameters; obtain a predicted accumulated gas field production sequence based on the gas field production prediction model, wherein the predicted accumulated gas field production in the predicted accumulated gas field production sequence corresponds one-to-one with the original accumulated gas field production in the original accumulated gas field production sequence; and perform cumulative subtraction processing on the predicted accumulated gas field production sequence to obtain a predicted gas field production sequence, wherein the predicted gas field production in the predicted gas field production sequence corresponds one-to-one with the original gas field production in the original gas field production sequence. The update unit is used to update the prediction parameters on which the current predicted gas field production sequence is based in response to the fact that the current predicted gas field production sequence does not meet the similarity condition with the original gas field production sequence, so as to obtain new prediction parameters. The third acquisition unit is used to acquire a new predicted gas field production sequence based on the new prediction parameters and the original gas field production sequence; this process is repeated until the current predicted gas field production sequence and the original gas field production sequence meet similarity conditions, and the target prediction parameters are acquired based on the prediction parameters on which the current predicted gas field production sequence is acquired. The prediction unit is used to predict the gas field production at a reference time based on the target prediction parameters and the original gas field production sequence. The process of accumulating the original gas field production sequence based on the time interval adjustment parameter to obtain the original accumulated gas field production sequence corresponding to the target time sequence is implemented based on the following formula 7: (Official 7) The original gas field production sequence The original cumulative gas field production sequence The For any time in the target time series; G and g are time interval adjustment parameters, and G and g are integer variables, with g being less than G; The process of constructing a gas field production prediction model that matches the original cumulative gas field production sequence based on model setting parameters is implemented according to the following formula 8: (Official 8) The This represents a gas field production prediction model that matches the original cumulative gas field production sequence. When the original gas field production sequence is a non-equal time interval sequence, the cumulative gas field production sequence is predicted based on the following formula 11. Perform cumulative reduction processing: (Official 11).

8. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one computer program, which is loaded and executed by the processor to enable the computer device to implement the gas field production prediction method as described in any one of claims 1 to 6.

Images