A method, device and computer equipment for generating gas production profile of coalbed methane well

Abstract

The present invention relates to the field of coalbed methane production capacity prediction, and discloses a coalbed methane well gas production profile generation method, device and computer equipment, which can obtain the static characteristic data of the coalbed methane well and the dynamic characteristic data in the first gas production period, pre-process the dynamic characteristic data in the first gas production period to obtain the pre-processed characteristic data, and construct the corresponding parameter value time series data according to the pre-processed characteristic data, input the parameter value time series data and the static characteristic data into the trained gas production prediction model to perform gas production time series prediction, and obtain the gas production data of the second gas production period predicted by the model, and the number of time steps of the drainage stage. According to the number of time steps of the drainage stage and the gas production data of each time step in the second gas production period, the gas production profile of the coalbed methane well is generated, and the intelligent generation of the coalbed methane well gas production profile is effectively realized, which improves the intelligent level of coalbed methane well gas production profile generation.

Description

Technical Field

[0001] The present invention relates to the field of coalbed methane production capacity prediction, and in particular to a method, device and computer equipment for generating a gas production profile of a coalbed methane well. Background Art

[0002] Coalbed methane is a clean energy with abundant reserves. The development and application of coalbed methane can reduce greenhouse gas emissions and improve coal mine production safety.

[0003] The gas production profile of coalbed methane wells is very important for coalbed methane development. It can be used to evaluate the production potential of coalbed methane wells and to optimize coalbed methane development planning schemes.

[0004] However, the relevant technology is generally used by technicians to determine the gas production profile of coalbed methane wells based on the historical gas production of the coalbed methane wells and subjective experience, and the level of intelligence is relatively low. Summary of the invention

[0005] The present invention provides a method, device and computer equipment for generating a gas production profile of a coalbed methane well, so as to solve the defect in the related technology that the gas production profile of a coalbed methane well is generally determined by technicians based on the historical gas production of the coalbed methane well and subjective experience, and the intelligence level is low, thereby improving the intelligence level of generating the gas production profile of a coalbed methane well.

[0006] In a first aspect, the present invention provides a method for generating a gas production profile of a coalbed methane well, comprising: Obtaining static characteristic data of the coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period; Preprocessing each of the target parameter values ​​respectively to obtain a processed parameter value corresponding to each of the target parameter values; According to each of the processed parameter values, construct corresponding parameter value time series data; Inputting the parameter value time series data and the static characteristic data into a trained gas production prediction model to perform gas production time series prediction, so as to obtain a predicted value of the gas production of the coalbed methane well at each time step in the second gas production period, and to obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period; Based on the number of time steps and the predicted value of gas production in each time step in the second gas production period, a gas production profile of the coalbed methane well is generated.

[0007] Optionally, the dynamic production parameters include at least one target production parameter, and the preprocessing is noise reduction processing; The preprocessing of each target parameter value to obtain a processed parameter value corresponding to each target parameter value comprises: For the target parameter value of any target production parameter in the Nth time step in the first gas production period, determine the first N time steps and the last N time steps adjacent to the Nth time step, obtain the parameter values ​​of the target production parameter in the first N time steps and the last N time steps, respectively, calculate the corresponding parameter mean according to the parameter values ​​of the target production parameter in the first N time steps, the last N time steps and the target parameter value, and input the parameter mean into the filter so that the filter performs noise reduction processing based on the parameter mean and a preset formula to obtain the corresponding processed parameter value; wherein the Nth time step is any time step in the first gas production period.

[0008] Optionally, the preset formula is:

[0009] in, is the parameter value after processing, is the mean value of the parameter, is the noise reduction order, , and is the matrix parameter of the fitting operator.

[0010] Optionally, when the dynamic production parameter includes a target production parameter, constructing corresponding parameter value time series data according to each of the processed parameter values ​​includes: For the Nth time step in the first gas production period, setting an order to be arranged equal to N for the processed parameter values ​​corresponding to the Nth time step; All the processed parameter values ​​are arranged based on the order to be arranged of all the processed parameter values ​​to obtain the parameter value time series data.

[0011] Optionally, when the dynamic production parameters include multiple target production parameters, constructing corresponding parameter value time series data according to each of the processed parameter values ​​includes: For the Nth time step in the first gas production period, based on the set parameter arrangement order, all the processed parameter values ​​corresponding to the Nth time step are sorted to obtain a parameter value subsequence corresponding to the Nth time step, and an order to be arranged is set for the parameter value subsequence corresponding to the Nth time step, and the order to be arranged is the same as N; All the parameter value subsequences are arranged based on the order to be arranged of all the parameter value subsequences to obtain the parameter value time series data.

[0012] Optionally, generating the gas production profile of the coalbed methane well based on the number of time steps and the predicted value of gas production at each time step in the second gas production period includes: Setting the gas production of the coalbed methane well at each time step in the time step number to 0; Determine the gas production of the coalbed methane well at each time step from a first time step to a second time step, wherein the first time step is the first time step in the gas production stage, and the second time step is the last time step in the first gas production period; According to the gas production of the coalbed methane well in each time step in the time step number, the gas production in each time step from the first time step to the second time step and the predicted value of gas production in each time step in the second gas production period, a curve of gas production changing with time step is drawn and used as the gas production profile.

[0013] Optionally, the training process of the gas production prediction model includes: Obtain static characteristic data and dynamic production data of sample coalbed methane wells; Performing data cleaning on the dynamic production data of the sample coalbed methane well to obtain cleaned production data; Dividing the post-cleaning production data into drainage stage data and gas production stage data, and determining corresponding drainage days according to the drainage stage data; In the gas production stage data, sample parameter values ​​of dynamic production parameters for each day in the first sample gas production period are obtained; Preprocessing each of the sample parameter values ​​respectively to obtain a processed sample parameter value corresponding to each of the sample parameter values; According to each of the processed parameter values, construct corresponding sample parameter value time series data; Input the sample parameter value time series data and the static characteristic data of the sample coalbed methane well into the gas production prediction model to be trained to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the sample coalbed methane well every day in the second sample gas production period, and obtain the predicted value of the number of drainage days; wherein the second sample gas production period is adjacent to the first sample gas production period and is after the first sample gas production period; Based on the difference between the predicted value of drainage days and the drainage days, and the difference between the predicted value of gas production per day in the second sample gas production period and the actual daily gas production, the gas production prediction model to be trained is updated to obtain the trained gas production prediction model.

[0014] Optionally, the static characteristic data includes geological characteristic data and / or engineering characteristic data, the geological characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume, and the engineering characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume; The at least one target production parameter includes at least one of gas production, water production, bottom hole flow pressure and casing pressure.

[0015] In a second aspect, the present invention provides a coalbed methane well gas production profile generation device, comprising: An acquisition unit, used for acquiring static characteristic data of the coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period; A preprocessing unit, used to preprocess each of the target parameter values ​​respectively to obtain a processed parameter value corresponding to each of the target parameter values; A construction unit, used to construct corresponding parameter value time series data according to each of the processed parameter values; A prediction unit, used for inputting the parameter value time series data and the static characteristic data into a trained gas production prediction model to perform gas production time series prediction, so as to obtain a predicted value of gas production of the coalbed methane well at each time step in a second gas production period, and to obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period; A generating unit is used to generate a gas production profile of the coalbed methane well based on the number of time steps and the predicted value of gas production in each time step in the second gas production period.

[0016] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for generating a gas production profile of a coalbed methane well according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0017] The coalbed methane well gas production profile generation method, device and computer equipment provided by the present invention can obtain the static characteristic data of the coalbed methane well and the dynamic characteristic data in the first gas production period, pre-process the dynamic characteristic data in the first gas production period to obtain the pre-processed characteristic data, and construct the corresponding parameter value time series data according to the pre-processed characteristic data, input the parameter value time series data and the static characteristic data into the trained gas production prediction model to perform gas production time series prediction, and obtain the gas production data of the second gas production period predicted by the model, and the number of time steps of the drainage stage. According to the number of time steps of the drainage stage and the gas production data of each time step in the second gas production period, the gas production profile of the coalbed methane well is generated, which effectively realizes the intelligent generation of the coalbed methane well gas production profile, and improves the intelligent level of coalbed methane well gas production profile generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a method for generating a gas production profile of a coalbed methane well provided in an embodiment of the present invention; Figure 2 A parameter value variation curve of a target production parameter provided by an embodiment of the present invention; Figure 3 A parameter value variation curve of another target production parameter provided by an embodiment of the present invention; Figure 4 A parameter value variation curve of a target production parameter after noise reduction provided by an embodiment of the present invention; Figure 5 A parameter value variation curve after noise reduction of another target production parameter provided by an embodiment of the present invention; Figure 6 A schematic diagram of the structure of a gas production prediction model provided by an embodiment of the present invention; Figure 7 A schematic diagram of a value range of a model structure hyperparameter provided in an embodiment of the present invention; Figure 8 A schematic diagram of a curve showing a change in daily gas production of a coalbed methane well predicted by a model provided in an embodiment of the present invention; Fig. 9 A schematic structural diagram of a coalbed methane well gas production profile generating device provided by an embodiment of the present invention; Fig.10 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Combine the following Figure 1-Figure 8 The method for generating gas production profile of coalbed methane well according to the present invention is described.

[0022] like Figure 1 As shown, this embodiment proposes a first method for generating a gas production profile of a coalbed methane well, which may include the following steps: S101. Obtain static characteristic data of a coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period.

[0023] Specifically, the static characteristic data may be certain parameter values ​​that do not change during the coalbed methane production process of the coalbed methane well, and the dynamic production parameters may be production parameters whose parameter values ​​may change during the coalbed methane production process of the coalbed methane well.

[0024] Optionally, the static characteristic data includes geological characteristic data and / or engineering characteristic data, the geological characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume, and the engineering characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume; The at least one target production parameter includes at least one of gas production, water production, bottom hole flow pressure and casing pressure.

[0025] The time step may be a specific time interval, such as one day, two days, or 20 hours.

[0026] It should be noted that the production process of coalbed methane wells has the production characteristics of drainage and gas production. The production process can be specifically divided into a drainage stage and a gas production stage. The gas production in each time step in the drainage stage is 0.

[0027] Specifically, the first gas production period is a certain production period of the coalbed methane well in the gas production stage, such as from the first day to the 100th day of the gas production stage.

[0028] Specifically, the target parameter value of the dynamic production parameter is the parameter value of the dynamic production parameter at a certain time step in the first gas production period. Figure 2 and Figure 3The target production parameters shown are parameter value variation curves of daily gas production, daily water production, casing pressure and bottom hole flowing pressure.

[0029] S102: Preprocess each target parameter value respectively to obtain a processed parameter value corresponding to each target parameter value.

[0030] Specifically, this embodiment can preprocess the target parameter value of each target production parameter in the dynamic production parameters in the first gas production period to reduce the impact of production fluctuations and human interference on the parameter value and improve the accuracy of subsequent coalbed methane well gas production prediction.

[0031] Optionally, in other coalbed methane well gas production profile generation methods proposed in this embodiment, the dynamic production parameters include at least one target production parameter, and the preprocessing is noise reduction processing. In this case, the above step S102 may include: For the target parameter value of any target production parameter in the Nth time step in the first gas production period, determine the first N time steps and the last N time steps adjacent to the Nth time step, obtain the parameter values ​​of the target production parameter in the first N time steps and the last N time steps, calculate the corresponding parameter mean according to the parameter values ​​of the target production parameter in the first N time steps and the last N time steps and the target parameter value, and input the parameter mean into the filter so that the filter performs noise reduction processing based on the parameter mean and a preset formula to obtain the corresponding processed parameter value; wherein the Nth time step is any time step in the first gas production period.

[0032] The value range of N may be from 1 to the number of time steps in the first gas production period, and N may be any value in the value range.

[0033] It can be understood that, in this embodiment, the target parameter value of each target production parameter at each time step in the first gas production period can be subjected to noise reduction processing.

[0034] Specifically, the filter may be an SG filter or other types of filters, which is not limited in this embodiment.

[0035] Optionally, the preset formula is:

[0036] in, is the parameter value after processing, is the parameter mean, is the noise reduction order, , and is the matrix parameter of the fitting operator.

[0037] This embodiment Figure 2 and Figure 3The processed parameter values ​​obtained by denoising the parameter values ​​in can be found in Figure 4 and Figure 5 The parameter value change curve after noise reduction processing is shown. Figure 4 and Figure 5 The parameter value change curves after noise reduction processing are shown when the target production parameters are daily gas production, daily water production, casing pressure and bottom hole flow pressure.

[0038] S103: Construct corresponding parameter value time series data according to each processed parameter value.

[0039] Specifically, in this embodiment, all processed parameter values ​​may be used to construct parameter value time series data. It is understandable that the parameter value time series data includes all processed parameter values.

[0040] Optionally, when the dynamic production parameters include a target production parameter, step S103 may include: For the Nth time step in the first gas production period, setting an order to be arranged equal to N for the processed parameter values ​​corresponding to the Nth time step; All processed parameter values ​​are arranged based on the order to be arranged of all processed parameter values ​​to obtain parameter value time series data.

[0041] It can be understood that when the dynamic production parameters include only one target production parameter, the processed parameter value corresponding to the Nth time step is the processed parameter value obtained after denoising the target parameter value corresponding to the Nth time step in the first gas production period in this embodiment.

[0042] Specifically, this embodiment can set the corresponding order to be arranged for the processed parameter values ​​corresponding to each time step, for example, the order to be arranged for the processed parameter values ​​corresponding to the first time step is set to 1, and the order to be arranged for the processed parameter values ​​corresponding to the second time step is set to 2.

[0043] Specifically, in this embodiment, the processed parameter values ​​may be arranged according to the order in which the processed parameter values ​​are to be arranged, so as to obtain the parameter value time series data.

[0044] Optionally, when the dynamic production parameters include multiple target production parameters, step S103 may include: For the Nth time step in the first gas production period, all processed parameter values ​​corresponding to the Nth time step are sorted based on the set parameter arrangement order to obtain a parameter value subsequence corresponding to the Nth time step, and an order to be arranged is set for the parameter value subsequence corresponding to the Nth time step, and the order to be arranged is the same as N; Based on the to-be-arranged order of all parameter value subsequences, all parameter value subsequences are arranged to obtain parameter value time series data.

[0045] It can be understood that when the dynamic production parameters include multiple target production parameters, the Nth time step corresponds to multiple processed parameter values, and all processed parameter values ​​corresponding to the Nth time step include the target parameter values ​​corresponding to the Nth time step in the first gas production period in this embodiment after noise reduction processing. For example, all processed parameter values ​​corresponding to the first time step include the target parameter values ​​corresponding to the first time step in the first gas production period in this embodiment after noise reduction processing, and for another example, all processed parameter values ​​corresponding to the second time step include the target parameter values ​​corresponding to the second time step in the first gas production period in this embodiment after noise reduction processing.

[0046] Among them, the parameter arrangement order can be set by technical personnel according to actual conditions. For example, when the dynamic production parameters include gas production, water production, bottom hole flow pressure and casing pressure, the parameter arrangement order can be gas production, water production, bottom hole flow pressure and casing pressure, or casing pressure, bottom hole flow pressure, water production and gas production.

[0047] Specifically, this embodiment can sort all processed parameter values ​​corresponding to the same time step according to the set parameter arrangement order to obtain a parameter value subsequence corresponding to the time step. For example, this embodiment can sort all processed parameter values ​​corresponding to the first time step according to the set parameter arrangement order to obtain a parameter value subsequence corresponding to the first time step. For another example, this embodiment can sort all processed parameter values ​​corresponding to the second time step according to the set parameter arrangement order to obtain a parameter value subsequence corresponding to the second time step.

[0048] Specifically, this embodiment can set corresponding to-be-arranged orders for parameter value subsequences corresponding to different time steps. For example, this embodiment sets to-be-arranged order 1 for the parameter value subsequence corresponding to the first time step, and sets to-be-arranged order 2 for the parameter value subsequence corresponding to the second time step. Afterwards, all parameter value subsequences are sorted according to the to-be-arranged orders of different parameter value subsequences to obtain parameter value time series data.

[0049] S104. Input the parameter value time series data and static feature data into the trained gas production prediction model to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the coalbed methane well at each time step in the second gas production period, and obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period.

[0050] Among them, the trained gas production prediction model can be used to predict gas production related data of coalbed methane wells.

[0051] Specifically, the time step number corresponding to the drainage stage of the coalbed methane well is the time step number that the drainage stage lasts.

[0052] Specifically, when the time step is one day, this embodiment can input the parameter value time series data and the static characteristic data of the coalbed methane well into the trained gas production prediction model, so that the gas production prediction model can predict the drainage days of the coalbed methane well and the daily gas production of the coalbed methane well in the second gas production period according to the parameter value time series data and the static characteristic data, and obtain the predicted drainage days and daily gas production in the second gas production period.

[0053] S105. Generate a gas production profile of the coalbed methane well based on the number of time steps and the predicted value of gas production in each time step in the second gas production period.

[0054] Specifically, this embodiment can draw the gas production profile of the coalbed methane well based on the number of time steps predicted by the gas production prediction model and the predicted value of the gas production in each time step in the second gas production period.

[0055] Optionally, step S105 may include: Set the gas production of the coalbed methane well in each time step to 0; Determine the gas production of the coalbed methane well at each time step from the first time step to the second time step, the first time step is the first time step in the gas production stage, and the second time step is the last time step in the first gas production period; According to the gas production of the coalbed methane well in each time step in the time step number, the gas production in each time step from the first time step to the second time step, and the predicted value of gas production in each time step in the second gas production period, a curve of gas production changing with time step is drawn and used as the gas production profile.

[0056] It should be noted that the gas production profile of a coalbed methane well is a curve showing the change of gas production over time. The formation process of a coalbed methane well is that the drainage stage appears first, and then enters the gas production stage.

[0057] Among them, the time step predicted by the gas production prediction model is the drainage stage of the coalbed methane well, and the gas production of each time step within this time step is 0.

[0058] Optionally, when the first gas production period starts from the first time step in the gas production stage, the first gas production period is from the first time step to the second time step. At this time, the gas production of each time step in the first gas production period is the gas production of each time step from the first time step to the second time step. At this time, if the dynamic production parameter includes the gas production, the target parameter value of the dynamic production parameter at each time step in the first gas production period obtained in step S101 of this embodiment includes the gas production of each time step from the first time step to the second time step, and the gas production can be directly determined without executing the step of obtaining the gas production.

[0059] Optionally, if the dynamic production parameters do not include gas production, this embodiment also needs to obtain the gas production of the coalbed methane well in each time step from the first time step to the second time step.

[0060] Specifically, when drawing the gas production profile, the present embodiment can first draw a curve according to the gas production in the drainage stage in the coordinate system where the gas production changes with the time step, that is, first draw a curve where the gas production is 0 in the time steps of the drainage stage, and then draw the curve where the gas production corresponding to the first gas production period and the second gas production period changes with the time step. The curve drawn at last is the gas production profile of the coalbed methane well.

[0061] The coalbed methane well gas production profile generation method proposed in this embodiment can obtain the static characteristic data of the coalbed methane well and the dynamic characteristic data in the first gas production period, pre-process the dynamic characteristic data in the first gas production period to obtain the pre-processed characteristic data, and construct the corresponding parameter value time series data according to the pre-processed characteristic data, input the parameter value time series data and the static characteristic data into the trained gas production prediction model to perform gas production time series prediction, and obtain the gas production data of the second gas production period predicted by the model, as well as the number of time steps of the drainage stage. According to the number of time steps of the drainage stage and the gas production data of each time step in the second gas production period, the gas production profile of the coalbed methane well is generated, which effectively realizes the intelligent generation of the coalbed methane well gas production profile, and improves the intelligent level of coalbed methane well gas production profile generation.

[0062] based on Figure 1 This embodiment proposes a second method for generating a gas production profile of a coalbed methane well. In this method, the training process of the gas production prediction model includes: Obtain static characteristic data and dynamic production data of sample coalbed methane wells; Perform data cleaning on the dynamic production data of sample coalbed methane wells to obtain cleaned production data; Divide the post-cleaning production data into drainage stage data and gas production stage data, and determine the corresponding drainage days based on the drainage stage data; In the gas production stage data, sample parameter values ​​of the dynamic production parameters for each day in the first sample gas production period are obtained; Preprocess each sample parameter value respectively to obtain a processed sample parameter value corresponding to each sample parameter value; According to each processed parameter value, construct the corresponding sample parameter value time series data; Input the sample parameter value time series data and the static characteristic data of the sample coalbed methane well into the gas production prediction model to be trained to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the sample coalbed methane well every day in the second sample gas production period, and obtain the predicted value of the number of drainage days; wherein the second sample gas production period is adjacent to the first sample gas production period and is after the first sample gas production period; Based on the difference between the predicted drainage days and the drainage days, and the difference between the predicted gas production value of each day in the second sample gas production period and the actual daily gas production, the gas production prediction model to be trained is updated to obtain a trained gas production prediction model.

[0063] Specifically, this embodiment can first collect the geological, engineering parameters and production parameters of existing coalbed methane wells. Data cleaning is performed on all data, including data outlier deletion and vacancy value supplementation.

[28] The coalbed methane production parameters after data cleaning are divided into two parts, one part is the coalbed methane drainage stage data, and the other part is the coalbed methane production stage data, and the number of production days in the drainage stage is counted.

[0064] Afterwards, the sample parameter values ​​of the dynamic production parameters for each day in the first sample gas production period are extracted from the coalbed methane production stage data, and then preprocessing such as noise reduction is performed to obtain the processed parameter values ​​of the dynamic production parameters.

[0065] Specifically, in this embodiment, a gas production prediction model to be trained may be designed in advance. Figure 6 As shown, the model includes a dynamic feature data processing layer, a static feature data processing layer, a merging layer and an output layer.

[0066] Among them, the dynamic feature data processing layer includes a dynamic input layer, a convolution layer and a recurrent neural network layer, the convolution layer can be a convolutional neural network (CNN), and the recurrent neural network layer can be a long short-term memory network (LSTM). In this embodiment, the coalbed methane well production data, that is, the processed parameter values ​​of the dynamic production parameters, can be input into the dynamic feature data processing layer, and the dynamic input layer, the convolution layer and the recurrent neural network layer can process them in sequence.

[0067] The static feature data processing layer includes a static input layer, a convolution layer and a static fully connected layer. The convolution layer may be a CNN. In this embodiment, static feature data such as parameter values ​​of formation parameters and engineering parameters may be input into the static feature data processing layer, and processed in sequence by the static input layer, the convolution layer and the static fully connected layer.

[0068] Specifically, the merging layer can be a fully connected neural network layer. The merging layer and the output layer are used to splice and predict the output data of the dynamic feature data processing layer and the static feature data processing layer, and output the number of time steps of the drainage stage and the predicted value of the daily gas production in the second sample gas production period. After that, the gas production profile is drawn.

[0069] Specifically, the value range of the model structure hyperparameters can be as follows: Figure 7 As shown, Figure 7 The value ranges of hyperparameters such as lag order, time step, number of parallel batches, number of convolutional layer output channels, and number of linear units are shown in Figure 2.

[0070] Specifically, this embodiment can respectively input the sorted coalbed methane well geological and engineering parameters and the daily gas production data after noise reduction in the corresponding neural network layer, and use the sliding window method to train the model. The optimal hyperparameter combination of the recurrent neural network model that can input dynamic and static parameters is determined and saved locally using the grid search method and the root mean square error algorithm. By calling the recurrent neural network model that can input dynamic and static parameters saved locally, the gas production of the coalbed methane well can be predicted at any time. The gas production prediction results can include the number of drainage days (i.e., drainage days) and the daily gas production data of the gas production stage. See Figure 8 . Figure 8 The horizontal axis is the production days of the sample coalbed methane well, the vertical axis is the daily gas production, and the two curves are the curves of the actual daily gas production of the sample coalbed methane well changing with the number of days, and the curves of the daily gas production predicted by the model changing with the number of days.

[0071] In the related technologies, the LSTM model is the most commonly used oil and gas prediction model. However, due to the unique production characteristics of drainage and gas production in coalbed methane, when the general LSTM model inputs the single well pressure or the historical daily gas production to predict the gas production profile of the coalbed methane well, it is difficult to learn and capture the production characteristics of the long-term zero gas production in the drainage stage. In view of the special production characteristics of drainage and gas production in coalbed methane, and only inputting the single well pressure or the historical daily gas production, the single long short-term memory neural network is not ideal for predicting the coalbed methane production profile, and even the predicted gas production in the drainage stage is negative.

[0072] This embodiment takes into account the development characteristics of coalbed methane drainage and gas production, and improves the accuracy of the neural network model in learning the coalbed methane production profile. It also takes into account the influence of coalbed methane geology and engineering parameters on the coalbed methane production profile, and improves the accuracy and interpretability of the neural network model.

[0073] The coalbed methane well gas production profile generation method proposed in this embodiment can reduce the data characteristics of the coalbed methane drainage stage by dividing the production stage and using the SG filter algorithm to reduce the noise of the coalbed methane production data, and reduce the large-scale impact of the dynamic fluctuations of production during multiple discontinuous drainage on the training process of the recurrent neural network model. It is also possible to input the geological and engineering parameters of the coalbed methane well into the model to ensure that the neural network model learns the nonlinear relationship of static parameters while learning the production relationship of the coalbed methane well, and more accurately predict the gas production profile of the coalbed methane well during the actual production process. This embodiment fully considers the drainage characteristics of the coalbed methane well, and can effectively improve the ability of the model to learn and predict the gas production profile of the coalbed methane well.

[0074] like Fig. 9 As shown, this embodiment provides a device for generating a gas production profile of a coalbed methane well, which may include: An acquisition unit 901 is used to acquire static characteristic data of the coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period; A preprocessing unit 902 is used to preprocess each target parameter value to obtain a processed parameter value corresponding to each target parameter value; A construction unit 903, configured to construct corresponding parameter value time series data according to each processed parameter value; The prediction unit 904 is used to input the parameter value time series data and the static feature data into the trained gas production prediction model to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the coalbed methane well at each time step in the second gas production period, and obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period; The generating unit 905 is used to generate the gas production profile of the coalbed methane well based on the number of time steps and the predicted value of the gas production in each time step in the second gas production period.

[0075] It should be noted that the processing of the acquisition unit 901, the preprocessing unit 902, the construction unit 903, the prediction unit 904 and the generation unit 905 and the beneficial effects thereof can be respectively referred to in Figure 1 Steps S101 to S105 in the above are not described in detail.

[0076] Optionally, the dynamic production parameters include at least one target production parameter, and the preprocessing is noise reduction processing; the preprocessing unit 902 is further used to: For the target parameter value of any target production parameter in the Nth time step in the first gas production period, determine the first N time steps and the last N time steps adjacent to the Nth time step, obtain the parameter values ​​of the target production parameter in the first N time steps and the last N time steps, calculate the corresponding parameter mean according to the parameter values ​​of the target production parameter in the first N time steps and the last N time steps and the target parameter value, and input the parameter mean into the filter so that the filter performs noise reduction processing based on the parameter mean and a preset formula to obtain the corresponding processed parameter value; wherein the Nth time step is any time step in the first gas production period.

[0077] Optionally, the preset formula is:

[0078] in, is the parameter value after processing, is the parameter mean, is the noise reduction order, , and is the matrix parameter of the fitting operator.

[0079] Optionally, when the dynamic production parameters include a target production parameter, the construction unit 903 is further configured to: For the Nth time step in the first gas production period, setting an order to be arranged equal to N for the processed parameter values ​​corresponding to the Nth time step; All processed parameter values ​​are arranged based on the order to be arranged of all processed parameter values ​​to obtain parameter value time series data.

[0080] Optionally, when the dynamic production parameters include multiple target production parameters, the construction unit 903 is further configured to: For the Nth time step in the first gas production period, all processed parameter values ​​corresponding to the Nth time step are sorted based on the set parameter arrangement order to obtain a parameter value subsequence corresponding to the Nth time step, and an order to be arranged is set for the parameter value subsequence corresponding to the Nth time step, and the order to be arranged is the same as N; Based on the to-be-arranged order of all parameter value subsequences, all parameter value subsequences are arranged to obtain parameter value time series data.

[0081] Optionally, the generating unit 905 is further configured to: Set the gas production of the coalbed methane well in each time step to 0; Determine the gas production of the coalbed methane well at each time step from the first time step to the second time step, the first time step is the first time step in the gas production stage, and the second time step is the last time step in the first gas production period; According to the gas production of the coalbed methane well in each time step in the time step number, the gas production in each time step from the first time step to the second time step, and the predicted value of gas production in each time step in the second gas production period, a curve of gas production changing with time step is drawn and used as the gas production profile.

[0082] Optionally, the training process of the gas production prediction model is set as follows: Obtain static characteristic data and dynamic production data of sample coalbed methane wells; Perform data cleaning on the dynamic production data of sample coalbed methane wells to obtain cleaned production data; Divide the post-cleaning production data into drainage stage data and gas production stage data, and determine the corresponding drainage days based on the drainage stage data; In the gas production stage data, sample parameter values ​​of the dynamic production parameters for each day in the first sample gas production period are obtained; Preprocess each sample parameter value respectively to obtain a processed sample parameter value corresponding to each sample parameter value; According to each processed parameter value, construct the corresponding sample parameter value time series data; Input the sample parameter value time series data and the static characteristic data of the sample coalbed methane well into the gas production prediction model to be trained to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the sample coalbed methane well every day in the second sample gas production period, and obtain the predicted value of the number of drainage days; wherein the second sample gas production period is adjacent to the first sample gas production period and is after the first sample gas production period; Based on the difference between the predicted drainage days and the drainage days, and the difference between the predicted gas production value of each day in the second sample gas production period and the actual daily gas production, the gas production prediction model to be trained is updated to obtain a trained gas production prediction model.

[0083] Optionally, the static characteristic data includes geological characteristic data and / or engineering characteristic data, the geological characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume, and the engineering characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume; The at least one target production parameter includes at least one of gas production, water production, bottom hole flow pressure and casing pressure.

[0084] The coalbed methane well gas production profile generation device proposed in this embodiment can obtain the static characteristic data of the coalbed methane well and the dynamic characteristic data in the first gas production period, pre-process the dynamic characteristic data in the first gas production period to obtain the pre-processed characteristic data, and construct the corresponding parameter value time series data according to the pre-processed characteristic data, input the parameter value time series data and the static characteristic data into the trained gas production prediction model to perform gas production time series prediction, and obtain the gas production data of the second gas production period predicted by the model, as well as the number of time steps of the drainage stage. According to the number of time steps of the drainage stage and the gas production data of each time step in the second gas production period, the gas production profile of the coalbed methane well is generated, which effectively realizes the intelligent generation of the coalbed methane well gas production profile, and improves the intelligent level of coalbed methane well gas production profile generation.

[0085] The coalbed methane well gas production profile generation device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0086] The embodiment of the present invention also provides a computer device having the above Fig. 9 The coalbed methane well gas production profile generating device shown.

[0087] See also Fig.10 , a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig.10 A processor 10 is taken as an example.

[0088] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0089] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0090] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function. The data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0091] The memory 20 may include a volatile memory, such as a random access memory. The memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive. The memory 20 may also include a combination of the above-mentioned types of memory.

[0092] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0093] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a gas production profile of a coalbed methane well, characterized in that: include: Obtaining static characteristic data of the coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period; Preprocessing each of the target parameter values ​​respectively to obtain a processed parameter value corresponding to each of the target parameter values; According to each of the processed parameter values, construct corresponding parameter value time series data; Inputting the parameter value time series data and the static characteristic data into a trained gas production prediction model to perform gas production time series prediction, so as to obtain a predicted value of the gas production of the coalbed methane well at each time step in the second gas production period, and to obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period; Based on the number of time steps and the predicted value of gas production in each time step in the second gas production period, a gas production profile of the coalbed methane well is generated.

2. The method according to claim 1, characterized in that The dynamic production parameters include at least one target production parameter, and the preprocessing is noise reduction processing; The preprocessing of each target parameter value to obtain a processed parameter value corresponding to each target parameter value comprises: For the target parameter value of any target production parameter in the Nth time step in the first gas production period, determine the first N time steps and the last N time steps adjacent to the Nth time step, obtain the parameter values ​​of the target production parameter in the first N time steps and the last N time steps, respectively, calculate the corresponding parameter mean according to the parameter values ​​of the target production parameter in the first N time steps, the last N time steps and the target parameter value, and input the parameter mean into the filter so that the filter performs noise reduction processing based on the parameter mean and a preset formula to obtain the corresponding processed parameter value; wherein the Nth time step is any time step in the first gas production period.

3. The method according to claim 2, characterized in that The preset formula is: in, is the parameter value after processing, is the mean value of the parameter, is the noise reduction order, , and is the matrix parameter of the fitting operator.

4. The method according to claim 2, characterized in that: When the dynamic production parameters include a target production parameter, the corresponding parameter value time series data is constructed according to each of the processed parameter values, including: For the Nth time step in the first gas production period, setting an order to be arranged equal to N for the processed parameter values ​​corresponding to the Nth time step; All the processed parameter values ​​are arranged based on the order to be arranged of all the processed parameter values ​​to obtain the parameter value time series data.

5. The method according to claim 2, characterized in that: When the dynamic production parameters include multiple target production parameters, constructing corresponding parameter value time series data according to each processed parameter value includes: For the Nth time step in the first gas production period, based on the set parameter arrangement order, all the processed parameter values ​​corresponding to the Nth time step are sorted to obtain a parameter value subsequence corresponding to the Nth time step, and an order to be arranged is set for the parameter value subsequence corresponding to the Nth time step, and the order to be arranged is the same as N; All the parameter value subsequences are arranged based on the order to be arranged of all the parameter value subsequences to obtain the parameter value time series data.

6. The method according to claim 1, characterized in that The generating of the gas production profile of the coalbed methane well based on the number of time steps and the predicted value of the gas production in each time step in the second gas production period comprises: Setting the gas production of the coalbed methane well at each time step in the time step number to 0; Determine the gas production of the coalbed methane well at each time step from a first time step to a second time step, wherein the first time step is the first time step in the gas production stage, and the second time step is the last time step in the first gas production period; According to the gas production of the coalbed methane well in each time step in the time step number, the gas production in each time step from the first time step to the second time step and the predicted value of gas production in each time step in the second gas production period, a curve of gas production changing with time step is drawn and used as the gas production profile.

7. The method according to claim 1, characterized in that The training process of the gas production prediction model includes: Obtain static characteristic data and dynamic production data of sample coalbed methane wells; Performing data cleaning on the dynamic production data of the sample coalbed methane well to obtain cleaned production data; Dividing the post-cleaning production data into drainage stage data and gas production stage data, and determining corresponding drainage days according to the drainage stage data; In the gas production stage data, sample parameter values ​​of dynamic production parameters for each day in the first sample gas production period are obtained; Preprocessing each of the sample parameter values ​​respectively to obtain a processed sample parameter value corresponding to each of the sample parameter values; According to each of the processed parameter values, construct corresponding sample parameter value time series data; Input the sample parameter value time series data and the static characteristic data of the sample coalbed methane well into the gas production prediction model to be trained to perform gas production time series prediction, so as to obtain the predicted value of the gas production of the sample coalbed methane well every day in the second sample gas production period, and obtain the predicted value of the number of drainage days; wherein the second sample gas production period is adjacent to the first sample gas production period and is after the first sample gas production period; Based on the difference between the predicted value of drainage days and the drainage days, and the difference between the predicted value of gas production per day in the second sample gas production period and the actual daily gas production, the gas production prediction model to be trained is updated to obtain the trained gas production prediction model.

8. The method according to claim 2, characterized in that: The static characteristic data includes geological characteristic data and / or engineering characteristic data, wherein the geological characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume, and the engineering characteristic data includes at least one of displacement, fracturing fluid volume, and fracturing sand volume; The at least one target production parameter includes at least one of gas production, water production, bottom hole flow pressure and casing pressure.

9. A device for generating a gas production profile of a coalbed methane well, characterized in that: include: An acquisition unit, used for acquiring static characteristic data of the coalbed methane well and target parameter values ​​of dynamic production parameters at each time step in the first gas production period; A preprocessing unit, used to preprocess each of the target parameter values ​​respectively to obtain a processed parameter value corresponding to each of the target parameter values; A construction unit, used to construct corresponding parameter value time series data according to each of the processed parameter values; A prediction unit, used for inputting the parameter value time series data and the static characteristic data into a trained gas production prediction model to perform gas production time series prediction, so as to obtain a predicted value of gas production of the coalbed methane well at each time step in a second gas production period, and to obtain the number of time steps corresponding to the drainage stage of the coalbed methane well; wherein the second period is adjacent to the first period and is after the first period; A generating unit is used to generate a gas production profile of the coalbed methane well based on the number of time steps and the predicted value of gas production in each time step in the second gas production period.

10. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for generating a gas production profile of a coalbed methane well as described in any one of claims 1 to 8 by executing the computer instructions.

Images