CN116502772A - Gas well daily gas production prediction method, determination device, electronic equipment and storage medium - Google Patents
Gas well daily gas production prediction method, determination device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116502772A CN116502772A CN202310750976.XA CN202310750976A CN116502772A CN 116502772 A CN116502772 A CN 116502772A CN 202310750976 A CN202310750976 A CN 202310750976A CN 116502772 A CN116502772 A CN 116502772A
- Authority
- CN
- China
- Prior art keywords
- gas well
- gas
- daily
- sparrow
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000003860 storage Methods 0.000 title claims abstract description 7
- 241000287127 Passeridae Species 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000012360 testing method Methods 0.000 claims abstract description 26
- 238000010845 search algorithm Methods 0.000 claims abstract description 19
- 238000004140 cleaning Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 10
- 238000009826 distribution Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 9
- 238000004364 calculation method Methods 0.000 abstract description 2
- 239000007789 gas Substances 0.000 description 124
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 6
- 239000003345 natural gas Substances 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Sampling And Sample Adjustment (AREA)
Abstract
The invention relates to the technical field of oil and gas field development, and provides a method for predicting daily gas yield of a gas well, a determining device, electronic equipment and a storage medium, wherein the method comprises the following steps of: s100, acquiring data required by daily gas production prediction of a gas well, establishing a data set and cleaning the data; s200, dividing a data set into a training set and a testing set, determining initial super parameters of a model, and establishing an LSTM gas well daily gas production prediction model; s300, optimizing a daily gas yield prediction model super parameter of the LSTM gas well by utilizing a sparrow search algorithm, and establishing a daily gas yield prediction model of the LSTM gas well by utilizing the sparrow search algorithm; s400, inputting data of the test set into a sparrow search algorithm to optimize a daily gas production prediction model of the LSTM gas well, so as to obtain a daily gas production prediction result of the gas well; the invention has the advantages of short time and high calculation precision.
Description
Technical Field
The invention relates to the technical field of oil and gas field development, in particular to a method for predicting daily gas production of a gas well, a determining device, electronic equipment and a storage medium.
Background
The natural gas reserves in China are rich, and the method has good development prospect. Accelerating the development of natural gas is beneficial to clean transformation of energy sources in China, and is a necessary choice for the natural gas industry. Before gas reservoir development, the prediction of the well-done yield has important guiding significance for the gas reservoir development, and the method can not only check the exploration effect, but also is beneficial to the development, deployment and planning of the well-done gas reservoir.
Currently, methods for predicting daily gas production of a gas well mainly comprise three methods, namely an empirical formula method, an analytic method and a numerical simulation method. The complex seepage characteristics of the gas reservoir are difficult to consider by the empirical formula and the analysis model, and the application conditions and the application stages of different models are different, so that the prediction result and the actual reservoir characteristic have larger difference. The gas well numerical simulation has the advantages of large related calculated amount, difficult history fitting, low yield prediction efficiency and large result uncertainty, so the gas well numerical simulation has general field popularization.
In summary, the conventional daily gas production prediction method has the defects of insufficient data feature extraction, limited application conditions and the like, and a new thought needs to be provided for the insufficient daily gas production prediction. Machine learning and deep learning are widely applied to various fields, and good effects are achieved. The deep neural network relies on a large number of data sets and a deep network structure, so that the characteristics of the data can be fully learned, and the prediction result has higher credibility.
Disclosure of Invention
The invention aims at: in order to solve the problems that the conventional method is low in daily gas production efficiency and poor in accuracy in determining the daily gas production of the gas well, the formulation of a next development scheme and the application condition of a process are affected, a daily gas production prediction model of the gas well for optimizing LSTM by a sparrow search algorithm is established on the basis of a machine learning principle based on production historical data of the production well, and the high-efficiency and accurate prediction of the daily gas production of the gas well is realized.
In order to achieve the above purpose, the invention provides a method for predicting daily gas production of a gas well, a determining device, electronic equipment and a storage medium, which are realized by adopting the following technical scheme:
the invention provides a method for predicting daily gas production of a gas well, which comprises the following steps:
s100, acquiring seven parameters of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield required by gas well daily gas yield prediction and gas well daily gas yield corresponding to the seven parameters, establishing a data set, and cleaning the data set;
s200, dividing a data set into a training set and a testing set, determining an initial learning rate, iteration times and hidden layer node number of the model, establishing an LSTM gas well daily gas yield prediction model, training the established LSTM gas well daily gas yield prediction model by using training set data, and testing prediction accuracy of the LSTM gas well daily gas yield prediction model by using data of the testing set;
s300, optimizing three super parameters of a learning rate, iteration times and hidden layer node number of a daily gas yield prediction model of the LSTM gas well by utilizing a sparrow search algorithm, continuously training and testing the model by using data of a training set and a testing set in the optimization process, keeping model parameters corresponding to the optimal model training, and then training the model by utilizing the obtained optimal model parameters and the data of the training set to establish the daily gas yield prediction model of the LSTM gas well optimized by utilizing the sparrow search algorithm;
s400, inputting data of the training set and the testing set into a sparrow search algorithm to optimize a daily gas yield prediction model of the LSTM gas well, and obtaining a daily gas yield prediction result of the gas well.
Further, the step S100 specifically includes the following steps:
s101, extracting historical data of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature, daily water yield and daily gas yield of the gas well;
s102, eliminating abnormal value and missing value data in historical data, and carrying out normalization processing;
s103, sorting historical data according to the production date of the gas well, creating a data set of an input model, wherein the input of the data set comprises seven parameters of the production date of the gas well, the production time of the gas well, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield, and the output of the data set is daily gas yield of the gas well.
Further, the step S300 specifically includes the following steps:
s301, initializing a sparrow population, and setting the total number of the sparrow population, the ratio of discoverers to joiners, the maximum iteration number and a safety value;
s302, calculating fitness values of all sparrows by taking the root mean square error as a fitness function, sequencing, and selecting the current optimal position and the worst position;
s303, updating the positions of the discoverer, the joiner and the early warning sparrow according to a position updating formula to obtain the firstiSparrow of the first kindjPosition information in the dimension and updating the fitness value thereof;
s304, judging whether the maximum iteration times are reached, stopping iteration if the maximum iteration times are met, taking the sparrow position with the lowest fitness value as the optimal solution, outputting the result, and otherwise, repeatedly executing the steps S302 to S303.
Further, the location update formula in step S303 includes:
the location of the finder updates the formula:
；
wherein,,trepresenting the number of current iterations and,j=1,2,3…,d；is a constant representing the maximum number of iterations; />Represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Is a random number; />And->Respectively representing an early warning value and a safety value;Qis a random number subject to normal distribution;Lis->And each element is a matrix of 1; exp represents an exponential function based on e;
the location of the enrollee updates the formula:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Indicating that the discoverer is at the first iteration when t+1 iterations occurjOptimal position information in dimension, < >>Then the current global worst position is indicated;Qis a random number subject to normal distribution;Lis thatAnd each element is a matrix of 1; exp represents an exponential function based on e;Afor a row of an n-dimensional matrix, the elements in the matrix are 1 or-1, and +.>；
The position update formula of the early warning sparrow:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Representing the current global optimal position when t iterations occur; />Representing a current global worst position when t iterations occur; />The step control parameter is a random number subject to normal distribution with a mean value of 0 and a variance of 1; />Is a random number,/->The fitness value of the current sparrow individual is the fitness value; />And->The current global optimal and worst fitness values respectively; />Is a non-zero constant.
Further, in the step S300, whether the model training is optimal is determined according to the root mean square error, and the model training with the minimum root mean square error corresponds to the optimal model training.
The present invention further provides a determining apparatus comprising:
the data acquisition module is used for acquiring data in the target data set;
the preprocessing module is used for preprocessing the data in the acquired target data set, deleting abnormal values and normalizing the abnormal values;
and the yield determining module is used for optimizing the daily gas yield prediction model hyper-parameters of the LSTM gas well by utilizing a sparrow search algorithm according to the target data set, improving the model prediction effect and determining the daily gas yield of the target gas well.
The present invention further provides an electronic device comprising:
a memory, a processor, a computer program executable in the processor and storable in the memory, an input device and an output device, wherein the processor implements the steps of the method as described above when the computer program is executed.
The invention further provides a storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the above method.
Drawings
Fig. 1 is a flowchart of a method for predicting daily gas production of a gas well provided in this embodiment.
Fig. 2 is a flowchart of step S100 of the method for predicting daily gas production of a gas well provided in this embodiment.
Fig. 3 is a flowchart of step S300 of the method for predicting daily gas production of a gas well provided in this embodiment.
Fig. 4 is a model training prediction effect diagram provided in the present embodiment.
Fig. 5 is a diagram showing the effect of model test prediction provided in this embodiment.
Detailed Description
The invention is further described below with reference to the embodiments and the accompanying drawings.
The embodiment provides a method for predicting daily gas production of a gas well, as shown in fig. 1, comprising the following steps:
s100, acquiring seven parameters of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield required by gas well daily gas yield prediction and gas well daily gas yield corresponding to the seven parameters, establishing a data set, and cleaning the data set;
s200, dividing a data set into a training set and a testing set, determining an initial learning rate, iteration times and hidden layer node number of the model, establishing an LSTM gas well daily gas yield prediction model, training the established LSTM gas well daily gas yield prediction model by using training set data, and testing prediction accuracy of the LSTM gas well daily gas yield prediction model by using data of the testing set;
s300, optimizing three super parameters of a learning rate, iteration times and hidden layer node number of a daily gas yield prediction model of the LSTM gas well by utilizing a sparrow search algorithm, continuously training and testing the model by using data of a training set and a testing set in the optimization process, keeping model parameters corresponding to the optimal model training, and then training the model by utilizing the obtained optimal model parameters and the data of the training set to establish the daily gas yield prediction model of the LSTM gas well optimized by utilizing the sparrow search algorithm;
s400, inputting data of the training set and the testing set into a sparrow search algorithm to optimize a daily gas yield prediction model of the LSTM gas well, and obtaining a daily gas yield prediction result of the gas well.
In a specific embodiment, as shown in fig. 2, the step S100 specifically includes the following steps:
s101, extracting historical data of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature, daily water yield and daily gas yield of the gas well;
s102, eliminating abnormal value and missing value data in historical data, and carrying out normalization processing;
s103, sorting historical data according to the production date of the gas well, creating a data set of an input model, wherein the input of the data set comprises seven parameters of the production date of the gas well, the production time of the gas well, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield, and the output of the data set is daily gas yield of the gas well.
In a specific embodiment, as shown in fig. 3, the step S300 specifically includes the following steps:
s301, initializing a sparrow population, and setting the total number of the sparrow population, the ratio of discoverers to joiners, the maximum iteration number and a safety value;
s302, calculating fitness values of all sparrows by taking the root mean square error as a fitness function, sequencing, and selecting the current optimal position and the worst position;
s303, updating the positions of the discoverer, the joiner and the early warning sparrow according to a position updating formula to obtain the firstiSparrow of the first kindjPosition information in the dimension and updating the fitness value thereof;
s304, judging whether the maximum iteration times are reached, stopping iteration if the maximum iteration times are met, taking the sparrow position with the lowest fitness value as the optimal solution, outputting the result, and otherwise, repeatedly executing the steps S302 to S303.
In a specific embodiment, the method for predicting daily gas production of the gas well updates the positions of discoverers, joiners and early warning sparks according to a position updating formula.
The location update formula for the finder is as follows:
；
wherein,,trepresenting the number of current iterations and,j=1,2,3…,d；is a constant representing the maximum number of iterations; />Represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Is a random number; />And->Respectively representing an early warning value and a safety value;Qis a random number subject to normal distribution;Lis->And each element is a matrix of 1; exp represents an exponential function based on e;
the location update formula of the enrollee is as follows:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Indicating that the discoverer is at the first iteration when t+1 iterations occurjOptimal position information in dimension, < >>Then the current global worst position is indicated;Qis a random number subject to normal distribution;Lis thatAnd each element is a matrix of 1; exp represents an exponential function based on e;Afor a row of an n-dimensional matrix, the elements in the matrix are 1 or-1, and +.>；
The position update formula of the early warning sparrow is as follows:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Representing the current global optimal position when t iterations occur; />Representing a current global worst position when t iterations occur; />The step control parameter is a random number subject to normal distribution with a mean value of 0 and a variance of 1; />Is a random number,/->The fitness value of the current sparrow individual is the fitness value; />And->The current global optimal and worst fitness values respectively; />Is a non-zero constant.
Taking a production well of a Fuling shale gas field as an example, 29 calendar history production data of the well of the gas field from 4/8/2014 to 3/2022 are collected and arranged. And acquiring the oil pressure, the casing pressure, the wellhead pipeline conveying pressure, the wellhead temperature, the production time of the gas well, the daily water yield, the production date of the gas well and the corresponding actual measured daily gas yield, creating a data set, and carrying out normalization processing. The total data set is 2838 groups, the data of the training set and the data of the testing set are divided according to the proportion of 8:2, then an LSTM gas well daily gas production prediction model is established, and three super parameters including the learning rate, the iteration times and the hidden layer node number of the LSTM gas well daily gas production prediction model are optimized by utilizing a sparrow search algorithm. Initializing a sparrow population, updating the positions of discoverers, joiners and early warning sparrows, selecting individuals according to the fitness value, updating the global optimal fitness value, obtaining the optimal super-parameters of the LSTM gas well daily gas production prediction model from the calculation result, and establishing a sparrow search algorithm to optimize the LSTM gas well daily gas production prediction model. Finally, training a sparrow search algorithm to optimize a daily gas yield prediction model of the LSTM gas well, wherein the correlation between the training prediction result and the actually measured daily gas yield is shown in figure 4. Inputting test data, and testing the reliability of a daily gas production prediction model of the LSTM by a sparrow search algorithm to obtain a daily gas production test prediction result of the LSTM. The correlation of the test prediction result and the measured daily gas production is shown in fig. 5.
The average relative error of the daily gas production of the gas well predicted by the model is 4.6%, the daily gas production of 2838 groups of data is predicted for 15 seconds, the daily gas production of 470 production wells of the Fuling shale gas field is predicted, the average relative error is 5.2%, the labor cost and the time cost are saved to a great extent on the premise of ensuring the prediction precision, the efficiency of the gas well in the production and development process is improved, and a foundation is laid for the application of big data analysis in the gas field.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A method for predicting daily gas production of a gas well, the method comprising the steps of:
s100, acquiring seven parameters of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield required by gas well daily gas yield prediction and gas well daily gas yield corresponding to the seven parameters, establishing a data set, and cleaning the data set;
s200, dividing a data set into a training set and a testing set, determining an initial learning rate, iteration times and hidden layer node number of the model, establishing an LSTM gas well daily gas yield prediction model, training the established LSTM gas well daily gas yield prediction model by using training set data, and testing prediction accuracy of the LSTM gas well daily gas yield prediction model by using data of the testing set;
s300, optimizing three super parameters of a learning rate, iteration times and hidden layer node number of a daily gas yield prediction model of the LSTM gas well by utilizing a sparrow search algorithm, continuously training and testing the model by using data of a training set and a testing set in the optimization process, keeping model parameters corresponding to the optimal model training, and then training the model by utilizing the obtained optimal model parameters and the data of the training set to establish the daily gas yield prediction model of the LSTM gas well optimized by utilizing the sparrow search algorithm;
s400, inputting data of the training set and the testing set into a sparrow search algorithm to optimize a daily gas yield prediction model of the LSTM gas well, and obtaining a daily gas yield prediction result of the gas well.
2. The method for predicting daily gas production in a gas well according to claim 1, wherein the step S100 specifically comprises the steps of:
s101, extracting historical data of gas well production date, gas well production time, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature, daily water yield and daily gas yield of the gas well;
s102, eliminating abnormal value and missing value data in historical data, and carrying out normalization processing;
s103, sorting historical data according to the production date of the gas well, creating a data set of an input model, wherein the input of the data set comprises seven parameters of the production date of the gas well, the production time of the gas well, oil pressure, casing pressure, wellhead pipeline conveying pressure, wellhead temperature and daily water yield, and the output of the data set is daily gas yield of the gas well.
3. The method for predicting daily gas production in a gas well according to claim 1, wherein the step S300 specifically comprises the steps of:
s301, initializing a sparrow population, and setting the total number of the sparrow population, the ratio of discoverers to joiners, the maximum iteration number and a safety value;
s302, calculating fitness values of all sparrows by taking the root mean square error as a fitness function, sequencing, and selecting the current optimal position and the worst position;
s303, updating the positions of the discoverer, the joiner and the early warning sparrow according to a position updating formula to obtain the firstiSparrow of the first kindjPosition information in the dimension and updating the fitness value thereof;
s304, judging whether the maximum iteration times are reached, stopping iteration if the maximum iteration times are met, taking the sparrow position with the lowest fitness value as the optimal solution, outputting the result, and otherwise, repeatedly executing the steps S302 to S303.
4. A method for predicting daily gas production in a gas well as claimed in claim 3, wherein the location update formula in step S303 comprises:
the location of the finder updates the formula:
；
wherein,,trepresenting the number of current iterations and,j=1,2,3…,d；is a constant representing the maximum number of iterations; />Represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Is a random number; />And->Respectively representing an early warning value and a safety value;Qis a random number subject to normal distribution;Lis->And each element is a matrix of 1; exp represents an exponential function based on e;
the location of the enrollee updates the formula:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Indicating that the discoverer is at the first iteration when t+1 iterations occurjOptimal position information in dimension, < >>Then the current global worst position is indicated;Qis a random number subject to normal distribution;Lis->And each element is a matrix of 1; exp represents an exponential function based on e;Afor a row of an n-dimensional matrix, the elements in the matrix are 1 or-1, and +.>；
The position update formula of the early warning sparrow:
；
wherein,,represent the firstiSparrow at time t+1 iterations occurjPosition information in the dimension; />Represent the firstiSparrow at the t-th iterationjPosition information in the dimension; />Representing the current global optimal position when t iterations occur; />Representing a current global worst position when t iterations occur; />The step control parameter is a random number subject to normal distribution with a mean value of 0 and a variance of 1; />Is a random number,/->The fitness value of the current sparrow individual is the fitness value; />And->The current global optimal and worst fitness values respectively; />Is a non-zero constant.
5. A method for predicting daily gas production in a gas well according to claim 3, wherein in step S300, whether the model training is optimal is determined based on a root mean square error, and the model training is the model training with the minimum root mean square error.
6. A determining apparatus, comprising:
the data acquisition module is used for acquiring data in the target data set;
the preprocessing module is used for preprocessing the data in the acquired target data set, deleting abnormal values and normalizing the abnormal values;
and the yield determining module is used for optimizing the daily gas yield prediction model hyper-parameters of the LSTM gas well by utilizing a sparrow search algorithm according to the target data set, improving the model prediction effect and determining the daily gas yield of the target gas well.
7. An electronic device, comprising:
memory, a processor, a computer program executable in the processor and storable in the memory, input means and output means, wherein the processor implements the steps of the method as claimed in any one of claims 1 to 5 when the computer program is executed.
8. A storage medium having stored thereon a computer program, which when executed by a processor performs the steps of the method according to any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750976.XA CN116502772A (en) | 2023-06-25 | 2023-06-25 | Gas well daily gas production prediction method, determination device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750976.XA CN116502772A (en) | 2023-06-25 | 2023-06-25 | Gas well daily gas production prediction method, determination device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116502772A true CN116502772A (en) | 2023-07-28 |
Family
ID=87318699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310750976.XA Pending CN116502772A (en) | 2023-06-25 | 2023-06-25 | Gas well daily gas production prediction method, determination device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116502772A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048932A (en) * | 2022-01-13 | 2022-02-15 | 西南石油大学 | Coal-bed gas well gas production rate prediction method based on LSTM |
| CN115374689A (en) * | 2022-01-17 | 2022-11-22 | 浙江科技学院 | Air quality index prediction method based on improved sparrow search algorithm optimization |
-
2023
- 2023-06-25 CN CN202310750976.XA patent/CN116502772A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048932A (en) * | 2022-01-13 | 2022-02-15 | 西南石油大学 | Coal-bed gas well gas production rate prediction method based on LSTM |
| CN115374689A (en) * | 2022-01-17 | 2022-11-22 | 浙江科技学院 | Air quality index prediction method based on improved sparrow search algorithm optimization |
Non-Patent Citations (1)
| Title |
|---|
| FEIYANG LIU 等: "Sparrow Search Algorithm-Optimized Long Short-Term Memory Model for Stock Trend Prediction", 《COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE》, pages 3 - 4 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110083125B (en) | Machine tool thermal error modeling method based on deep learning | |
| CN110942194A (en) | Wind power prediction error interval evaluation method based on TCN | |
| CN112364560B (en) | Intelligent prediction method for working hours of mine rock drilling equipment | |
| CN114488311A (en) | Transverse wave time difference prediction method based on SSA-ELM algorithm | |
| CN117571945A (en) | Groundwater pollution prediction method coupling time sequence prediction model and mechanism model | |
| CN115458071B (en) | Method, device and equipment for predicting heavy metal content in soil | |
| CN115906626A (en) | Underground water organic pollution mutual-feedback inversion tracing method based on artificial intelligence | |
| CN118246714B (en) | Water conservancy and hydropower engineering construction energy consumption analysis method and system | |
| CN118095659B (en) | Lake water level conformal prediction method integrating Copula function and deep learning | |
| Preis et al. | Online hydraulic state prediction for water distribution systems | |
| CN117669008B (en) | Foundation settlement prediction method and system based on deep learning | |
| CN116502772A (en) | Gas well daily gas production prediction method, determination device, electronic equipment and storage medium | |
| CN114718556A (en) | Method, device and equipment for acquiring artificial crack parameters | |
| CN116881676B (en) | Prediction method for water inflow of open pit | |
| CN117392813A (en) | Advanced prediction alarm method for lateral displacement of high-fill soil body | |
| CN116953528A (en) | Battery attenuation model construction method, prediction method, system, device and medium | |
| CN117365430A (en) | Shale gas single well daily output prediction method and system | |
| CN115860396A (en) | Intelligent planning method based on land reserve analysis model | |
| CN114862007A (en) | Short-period gas production rate prediction method and system for carbonate gas well | |
| CN117522173B (en) | Natural gas hydrate depressurization exploitation productivity prediction method based on deep neural network | |
| CN118606806B (en) | Average air volume prediction method and system based on roadway section sensor data | |
| CN109886420A (en) | A kind of adaptive coalcutter cutting height intelligent predicting system | |
| CN117684947B (en) | Deep learning-based oil well bottom hole flow pressure soft measurement method | |
| CN114707705B (en) | Warehouse-in flow prediction method, equipment and storage medium | |
| Yu et al. | Efficient Approaches for Online Training of Gaussian Process Models in Multi-objective Coastal Groundwater Management. Presented at Early Career Hydrogeologists’ Conference-University of Birmingham, UK (30 November 2023) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230728 |