A method, device, equipment and medium for real-time adjustment of oil exploration and production

Abstract

The present application discloses a method, device, equipment and medium for real-time adjustment of oil exploration and production, and relates to the technical field of oil exploration pumping unit control, including training of an oil pumping increment prediction model and an oil pumping energy consumption prediction model; generating an initial oil pumping parameter set based on actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, pumping oil from the oil field according to the initial oil pumping parameter set, collecting actual oil pumping increment data and actual oil pumping energy consumption data; fine-tuning the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model using actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data; generating an oil pumping adjustment parameter set according to the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data. Through the technical solution of the present application, it is possible to realize real-time adjustment of the oil pumping process of oil exploration, and improve the speed, flexibility and accuracy of oil pumping.

Description

Technical Field

[0001] The invention relates to the technical field of oil exploration pumping unit control, and in particular to a real-time adjustment method, device, equipment and medium for oil exploration and production. Background Art

[0002] The pumping speed of a petroleum exploration pumping unit is a key performance indicator affected by many factors. It is jointly regulated by the geological conditions of the oil field and the control parameters of the pumping unit, forming a complex and interrelated system. The existing intelligent control scheme for the pumping unit is: match the historical parameter library of the corresponding working oil well according to the real-time operating parameters of the current pumping unit, and then generate the empty pumping conditions of the working oil well, so as to determine whether the pumping unit is in an empty pumping state according to the empty pumping conditions. However, the existing technology can only determine whether the pumping unit is in an empty pumping state, which makes it impossible to rationally set the speed of the pumping unit, etc., and the traditional pumping unit control system relies on preset parameters and manual experience adjustment. Under the complex and changeable geological conditions of the oil field, the flexibility and accuracy are relatively low.

[0003] As can be seen from the above, how to achieve real-time adjustment of the oil extraction process in oil exploration and improve the speed, flexibility and accuracy of oil extraction is a problem to be solved in this field. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for real-time adjustment of oil exploration and production, which can realize real-time adjustment of the oil extraction process of oil exploration and improve the speed, flexibility and accuracy of oil extraction. The specific scheme is as follows:

[0005] In a first aspect, the present application discloses a method for real-time adjustment of oil exploration and production, comprising:

[0006] Acquire oil pumping model training data for oil exploration, and use the oil pumping model training data to train an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by an oil pumping unit and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time by an oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0007] Acquire actual oil pumping geological data of the oil field, generate an initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, pump oil from the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data;

[0008] Using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data, fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0009] An oil pumping adjustment parameter set is generated based on the finely tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and real-time oil pumping adjustment is performed on the oil field based on the oil pumping adjustment parameter set.

[0010] Optionally, the step of obtaining oil extraction model training data for oil exploration includes:

[0011] Acquire geological sample data of oil extraction for petroleum exploration; the geological sample data of oil extraction includes geographical parameters that affect the oil extraction speed of the oil pumping unit;

[0012] Obtaining different pumping unit parameter combinations, and generating pumping unit parameter sample data according to all the pumping unit parameter combinations; the pumping unit parameter combinations include rotation speed and frequency;

[0013] The oil pumping volume per unit time after adjusting the pumping unit parameter combination is used as an oil pumping increment label to generate oil pumping increment label data;

[0014] The power consumption per unit time after adjusting the pumping unit parameter combination is used as the pumping energy consumption label to generate pumping energy consumption label data; the pumping model training data includes the geological sample data, the pumping unit parameter sample data, the pumping increment label data and the pumping energy consumption label data.

[0015] Optionally, the pumping increment prediction model for predicting the amount of oil pumped per unit time of the pumping unit is trained using the pumping model training data to obtain a trained pumping increment prediction model, including:

[0016] The oil extraction geological sample data and the oil pumping unit parameter sample data are combined to generate each group of first sample feature vectors, and each group of the first sample feature vectors is used as the input of an oil extraction increment prediction model for predicting the oil extraction amount per unit time of the oil pumping unit, so as to output an oil extraction increment prediction value corresponding to each group of fault predicted duration feature vectors; the oil extraction increment prediction model is a polynomial regression model;

[0017] Taking the oil pumping increment label in the oil pumping increment label data as a prediction target, calculating the difference between the oil pumping increment prediction value and the oil pumping increment label, and taking the difference as a first prediction error;

[0018] The training objective is to minimize the sum of the first prediction errors, and the oil pumping increment prediction model is trained. When the sum of the first prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping increment prediction model.

[0019] Optionally, the pumping energy consumption prediction model for predicting the energy consumption per unit time of the pumping unit is trained using the pumping model training data to obtain a trained pumping energy consumption prediction model, including:

[0020] The oil pumping geological sample data and the oil pumping unit parameter sample data are combined to generate each group of second sample feature vectors, and each group of the second sample feature vectors is used as the input of an oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to output the oil pumping energy consumption prediction value corresponding to each group of fault predicted duration feature vectors; the oil pumping energy consumption prediction model is a polynomial regression model;

[0021] Taking the oil pumping energy consumption label in the oil pumping energy consumption label data as a prediction target, calculating the difference between the oil pumping energy consumption prediction value and the oil pumping energy consumption label, and taking the difference as a second prediction error;

[0022] The minimization of the sum of the second prediction errors is used as a training goal, and the oil pumping energy consumption prediction model is trained. When the sum of the second prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping energy consumption prediction model.

[0023] Optionally, the generating of an initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model includes:

[0024] Determine a parameter set, an actual rotation speed, and an actual frequency of various parameter combinations from the actual oil extraction geological data;

[0025] Generate a first increment function expression corresponding to the trained oil pumping increment prediction model and a first energy consumption function expression corresponding to the trained oil pumping energy consumption prediction model based on the parameter set, the actual rotation speed and the actual frequency;

[0026] Constructing an initial optimization function using the first increment function expression and the first energy consumption function expression;

[0027] The initial optimization function is solved to obtain an initial rotation speed and an initial frequency corresponding to a maximum value of the initial optimization function, and an initial oil pumping parameter set is generated based on the initial rotation speed and the initial frequency.

[0028] Optionally, the fine-tuning of the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model by using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data, and the actual oil pumping energy consumption data includes:

[0029] Determining the incremental initial parameters and the energy consumption initial parameters, combining the actual oil pumping geological data and the initial oil pumping parameter set to generate each group of third sample feature vectors;

[0030] Inputting the third sample feature vector into the trained oil pumping increment prediction model with the increment initial parameter as a parameter, and taking the actual oil pumping increment data as a prediction target, fine-tuning the trained oil pumping increment prediction model;

[0031] The third sample feature vector is input into the trained oil pumping energy consumption prediction model with the initial energy consumption parameter as a parameter, and the actual oil pumping energy consumption data is used as a prediction target to fine-tune the trained oil pumping energy consumption prediction model.

[0032] Optionally, generating a set of oil pumping adjustment parameters according to the finely tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data includes:

[0033] Generate a second increment function expression corresponding to the fine-tuned oil pumping increment prediction model and a second energy consumption function expression corresponding to the fine-tuned oil pumping energy consumption prediction model;

[0034] Constructing a target optimization function using the second increment function expression and the second energy consumption function expression;

[0035] The target optimization function is solved by using a derivation method or a gradient descent method to obtain a target rotation speed and a target frequency corresponding to a maximum value of the target optimization function, and a set of oil pumping adjustment parameters is generated based on the target rotation speed and the target frequency.

[0036] In a second aspect, the present application discloses a real-time adjustment device for oil exploration and production, comprising:

[0037] A model training module, used to obtain oil pumping model training data for oil exploration, and to train an oil pumping increment prediction model for predicting the oil pumping amount per unit time and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time of an oil pumping unit using the oil pumping model training data, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0038] An initial pumping module is used to obtain actual pumping geological data of the oil field, generate an initial pumping parameter set based on the actual pumping geological data, the trained pumping increment prediction model and the pumping energy consumption prediction model, pump the oil field according to the initial pumping parameter set, and collect actual pumping increment data and actual pumping energy consumption data;

[0039] A model fine-tuning module, used to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model by using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data, so as to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0040] The real-time oil pumping adjustment module is used to generate an oil pumping adjustment parameter set based on the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and to make real-time oil pumping adjustments to the oil field based on the oil pumping adjustment parameter set.

[0041] In a third aspect, the present application discloses an electronic device, comprising:

[0042] Memory, used to store computer programs;

[0043] The processor is used to execute the computer program to implement the aforementioned real-time adjustment method for oil exploration and production.

[0044] In a fourth aspect, the present application discloses a computer storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the aforementioned disclosed method for real-time adjustment of oil exploration and production are implemented.

[0045] It can be seen that the present application provides a method for real-time adjustment of oil exploration and production, including obtaining oil pumping model training data for oil exploration, using the oil pumping model training data to train an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by the pumping unit and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; obtaining actual oil pumping geological data of the oil field, generating an initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, and adjusting the oil pumping parameter set according to the initial oil pumping parameter set. The oil field is pumping oil, and actual oil pumping increment data and actual oil pumping energy consumption data are collected; the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model are fine-tuned using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; a pumping adjustment parameter set is generated based on the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and real-time adjustment of oil pumping is performed on the oil field based on the pumping adjustment parameter set. The present application trains the oil pumping increment prediction model and the oil pumping energy consumption prediction model, and generates an initial oil pumping parameter set in combination with actual oil pumping geological data, so as to carry out initial oil pumping on the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data, so as to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, improve the oil pumping efficiency and oil pumping effect of the oil exploration pumping unit, and generate an oil pumping adjustment parameter set by using the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model, so as to realize real-time adjustment of the oil pumping process of oil exploration, thereby improving the speed, flexibility and accuracy of oil pumping. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0047] Figure 1 A flow chart of a real-time adjustment method for oil exploration and production disclosed in this application;

[0048] Figure 2 A flow chart of a real-time adjustment method for oil exploration and production disclosed in this application;

[0049] Figure 3 This is a structural schematic diagram of a real-time adjustment device for oil exploration and production disclosed in this application;

[0050] Figure 4 A structural diagram of an electronic device provided for this application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0052] The pumping speed of a pumping unit in oil exploration is a key performance indicator affected by many factors. It is jointly regulated by the geological conditions of the oil field and the control parameters of the pumping unit, forming a complex and interrelated system. The existing intelligent control scheme for the pumping unit is: match the historical parameter library of the corresponding working oil well according to the real-time operating parameters of the current pumping unit, and then generate the empty pumping conditions of the working oil well, so as to determine whether the pumping unit is in an empty pumping state according to the empty pumping conditions. However, the existing technology can only determine whether the pumping unit is in an empty pumping state, which makes it impossible to rationally set the speed of the pumping unit, etc., and the traditional pumping unit control system relies on preset parameters and manual experience adjustment. Under complex and changeable oil field geological conditions, the flexibility and accuracy are relatively low. As can be seen from the above, how to achieve real-time adjustment of the pumping process of oil exploration and improve the speed, flexibility and accuracy of pumping are problems to be solved in this field.

[0053] See also Figure 1 As shown, the embodiment of the present invention discloses a method for real-time adjustment of oil exploration and production, which may specifically include:

[0054] Step S11: Acquire oil pumping model training data for oil exploration, and use the oil pumping model training data to train an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by the pumping unit and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time by the pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model.

[0055] In this embodiment, the pumping model training data includes the geological sample data, the pumping unit parameter sample data, the pumping increment label data and the pumping energy consumption label data. The specific process of obtaining the pumping model training data for oil exploration is as follows:

[0056] (1) Before using a petroleum exploration pumping unit to pump oil from an oil field each time, obtain geological sample data for oil exploration; the geological sample data for oil exploration includes geographical parameters that affect the pumping speed of the pumping unit; the geological parameters include but are not limited to oil layer permeability, porosity, reservoir pressure, rock leakage, formation structure, formation temperature, reservoir water content, and reservoir physical properties;

[0057] The permeability of an oil layer refers to how easily oil can move in the formation. A high permeability usually means that oil flows more easily, thus affecting the pumping efficiency of the pumping unit; porosity indicates the percentage of pores in a formation, that is, the space in the formation that can accommodate fluids. Higher porosity is usually more beneficial to the pumping effect of a pumping unit; the pressure inside the reservoir can affect the fluidity of the oil. Higher reservoir pressure helps to improve pumping efficiency; the seepage properties of rocks affect the movement of fluids in the formation. Different types of rocks affect the migration rate of oil, thereby affecting the amount of oil pumped by the pumping unit; the stratigraphic structure of the oil field, such as the distribution and connectivity of the pumping layers, will also affect the working effect of the pumping unit; the temperature of the oil layer can affect the viscosity of the oil, thereby affecting the pumping capacity of the pumping unit; the water content in the reservoir affects the relative permeability and fluidity of the oil, which may affect the oil-water separation efficiency of the pumping unit; physical parameters including the density and viscosity of crude oil also have a certain impact on the performance of the pumping unit;

[0058] Geological parameters are all related to the oil field itself, and because pumping oil will cause the depth of the oil field to change, the geological parameters may change each time the oil is pumped, so it is necessary to collect geological parameters before each pumping. Among them, the oil layer permeability uses logging instruments to record the formation permeability at different depths in the wellbore; the porosity is calculated by measuring the ratio of the volume of pores in the core to the total volume; the reservoir pressure can obtain the pressure information of the oil layer through logging data, especially the pressure logging tool in logging; rock leakage uses laboratory equipment to conduct permeability tests on rock samples to obtain the permeability characteristics of the rock; the formation structure obtains the structure and distribution information of the underground formation through seismic exploration; the formation temperature uses logging instruments to measure the formation temperature; the water content of the reservoir can be estimated through logging data, especially the resistivity logging tool in logging; the reservoir physical property information is obtained by sampling the reservoir fluid in the wellbore and analyzing it in the laboratory;

[0059] (2) obtaining different pumping unit parameter combinations, and generating pumping unit parameter sample data according to all the pumping unit parameter combinations; the pumping unit parameter combination includes rotation speed and frequency; that is, when pumping oil in an oil field, the pumping personnel manually control the oil exploration pumping unit to adopt different pumping unit parameter combinations, and collect the oil pumping volume per unit time and the power consumption per unit time during pumping each time the pumping unit parameter combination is changed. The pumping unit parameter combination is an adjustable functional parameter on the oil exploration pumping unit to control the speed of pumping oil;

[0060] (3) using the oil pumping volume per unit time after adjusting the pumping unit parameter combination as an oil pumping increment label to generate oil pumping increment label data;

[0061] (4) The power consumption per unit time after adjusting the pumping unit parameter combination is used as the pumping energy consumption label to generate pumping energy consumption label data.

[0062] In addition, the oil fields containing geological sample data, pumping unit parameter sample data, pumping increment label data, and pumping energy consumption label data may not be the oil fields to be pumped. Instead, several oil fields may be selected as experimental oil fields to collect more sufficient and comprehensive training data for the pumping model.

[0063] In this embodiment, the process of training the incremental oil pumping prediction model is as follows: the oil pumping geological sample data and the pumping unit parameter sample data are combined to generate each group of first sample feature vectors, and each group of the first sample feature vectors is used as the input of the incremental oil pumping prediction model for predicting the amount of oil pumped per unit time of the pumping unit, so as to output the incremental oil pumping prediction value corresponding to each group of fault predicted duration feature vectors; the incremental oil pumping prediction model is a polynomial regression model; the incremental oil pumping label in the incremental oil pumping label data is used as the prediction target, the difference between the incremental oil pumping prediction value and the incremental oil pumping label is calculated, and the difference is used as the first prediction error; the minimization of the sum of the first prediction errors is used as the training target, and the incremental oil pumping prediction model is trained. When the sum of the first prediction errors reaches convergence, the training is stopped to obtain the trained incremental oil pumping prediction model. The sum of the first prediction errors here can be the mean square error.

[0064] In this embodiment, the process of training the oil pumping energy consumption prediction model is as follows: the oil pumping geological sample data and the oil pumping unit parameter sample data are combined to generate each group of second sample feature vectors, and each group of the second sample feature vectors is used as the input of the oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to output the oil pumping energy consumption prediction value corresponding to each group of fault expected duration feature vectors; the oil pumping energy consumption prediction model is a polynomial regression model; the oil pumping energy consumption label in the oil pumping energy consumption label data is used as the prediction target, the difference between the oil pumping energy consumption prediction value and the oil pumping energy consumption label is calculated, and the difference is used as the second prediction error; the minimization of the sum of the second prediction errors is used as the training target, and the oil pumping energy consumption prediction model is trained. When the sum of the second prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping energy consumption prediction model. The sum of the second prediction errors here can also be the mean square error.

[0065] Step S12: Acquire actual oil pumping geological data of the oil field, generate an initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, pump the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data.

[0066] Step S13: Use the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model.

[0067] Step S14: generating an oil pumping adjustment parameter set according to the finely adjusted oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and performing real-time oil pumping adjustment on the oil field according to the oil pumping adjustment parameter set.

[0068] In this embodiment, a second incremental function expression corresponding to the fine-tuned oil pumping increment prediction model and a second energy consumption function expression corresponding to the fine-tuned oil pumping energy consumption prediction model are generated; the target optimization function is constructed using the second incremental function expression and the second energy consumption function expression; the target optimization function is solved using a derivative method or a gradient descent method to obtain a target speed and a target frequency corresponding to the maximum value of the target optimization function, a set of oil pumping adjustment parameters is generated based on the target speed and the target frequency, and then real-time oil pumping adjustments are performed on the oil field according to the set of oil pumping adjustment parameters.

[0069] Specifically, a second incremental function expression h3(D, v, f) corresponding to the fine-tuned oil pumping increment prediction model is generated, and a second energy consumption function expression h4(D, v, f) corresponding to the fine-tuned oil pumping energy consumption prediction model is generated. A target optimization function Y1=a1×h3(D, v, f)+a2×h4(D, v, f) is constructed, wherein a1 and a2 are respectively preset proportional coefficients, and a derivative method or a gradient descent method is used to solve the speed and frequency that make the target optimization function Y1 reach the maximum value. The target speed and target frequency when the target optimization function Y1 reaches the maximum value are marked as v1 and f1 respectively, and an oil pumping adjustment parameter set is generated based on the target speed v1 and the target frequency f1, and then the oil field is adjusted in real time according to the oil pumping adjustment parameter set.

[0070] The present application trains an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by an oil pumping unit and an oil pumping energy consumption prediction model for predicting the amount of energy consumed per unit time by an oil pumping unit through oil pumping model training data to generate an initial oil pumping parameter set. Before pumping oil from an oil field to be pumped, more appropriate oil pumping is performed using the initial oil pumping parameter set. Then, the oil pumping increment prediction model and the oil pumping energy consumption prediction model are fine-tuned using actual oil pumping geological data to generate a final oil pumping adjustment parameter set, thereby making real-time adjustments to oil pumping from the oil field based on the oil pumping adjustment parameter set to improve oil pumping efficiency and achieve better oil pumping effects. Furthermore, the oil pumping increment prediction model and the oil pumping energy consumption prediction model are fine-tuned based on the actual amount of oil pumped and energy consumption generated by pumping oil from the oil field to be pumped, to generate more accurate oil pumping increment prediction models and oil pumping energy consumption prediction models, thereby specifically improving the oil pumping effects of the oil fields to be pumped.

[0071] In this embodiment, the oil pumping model training data of oil exploration is obtained, and the oil pumping model training data is used to train the oil pumping increment prediction model for predicting the oil pumping amount per unit time of the oil pumping unit and the oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; the actual oil pumping geological data of the oil field is obtained, and an initial oil pumping parameter set is generated based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, and the oil field is pumped according to the initial oil pumping parameter set, and the oil field is sampled. Collect actual oil pumping increment data and actual oil pumping energy consumption data; use the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; generate an oil pumping adjustment parameter set based on the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and perform real-time oil pumping adjustments on the oil field based on the oil pumping adjustment parameter set. The present application trains the oil pumping increment prediction model and the oil pumping energy consumption prediction model, and generates an initial oil pumping parameter set in combination with actual oil pumping geological data, so as to carry out initial oil pumping on the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data, so as to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, improve the oil pumping efficiency and oil pumping effect of the oil exploration pumping unit, and generate an oil pumping adjustment parameter set by using the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model, so as to realize real-time adjustment of the oil pumping process of oil exploration, thereby improving the speed, flexibility and accuracy of oil pumping.

[0072] See also Figure 2 As shown, the embodiment of the present invention discloses a method for real-time adjustment of oil exploration and production, which may specifically include:

[0073] Step S21: Acquire oil pumping model training data for oil exploration, and use the oil pumping model training data to train an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by the pumping unit and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time by the pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model.

[0074] Step S22: Acquire actual oil pumping geological data of the oil field, determine a parameter set, an actual rotational speed, and an actual frequency of various parameter combinations from the actual oil pumping geological data, generate a first incremental function expression corresponding to the trained oil pumping increment prediction model and a first energy consumption function expression corresponding to the trained oil pumping energy consumption prediction model based on the parameter set, the actual rotational speed, and the actual frequency, construct an initial optimization function using the first incremental function expression and the first energy consumption function expression, solve the initial optimization function to obtain an initial rotational speed and an initial frequency corresponding to the maximum value of the initial optimization function, and generate an initial oil pumping parameter set based on the initial rotational speed and the initial frequency.

[0075] In this embodiment, the parameter set composed of various parameters in the actual oil pumping geological data is marked as D, the actual rotation speed is marked as v, and the actual frequency is marked as f. The first incremental function expression corresponding to the trained oil pumping increment prediction model is generated as h1(D, v, f), and the first energy consumption function expression corresponding to the trained oil pumping energy consumption prediction model is generated as h2(D, v, f). An initial optimization function Y=a1×h1(D, v, f)+a2×h2(D, v, f) is constructed, where a1 and a2 are preset proportional coefficients, respectively. The initial rotation speed and initial frequency that make the initial optimization function Y reach the maximum value are solved using the derivation method or the gradient descent method. The initial rotation speed and initial frequency when the initial optimization function Y reaches the maximum value are marked as v0 and f0, respectively. Finally, an initial oil pumping parameter set is formed based on the initial rotation speed v0 and the initial frequency f0.

[0076] Step S23: Pumping oil from the oil field according to the initial pumping parameter set, and collecting actual pumping increment data and actual pumping energy consumption data.

[0077] Step S24: Determine the incremental initial parameters and the energy consumption initial parameters, combine the actual oil pumping geological data and the initial oil pumping parameter set to generate each group of third sample feature vectors, input the third sample feature vectors into the trained oil pumping incremental prediction model with the incremental initial parameters as parameters, and use the actual oil pumping incremental data as the prediction target to fine-tune the trained oil pumping incremental prediction model, input the third sample feature vectors into the trained oil pumping energy consumption prediction model with the energy consumption initial parameters as parameters, and use the actual oil pumping energy consumption data as the prediction target to fine-tune the trained oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping incremental prediction model and the oil pumping energy consumption prediction model.

[0078] In this embodiment, the parameters of the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model are used as the incremental initial parameters and the energy consumption initial parameters respectively; the actual oil pumping geological data and the initial recommended parameter set are combined into a group of new sample feature vectors, and the new sample feature vectors are used as the incremental initial parameters in the new oil pumping increment prediction model. The new oil pumping increment prediction model takes the actual oil pumping increment data as the prediction target, and the new oil pumping increment prediction model is retrained. The trained new oil pumping increment prediction model is the fine-tuned oil pumping increment prediction model; the new sample feature vectors are used as the new oil pumping energy consumption prediction model with the energy consumption initial parameters as the parameters. The new oil pumping energy consumption prediction model takes the actual oil pumping energy consumption data as the prediction target, and the new oil pumping energy consumption prediction model is retrained. The trained new oil pumping energy consumption prediction model is the fine-tuned oil pumping energy consumption prediction model.

[0079] Step S25: generating an oil pumping adjustment parameter set according to the finely adjusted oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and performing real-time oil pumping adjustment on the oil field according to the oil pumping adjustment parameter set.

[0080] In this embodiment, the oil pumping model training data of oil exploration is obtained, and the oil pumping model training data is used to train the oil pumping increment prediction model for predicting the oil pumping amount per unit time of the oil pumping unit and the oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; the actual oil pumping geological data of the oil field is obtained, and an initial oil pumping parameter set is generated based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, and the oil field is pumped according to the initial oil pumping parameter set, and the oil field is sampled. Collect actual oil pumping increment data and actual oil pumping energy consumption data; use the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; generate an oil pumping adjustment parameter set based on the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and perform real-time oil pumping adjustments on the oil field based on the oil pumping adjustment parameter set. The present application trains the oil pumping increment prediction model and the oil pumping energy consumption prediction model, and generates an initial oil pumping parameter set in combination with actual oil pumping geological data, so as to carry out initial oil pumping on the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data, so as to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, improve the oil pumping efficiency and oil pumping effect of the oil exploration pumping unit, and generate an oil pumping adjustment parameter set by using the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model, so as to realize real-time adjustment of the oil pumping process of oil exploration, thereby improving the speed, flexibility and accuracy of oil pumping.

[0081] See also Figure 3 As shown, the embodiment of the present invention discloses a real-time adjustment device for oil exploration and production, which may specifically include:

[0082] The model training module 11 is used to obtain the oil pumping model training data for oil exploration, and use the oil pumping model training data to train the oil pumping increment prediction model for predicting the oil pumping amount per unit time of the oil pumping unit and the oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0083] The initial pumping module 12 is used to obtain actual pumping geological data of the oil field, generate an initial pumping parameter set based on the actual pumping geological data, the trained pumping increment prediction model and the pumping energy consumption prediction model, pump the oil field according to the initial pumping parameter set, and collect actual pumping increment data and actual pumping energy consumption data;

[0084] A model fine-tuning module 13 is used to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model by using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data, so as to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model;

[0085] The real-time oil pumping adjustment module 14 is used to generate an oil pumping adjustment parameter set according to the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and to perform real-time oil pumping adjustment on the oil field according to the oil pumping adjustment parameter set.

[0086] In this embodiment, the oil pumping model training data of oil exploration is obtained, and the oil pumping model training data is used to train the oil pumping increment prediction model for predicting the oil pumping amount per unit time of the oil pumping unit and the oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; the actual oil pumping geological data of the oil field is obtained, and an initial oil pumping parameter set is generated based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, and the oil field is pumped according to the initial oil pumping parameter set, and the oil field is sampled. Collect actual oil pumping increment data and actual oil pumping energy consumption data; use the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; generate an oil pumping adjustment parameter set based on the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and perform real-time oil pumping adjustments on the oil field based on the oil pumping adjustment parameter set. The present application trains the oil pumping increment prediction model and the oil pumping energy consumption prediction model, and generates an initial oil pumping parameter set in combination with actual oil pumping geological data, so as to carry out initial oil pumping on the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data, so as to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, improve the oil pumping efficiency and oil pumping effect of the oil exploration pumping unit, and generate an oil pumping adjustment parameter set by using the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model, so as to realize real-time adjustment of the oil pumping process of oil exploration, thereby improving the speed, flexibility and accuracy of oil pumping.

[0087] In some specific embodiments, the model training module 11 may specifically include:

[0088] The oil pumping geological sample data acquisition module is used to acquire the oil pumping geological sample data for oil exploration; the oil pumping geological sample data includes geographical parameters that affect the oil pumping speed of the pumping unit;

[0089] A pumping unit parameter sample data generation module is used to obtain different pumping unit parameter combinations and generate pumping unit parameter sample data according to all the pumping unit parameter combinations; the pumping unit parameter combinations include rotation speed and frequency;

[0090] A label data generating module, used for taking the oil pumping volume per unit time after adjusting the pumping unit parameter combination as an oil pumping increment label to generate oil pumping increment label data;

[0091] The oil pumping energy consumption label data generation module is used to use the power consumption per unit time after adjusting the pumping unit parameter combination as the oil pumping energy consumption label to generate the oil pumping energy consumption label data; the oil pumping model training data includes the geological sample data, the pumping unit parameter sample data, the oil pumping increment label data and the oil pumping energy consumption label data.

[0092] In some specific embodiments, the model training module 11 may specifically include:

[0093] The pumping increment prediction value output module is used to combine the pumping geological sample data and the pumping unit parameter sample data to generate each group of first sample feature vectors, and use each group of the first sample feature vectors as the input of the pumping increment prediction model for predicting the pumping volume per unit time of the pumping unit, so as to output the pumping increment prediction value corresponding to each group of fault predicted duration feature vectors; the pumping increment prediction model is a polynomial regression model;

[0094] A first prediction error calculation module is used to take the oil pumping increment label in the oil pumping increment label data as a prediction target, calculate the difference between the oil pumping increment prediction value and the oil pumping increment label, and take the difference as a first prediction error;

[0095] The oil pumping increment prediction model training module is used to train the oil pumping increment prediction model by minimizing the sum of the first prediction errors as a training goal. When the sum of the first prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping increment prediction model.

[0096] In some specific embodiments, the model training module 11 may specifically include:

[0097] The pumping energy consumption prediction value output module is used to combine the pumping geological sample data and the pumping unit parameter sample data to generate each group of second sample feature vectors, and use each group of the second sample feature vectors as the input of the pumping energy consumption prediction model for predicting the energy consumption per unit time of the pumping unit, so as to output the pumping energy consumption prediction value corresponding to each group of fault expected duration feature vectors; the pumping energy consumption prediction model is a polynomial regression model;

[0098] A second prediction error calculation module is used to take the oil pumping energy consumption label in the oil pumping energy consumption label data as a prediction target, calculate the difference between the oil pumping energy consumption prediction value and the oil pumping energy consumption label, and take the difference as a second prediction error;

[0099] The oil pumping energy consumption prediction model training module is used to train the oil pumping energy consumption prediction model with minimizing the sum of the second prediction errors as the training goal. When the sum of the second prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping energy consumption prediction model.

[0100] In some specific embodiments, the initial oil extraction module 12 may specifically include:

[0101] A parameter determination module, used to determine a parameter set of various parameter combinations, an actual rotation speed and an actual frequency from the actual oil extraction geological data;

[0102] A first function expression generating module, for generating a first increment function expression corresponding to the trained oil pumping increment prediction model and a first energy consumption function expression corresponding to the trained oil pumping energy consumption prediction model based on the parameter set, the actual rotation speed and the actual frequency;

[0103] An initial optimization function construction module, used to construct an initial optimization function using the first increment function expression and the first energy consumption function expression;

[0104] The initial pumping parameter set generation module is used to solve the initial optimization function to obtain the initial rotation speed and initial frequency corresponding to the maximum value of the initial optimization function, and generate an initial pumping parameter set based on the initial rotation speed and the initial frequency.

[0105] In some specific embodiments, the model fine-tuning module 13 may specifically include:

[0106] A third sample characteristic vector generation module, used to determine the incremental initial parameters and the energy consumption initial parameters, and to combine the actual oil pumping geological data with the initial oil pumping parameter set to generate each group of third sample characteristic vectors;

[0107] A pumping increment prediction model fine-tuning module, used for inputting the third sample feature vector into the trained pumping increment prediction model with the increment initial parameter as a parameter, and taking the actual pumping increment data as a prediction target to fine-tune the trained pumping increment prediction model;

[0108] The oil pumping energy consumption prediction model fine-tuning module is used to input the third sample feature vector into the trained oil pumping energy consumption prediction model with the initial energy consumption parameters as parameters, and use the actual oil pumping energy consumption data as the prediction target to fine-tune the trained oil pumping energy consumption prediction model.

[0109] In some specific embodiments, the oil pumping real-time adjustment module 14 may specifically include:

[0110] A second function expression generating module, used to generate a second increment function expression corresponding to the fine-tuned oil pumping increment prediction model and a second energy consumption function expression corresponding to the fine-tuned oil pumping energy consumption prediction model;

[0111] A target optimization function construction module, used to construct a target optimization function using the second increment function expression and the second energy consumption function expression;

[0112] The oil pumping adjustment parameter set generation module is used to solve the target optimization function by using the derivation method or the gradient descent method to obtain the target speed and target frequency corresponding to the maximum value of the target optimization function, and generate the oil pumping adjustment parameter set based on the target speed and the target frequency.

[0113] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the real-time adjustment method for oil exploration and production performed by the electronic device disclosed in any of the aforementioned embodiments.

[0114] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0115] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon include an operating system 221, a computer program 222 and data 223, etc. The storage method can be temporary storage or permanent storage.

[0116] The operating system 221 is used to manage and control the hardware devices and computer programs 222 on the electronic device 20, so as to realize the operation and processing of the data 223 in the memory 22 by the processor 21, which can be Windows, Unix, Linux, etc. In addition to including a computer program that can be used to complete the oil exploration and production real-time adjustment method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can further include a computer program that can be used to complete other specific tasks. In addition to data transmitted from an external device received by the oil exploration and production real-time adjustment device, the data 223 can also include data collected by its own input and output interface 25, etc.

[0117] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0118] Furthermore, an embodiment of the present application also discloses a computer-readable storage medium, in which a computer program is stored. When the computer program is loaded and executed by a processor, the steps of the real-time adjustment method for oil exploration and production disclosed in any of the aforementioned embodiments are implemented.

[0119] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0120] The above is a detailed introduction to a method, device, equipment and storage medium for real-time adjustment of oil exploration and production provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for a person skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for real-time adjustment of oil exploration and production, characterized in that: include: Acquire oil pumping model training data for oil exploration, and use the oil pumping model training data to train an oil pumping increment prediction model for predicting the amount of oil pumped per unit time by an oil pumping unit and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time by an oil pumping unit, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; Acquire actual oil pumping geological data of the oil field, generate an initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model, pump oil from the oil field according to the initial oil pumping parameter set, and collect actual oil pumping increment data and actual oil pumping energy consumption data; Using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data, fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; Generate an oil pumping adjustment parameter set according to the finely adjusted oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and perform real-time oil pumping adjustment on the oil field according to the oil pumping adjustment parameter set; The step of obtaining the oil extraction model training data for oil exploration includes: Acquire geological sample data of oil extraction for oil exploration; the geological sample data of oil extraction includes geographical parameters that affect the oil extraction speed of the oil pumping unit; Obtaining different pumping unit parameter combinations, and generating pumping unit parameter sample data according to all the pumping unit parameter combinations; the pumping unit parameter combinations include rotation speed and frequency; The oil pumping volume per unit time after adjusting the pumping unit parameter combination is used as an oil pumping increment label to generate oil pumping increment label data; The power consumption per unit time after adjusting the pumping unit parameter combination is used as the pumping energy consumption label to generate pumping energy consumption label data; the pumping model training data includes the geological sample data, the pumping unit parameter sample data, the pumping increment label data and the pumping energy consumption label data; The pumping increment prediction model for predicting the amount of oil pumped per unit time is trained using the pumping model training data to obtain the trained pumping increment prediction model, including: The oil extraction geological sample data and the oil pumping unit parameter sample data are combined to generate each group of first sample feature vectors, and each group of the first sample feature vectors is used as the input of an oil extraction increment prediction model for predicting the oil extraction amount per unit time of the oil pumping unit, so as to output an oil extraction increment prediction value corresponding to each group of fault predicted duration feature vectors; the oil extraction increment prediction model is a polynomial regression model; Taking the oil pumping increment label in the oil pumping increment label data as a prediction target, calculating the difference between the oil pumping increment prediction value and the oil pumping increment label, and taking the difference as a first prediction error; Taking minimizing the sum of the first prediction errors as a training objective, the oil pumping increment prediction model is trained, and when the sum of the first prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping increment prediction model; The pumping energy consumption prediction model for predicting the energy consumption per unit time of the pumping unit is trained using the pumping model training data to obtain the trained pumping energy consumption prediction model, including: The oil pumping geological sample data and the oil pumping unit parameter sample data are combined to generate each group of second sample feature vectors, and each group of the second sample feature vectors is used as the input of an oil pumping energy consumption prediction model for predicting the energy consumption per unit time of the oil pumping unit, so as to output the oil pumping energy consumption prediction value corresponding to each group of fault predicted duration feature vectors; the oil pumping energy consumption prediction model is a polynomial regression model; Taking the oil pumping energy consumption label in the oil pumping energy consumption label data as a prediction target, calculating the difference between the oil pumping energy consumption prediction value and the oil pumping energy consumption label, and taking the difference as a second prediction error; The minimization of the sum of the second prediction errors is used as a training goal, and the oil pumping energy consumption prediction model is trained. When the sum of the second prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping energy consumption prediction model.

2. The method for real-time adjustment of oil exploration and production according to claim 1, characterized in that: The generating of the initial oil pumping parameter set based on the actual oil pumping geological data, the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model comprises: Determine a parameter set, an actual rotation speed, and an actual frequency of various parameter combinations from the actual oil extraction geological data; Generate a first increment function expression corresponding to the trained oil pumping increment prediction model and a first energy consumption function expression corresponding to the trained oil pumping energy consumption prediction model based on the parameter set, the actual rotation speed and the actual frequency; Constructing an initial optimization function using the first increment function expression and the first energy consumption function expression; The initial optimization function is solved to obtain an initial rotation speed and an initial frequency corresponding to a maximum value of the initial optimization function, and an initial oil pumping parameter set is generated based on the initial rotation speed and the initial frequency.

3. The method for real-time adjustment of oil exploration and production according to claim 1, characterized in that: The fine-tuning of the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model by using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data, and the actual oil pumping energy consumption data comprises: Determine the incremental initial parameters and the energy consumption initial parameters, and combine the actual oil pumping geological data with the initial oil pumping parameter set to generate each group of third sample feature vectors; Inputting the third sample feature vector into the trained oil pumping increment prediction model with the increment initial parameter as a parameter, and taking the actual oil pumping increment data as a prediction target, fine-tuning the trained oil pumping increment prediction model; The third sample feature vector is input into the trained oil pumping energy consumption prediction model with the initial energy consumption parameter as a parameter, and the actual oil pumping energy consumption data is used as a prediction target to fine-tune the trained oil pumping energy consumption prediction model.

4. The method for real-time adjustment of oil exploration and production according to any one of claims 1 to 3, characterized in that: The step of generating a set of oil pumping adjustment parameters according to the finely adjusted oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data includes: Generate a second increment function expression corresponding to the fine-tuned oil pumping increment prediction model and a second energy consumption function expression corresponding to the fine-tuned oil pumping energy consumption prediction model; Constructing a target optimization function using the second increment function expression and the second energy consumption function expression; The target optimization function is solved by using a derivation method or a gradient descent method to obtain a target rotation speed and a target frequency corresponding to a maximum value of the target optimization function, and a set of oil pumping adjustment parameters is generated based on the target rotation speed and the target frequency.

5. A real-time adjustment device for oil exploration and production, characterized in that: include: A model training module, used to obtain oil pumping model training data for oil exploration, and to train an oil pumping increment prediction model for predicting the oil pumping amount per unit time and an oil pumping energy consumption prediction model for predicting the energy consumption per unit time of an oil pumping unit using the oil pumping model training data, so as to obtain the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model; An initial pumping module is used to obtain actual pumping geological data of the oil field, generate an initial pumping parameter set based on the actual pumping geological data, the trained pumping increment prediction model and the pumping energy consumption prediction model, pump the oil field according to the initial pumping parameter set, and collect actual pumping increment data and actual pumping energy consumption data; A model fine-tuning module, used to fine-tune the trained oil pumping increment prediction model and the oil pumping energy consumption prediction model by using the actual oil pumping geological data, the initial oil pumping parameter set, the actual oil pumping increment data and the actual oil pumping energy consumption data, so as to obtain the fine-tuned oil pumping increment prediction model and the oil pumping energy consumption prediction model; A real-time oil pumping adjustment module is used to generate an oil pumping adjustment parameter set according to the finely adjusted oil pumping increment prediction model and the oil pumping energy consumption prediction model and the actual oil pumping geological data, and to perform real-time oil pumping adjustment on the oil field according to the oil pumping adjustment parameter set; Wherein, the model training module includes: The oil pumping geological sample data acquisition module is used to acquire the oil pumping geological sample data for oil exploration; the oil pumping geological sample data includes geographical parameters that affect the oil pumping speed of the pumping unit; A pumping unit parameter sample data generation module is used to obtain different pumping unit parameter combinations and generate pumping unit parameter sample data according to all the pumping unit parameter combinations; the pumping unit parameter combinations include rotation speed and frequency; A label data generating module, used for taking the oil pumping volume per unit time after adjusting the pumping unit parameter combination as an oil pumping increment label to generate oil pumping increment label data; The pumping energy consumption label data generating module is used to use the power consumption per unit time after adjusting the pumping unit parameter combination as the pumping energy consumption label to generate the pumping energy consumption label data; the pumping model training data includes the geological sample data, the pumping unit parameter sample data, the pumping increment label data and the pumping energy consumption label data; Wherein, the model training module includes: The pumping increment prediction value output module is used to combine the pumping geological sample data and the pumping unit parameter sample data to generate each group of first sample feature vectors, and use each group of the first sample feature vectors as the input of the pumping increment prediction model for predicting the pumping volume per unit time of the pumping unit, so as to output the pumping increment prediction value corresponding to each group of fault predicted duration feature vectors; the pumping increment prediction model is a polynomial regression model; A first prediction error calculation module is used to take the oil pumping increment label in the oil pumping increment label data as a prediction target, calculate the difference between the oil pumping increment prediction value and the oil pumping increment label, and take the difference as a first prediction error; The oil pumping increment prediction model training module is used to train the oil pumping increment prediction model by minimizing the sum of the first prediction errors as a training target, and to stop the training when the sum of the first prediction errors reaches convergence, thereby obtaining the trained oil pumping increment prediction model; Wherein, the model training module includes: The pumping energy consumption prediction value output module is used to combine the pumping geological sample data and the pumping unit parameter sample data to generate each group of second sample feature vectors, and use each group of the second sample feature vectors as the input of the pumping energy consumption prediction model for predicting the energy consumption per unit time of the pumping unit, so as to output the pumping energy consumption prediction value corresponding to each group of fault expected duration feature vectors; the pumping energy consumption prediction model is a polynomial regression model; A second prediction error calculation module is used to take the oil pumping energy consumption label in the oil pumping energy consumption label data as a prediction target, calculate the difference between the oil pumping energy consumption prediction value and the oil pumping energy consumption label, and take the difference as a second prediction error; The oil pumping energy consumption prediction model training module is used to train the oil pumping energy consumption prediction model with minimizing the sum of the second prediction errors as the training goal. When the sum of the second prediction errors reaches convergence, the training is stopped to obtain the trained oil pumping energy consumption prediction model.

6. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the real-time adjustment method for oil exploration and production as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: Used to store computer programs; wherein, when the computer program is executed by a processor, the real-time adjustment method for oil exploration and production as described in any one of claims 1 to 4 is implemented.

Images