Regulation and control method, device and equipment of electric submersible pump layered producing well simulation model and medium

Abstract

The invention discloses a regulation and control method, device and equipment for an electric submersible pump layered producing well simulation model and a medium, relates to the technical field of computers, is applied to an intelligent agent and comprises the steps that current state information of the current environment where the electric submersible pump layered producing well simulation model is located is obtained; determining a regulation action according to a depth deterministic strategy gradient algorithm and based on the current state information; the regulation action is a regulation preset parameter; the preset parameters are one or more of the motor operation frequency of the electric submersible pump layered producing well, the diameter of a wellhead oil nozzle and the opening degree of each layer of inflow control device; the electric submersible pump layered producing well simulation model is regulated and controlled based on the regulation and control action, and the working performance corresponding to the regulated and controlled model is determined; judging whether the working performance meets a preset condition or not; and if not, skipping to the step of acquiring the current state information corresponding to the current environment where the electric submersible pump layered producing well simulation model is located until the working performance meets the preset condition. The speed and precision of regulation and control can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method, device, equipment and medium for regulating an electric submersible pump stratified oil production well simulation model. Background Art

[0002] Optimization of oil well production parameters and operation regulation is a multi-parameter collaborative decision-making problem. Since the wellhead and downhole parameters in the oil well environment are interrelated and dynamically changing, the level values of each parameter cannot be determined relatively independently. The problem of optimizing the inflow-lifting parameters of a separate production electric pump well to improve the working performance is a typical high-dimensional and complex problem.

[0003] Many scholars have studied the production optimization and regulation technology of oil and gas wells. Zi-ming Feng et al. established a dynamic model and optimization method for a variable-speed drive pumping unit in order to improve the efficiency of the pumping unit well system, make full use of the inertial load of moving parts, and can improve the comprehensive performance of the pumping unit well system, increase production, and reduce energy consumption. Gao xiaoyong et al. proposed a multi-period mixed-integer nonlinear programming (MINLP) model considering the production status of offshore electric pump wells, polymer flooding, energy consumption, platform inventory, and flow assurance to achieve the goal of minimizing the total operating cost. Sharma, R et al. proposed a nonlinear steady-state optimization method based on sequential quadratic programming (SQP) with the optimization goal of reducing energy consumption and maintaining stable production, and optimized the optimal oil and gas production and optimal pump speed of each well. K.M. Patel adopted APC multivariable control technology to achieve the coordinated control of the wellhead choke and operating frequency of an electric submersible pump well. Delou, P.D. et al. considered the safety operation constraint conditions of an electric submersible pump well and established a predictive control model for a submersible electric pump based on an adaptive strategy and real-time monitoring data in a throttling and variable-frequency regulation mode. Krishnamoorthy, Dinesh et al. developed a high-fidelity simulator for lifting heavy oil by an electric pump, and achieved the adaptive control of an electric submersible pump well through an MPC controller. The controller has sufficient robustness to the nonlinearity, operating conditions, disturbances, and measurement noise of the automatic control system. The above studies have achieved the closed-loop dynamic regulation of electric pump lifting wells through methods such as nonlinear steady-state optimization, predictive control, and adaptive control.

[0004] However, for the complex and changeable production environment of oil wells and the multi-parameter system collaborative optimization model of nonlinear systems, there are problems of slow convergence speed and low prediction accuracy. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for regulating an electric submersible pump stratified oil production well simulation model, which can improve the regulation speed and accuracy. The specific scheme is as follows:

[0006] In a first aspect, the present application discloses a method for regulating an electric submersible pump stratified oil production well simulation model, which is applied to an intelligent agent and includes:

[0007] Obtain the current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model;

[0008] Determine a regulation action according to the deep deterministic policy gradient algorithm and based on the current state information; the regulation action is to regulate a preset parameter; the preset parameter is one or more of the motor operating frequency of the electric submersible pump stratified oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device;

[0009] Regulate the electric submersible pump stratified oil production well simulation model based on the regulation action, and determine the working performance corresponding to the regulated model;

[0010] Judge whether the working performance meets a preset condition;

[0011] If not, jump to the step of obtaining the current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model until the working performance meets the preset condition.

[0012] Optionally, determining the working performance corresponding to the regulated model includes:

[0013] Determine the oil production rate of the electric submersible pump and the pump efficiency of the electric submersible pump corresponding to the regulated model.

[0014] Among them, judging whether the working performance meets a preset condition includes:

[0015] Judge whether the oil production rate of the electric submersible pump is within a preset production range, and judge whether the pump efficiency of the electric submersible pump is the best pump efficiency corresponding to the regulated environment;

[0016] If so, it is determined that the preset condition is met; if not, it is determined that the preset condition is not met.

[0017] Optionally, the current state information includes the motor operating frequency of the electric submersible pump stratified oil production well, the wellhead choke diameter, the opening degree of each layer of inflow control device, the oil production rate of the electric submersible pump, the pump efficiency of the electric submersible pump, the pump inlet pressure, the pump outlet pressure, the oil pressure, and the pressure behind the choke.

[0018] Optionally, determining the regulation action according to the deep deterministic policy gradient algorithm and based on the current state information includes:

[0019] Determine the opening degree of each layer of inflow control device to be adjusted according to the deep deterministic policy gradient algorithm and based on the current state information to change the oil well inflow capacity;

[0020] And / or, determine to adjust the diameter of the wellhead choke according to the deep deterministic policy gradient algorithm and based on the current state information to change the pipeline characteristics;

[0021] And / or, determine to adjust the operating frequency of the motor according to the deep deterministic policy gradient algorithm and based on the current state information.

[0022] Optionally, after determining the working performance corresponding to the regulated model, it further includes:

[0023] Based on the reward and punishment function corresponding to the agent, and determine the reward and punishment value corresponding to the action of regulating the preset parameter according to the working performance, so as to improve the determination process of the regulation action based on the reward and punishment value subsequently.

[0024] In a second aspect, the present application discloses a regulation device for an ESP stratified oil production well simulation model, which is applied to an agent and includes:

[0025] An information acquisition module, configured to acquire the current state information corresponding to the current environment where the ESP stratified oil production well simulation model is located;

[0026] A first determination module, configured to determine a regulation action according to the deep deterministic policy gradient algorithm and based on the current state information; the regulation action is a regulation preset parameter; the preset parameter is one or more of the operating frequency of the motor of the ESP stratified oil production well, the diameter of the wellhead choke, and the opening degree of each layer of inflow control device;

[0027] A regulation module, configured to regulate the ESP stratified oil production well simulation model based on the regulation action;

[0028] A second determination module, configured to determine the working performance corresponding to the regulated model;

[0029] A judgment module, configured to judge whether the working performance meets a preset condition;

[0030] A jump module, configured to, if not satisfied, jump to the step of acquiring the current state information corresponding to the current environment where the ESP stratified oil production well simulation model is located until the working performance meets the preset condition.

[0031] Optionally, the regulation device for the ESP stratified oil production well simulation model further includes:

[0032] A reward and punishment module, configured to determine the reward and punishment value corresponding to the action of regulating the preset parameter based on the reward and punishment function corresponding to the agent and according to the working performance, so as to improve the determination process of the regulation action based on the reward and punishment value subsequently.

[0033] In a third aspect, the present application discloses an electronic device, including:

[0034] A memory for storing a computer program;

[0035] A processor for executing the computer program to implement the regulation method of the ESP stratified oil production well simulation model disclosed above.

[0036] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the regulation method of the ESP stratified oil production well simulation model disclosed above is implemented.

[0037] It can be seen that the present application obtains the current state information corresponding to the current environment of the ESP stratified oil production well simulation model, determines a regulation action according to the deep deterministic policy gradient algorithm and based on the current state information, and regulates the ESP stratified oil production well simulation model based on the regulation action; the regulation action is to regulate a preset parameter; the preset parameter is one or more of the motor operating frequency of the ESP stratified oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device; determines the working performance corresponding to the regulated model; determines whether the working performance meets a preset condition; if not, jumps to the step of obtaining the current state information corresponding to the current environment of the ESP stratified oil production well simulation model until the working performance meets the preset condition. Thus, it can be seen that the present application realizes reinforcement learning through the deep deterministic policy gradient algorithm to determine the parameter regulation process, and obtains the final optimal result at a relatively fast speed, improving the speed and accuracy; the present application changes the working performance by adjusting one or more of the motor operating frequency of the ESP stratified oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device, so as to achieve the optimal working performance. In summary, the present application regulates the ESP stratified oil production well simulation model to improve the working performance of the model. Description of the Drawings

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0039] Figure 1 It is a flowchart of a regulation method for an ESP stratified oil production well simulation model disclosed in the present application;

[0040] Figure 2 It is a flowchart of a specific regulation method for an ESP stratified oil production well simulation model disclosed in the present application;

[0041] Figure 3Schematic diagram of the operating characteristic curve of an electric submersible pump disclosed in this application;

[0042] Figure 4 Schematic diagram of an optimal working area disclosed in this application;

[0043] Figure 5 Schematic diagram of a partial process of reinforcement learning disclosed in this application;

[0044] Figure 6 Schematic diagram of the exploration process of an agent disclosed in this application;

[0045] Figure 7 Schematic diagram of a regulation process disclosed in this application;

[0046] Figure 8 Schematic diagram of the structure of a control device for an electric submersible pump stratified oil production well simulation model disclosed in this application;

[0047] Figure 9 Schematic diagram of the structure of an electronic device disclosed in this application. Detailed implementation manners

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] For the complex and changeable production environment of oil wells and the multi-parameter system collaborative optimization model of non-linear systems, there are problems of slow convergence speed and low prediction accuracy.

[0050] Therefore, the embodiments of this application propose a control scheme for an electric submersible pump stratified oil production well simulation model, which can improve the speed and accuracy of control.

[0051] The embodiments of this application disclose a control method for an electric submersible pump stratified oil production well simulation model. As shown in Figure 1 the following, this method includes:

[0052] Step S11: Obtain the current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model.

[0053] In this embodiment, the current state information includes the motor operating frequency of the electric submersible pump stratified oil production well, the wellhead choke diameter, the opening degree of the inflow control device for each layer, the oil production output of the electric submersible pump, the pump efficiency of the electric submersible pump, the pump inlet pressure, the pump outlet pressure, the oil pressure, and the pressure behind the choke. See Table 1 below:

[0054] Table 1

[0055]

[0056] Step S12: Determine a control action according to the deep deterministic policy gradient algorithm and based on the current state information; the control action is to adjust a preset parameter; the preset parameter is one or more of the motor operating frequency of the electrical submersible pump stratified oil production well, the wellhead choke diameter, and the opening degrees of the inflow control devices of each layer.

[0057] In this embodiment, the determining the control action according to the deep deterministic policy gradient algorithm and based on the current state information includes: determining, according to the deep deterministic policy gradient algorithm and based on the current state information, to adjust the opening degrees of the inflow control devices of each layer to change the oil well inflow capacity; and / or, determining, according to the deep deterministic policy gradient algorithm and based on the current state information, to adjust the wellhead choke diameter to change the pipeline characteristics; and / or, determining, according to the deep deterministic policy gradient algorithm and based on the current state information, to adjust the motor operating frequency.

[0058] It should be noted that there are three methods to improve the performance of the electrical submersible pump stratified oil production well simulation model: one is to coordinate the supply and discharge relationship by adjusting the opening degree of the downhole ICD (Inflow Control Device) in the well, that is, by changing the oil well inflow capacity to improve the pump efficiency of the lifting system; the second is to improve the system efficiency of the electrical submersible pump well by adjusting the wellhead choke diameter, that is, by changing the pipeline characteristics; the third is to adopt a variable frequency adjustment method to optimize the production system working condition by changing the working characteristic curve of the submersible pump, thereby improving the system efficiency of the electrical submersible pump well. In actual production, it is necessary to jointly adjust the three methods so that the electrical submersible pump well is in the optimal pump efficiency within a certain range of liquid production. It should be noted that it may not be the optimal pump efficiency, but the pump efficiency value closest to the optimal pump efficiency in all cases.

[0059] It should be noted that the number of parameters of the preset parameter corresponding to different control actions and / or the degree of parameter change of the preset parameter are different.

[0060] Step S13: Regulate the electrical submersible pump stratified oil production well simulation model based on the control action, and determine the working performance corresponding to the regulated model.

[0061] In this embodiment, after determining the working performance corresponding to the regulated model, it further includes: determining the reward and punishment value corresponding to the action of the regulated preset parameter based on the reward and punishment function corresponding to the intelligent agent and according to the working performance, so as to improve the determination process of the control action based on the reward and punishment value subsequently.

[0062] It should be noted that the existence of the reward and punishment function enables the intelligent agent to clarify which actions are beneficial and which are not. After that, when determining the control actions, it can try to avoid unfavorable actions as much as possible.

[0063] Step S14: Determine whether the working performance meets the preset conditions.

[0064] Step S15: If not, jump to the step of obtaining the current state information corresponding to the current environment of the ESP layered oil production well simulation model until the working performance meets the preset conditions.

[0065] It can be seen that in this application, the current state information corresponding to the current environment of the ESP layered oil production well simulation model is obtained, the control action is determined according to the deep deterministic policy gradient algorithm and based on the current state information, and the ESP layered oil production well simulation model is controlled based on the control action; the control action is to control the preset parameters; the preset parameters are one or more of the motor operating frequency of the ESP layered oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device; determine the working performance corresponding to the model after regulation; determine whether the working performance meets the preset conditions; if not, jump to the step of obtaining the current state information corresponding to the current environment of the ESP layered oil production well simulation model until the working performance meets the preset conditions. Thus, it can be seen that this application realizes reinforcement learning through the deep deterministic policy gradient algorithm to determine the parameter regulation process, and obtains the final optimal result at a relatively fast speed, improving the speed and accuracy; this application changes the working performance by adjusting one or more of the motor operating frequency of the ESP layered oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device so as to achieve the optimal working performance. In summary, this application regulates the ESP layered oil production well simulation model to improve the working performance of the model.

[0066] The embodiment of the present application discloses a specific method for regulating an ESP layered oil production well simulation model. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution. See Figure 2 as shown, specifically including:

[0067] Step S21: Obtain the current state information corresponding to the current environment of the ESP layered oil production well simulation model.

[0068] Step S22: Determine the control action according to the deep deterministic policy gradient algorithm and based on the current state information; the control action is to control the preset parameters; the preset parameters are one or more of the motor operating frequency of the ESP layered oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device.

[0069] Step S23: Based on the regulation action, regulate the electric submersible pump stratified oil production well simulation model, and determine the oil production rate of the electric submersible pump and the pump efficiency of the electric submersible pump corresponding to the regulated model.

[0070] In this embodiment, it is necessary to determine the oil production rate of the electric submersible pump and the pump efficiency of the electric submersible pump corresponding to the regulated model, so as to further determine whether further regulation is required based on the oil production rate of the electric submersible pump and the pump efficiency of the electric submersible pump.

[0071] Step S24: Judge whether the oil production rate of the electric submersible pump is within the preset production range, and judge whether the pump efficiency of the electric submersible pump is the best pump efficiency corresponding to the regulated environment; if so, it is determined that the preset conditions are met, and if not, it is determined that the preset conditions are not met.

[0072] In this embodiment, the preset production range is preset, and can be specifically set according to the actual situation and can be changed.

[0073] In this embodiment, the process of judging whether the pump efficiency of the electric submersible pump is the best pump efficiency corresponding to the regulated environment can also be to judge whether the pump efficiency of the electric submersible pump is the pump efficiency closest to the best pump efficiency corresponding to the regulated environment.

[0074] See Figure 3 shown, which is a schematic diagram of the working characteristic curve of an electric submersible pump; the working characteristic curve of the electric submersible pump is an important basis for pump selection design and production operation optimization. Figure 3 It shows that when the operating frequency is 60Hz, the working characteristic curve reflects the relationship between flow rate, pump power, pump efficiency and head (where, Head represents head (unit: ft), Efficiency represents pump efficiency (unit: %), Power represents pump power (unit: hp)). In the figure, the abscissa represents the flow rate (Flowrate). From top to bottom on the left, the first line represents the head, the second line represents the pump power, and the third line represents the pump efficiency (pump efficiency). Among them, the highest point of the third line represents the best pump efficiency when the operating frequency is 60Hz. By determining the best pump efficiency at different frequencies, the content shown in Figure 4 can be obtained. Figure 4 is a schematic diagram of the best working area (the abscissa Flowrate is the flow rate, Head is the depth). In the figure, the best pump efficiency lines corresponding to each operating efficiency are used as the center to determine the best working area (the middle dotted line part is the best pump efficiency line, the upper dotted line area is the flow rate lower limit line, and the lower dotted line area is the flow rate upper limit line). Within the working range area, the electric pump operates in a high pump efficiency range. When the production point is not in this interval, the pump efficiency is too low, and long-term operation outside the interval will cause faults. Therefore, the goal of optimizing the electric submersible pump is to make the production point always fall into the area near the optimal pump efficiency line, and at this time, the pump efficiency of the electric pump well is the highest. It should be noted that Figure 4The intelligent agent exploration process is also the process of model regulation. During the intelligent agent exploration process, the Initial point is the starting point of the process, and the End point is the ending point of the process. The ending part is located on the optimal pump efficiency line or infinitely close to the optimal pump efficiency line.

[0075] Step S25: If not satisfied, jump to the step of obtaining the current state information corresponding to the current environment where the ESP stratified oil production well simulation model is located until the working performance meets the preset conditions.

[0076] It can be seen that this application obtains the current state information corresponding to the current environment where the ESP stratified oil production well simulation model is located; determines the regulation action according to the deep deterministic policy gradient algorithm and based on the current state information; the regulation action is to regulate the preset parameters; the preset parameters are one or more of the motor operating frequency of the ESP stratified oil production well, the wellhead choke diameter, and the opening degrees of the inflow control devices of each layer; regulates the ESP stratified oil production well simulation model based on the regulation action, and determines the oil production rate of the ESP and the pump efficiency of the ESP corresponding to the regulated model; determines whether the oil production rate of the ESP is within the preset production range, and determines whether the pump efficiency of the ESP is the optimal pump efficiency corresponding to the regulated environment; if so, it is determined that the preset conditions are met, and if not, it is determined that the preset conditions are not met; if not satisfied, jump to the step of obtaining the current state information corresponding to the current environment where the ESP stratified oil production well simulation model is located until the working performance meets the preset conditions. Thus, it can be seen that this application uses the oil production rate of the ESP and the pump efficiency of the ESP to determine whether to continue regulating the model, and can obtain the oil production rate of the ESP and the pump efficiency of the ESP that meet the requirements, which is beneficial to improving the regulation efficiency and accuracy.

[0077] In a specific embodiment, the oil well production is adjusted by the coordinated optimization of the electric pump frequency, the wellhead choke diameter, and the opening degrees of each ICD to improve the system efficiency. The DDPG (Deep Deterministic Policy Gradient) reinforcement learning algorithm is used. The basic elements of reinforcement learning include state Status, action Action, policy Policy, and reward function Reward. Through the communication and feedback between the intelligent agent AGENT and the environment Environment, the optimal policy is learned from random continuous attempts, and a machine learning method for obtaining the maximum long-term cumulative reward is used. Its process can be described by the Markov decision process (MDP), which is generally represented by a 5-tuple (S, A, P, R, γ), where S is the state space, A is the action space, P is the state transition probability, R is the reward function, and γ is the discount factor. Since the Markov decision process is prior art, the use of the state transition probability, the reward function, and the discount factor will not be specifically introduced.

[0078] Specifically, first, a DDPG-environment will be constructed, that is, an electric submersible pump stratified oil production well simulation model will be constructed; then, the DDPG-state will be determined, that is, the current state information corresponding to the current environment will be determined. The state space of the specific environment is expressed as: , where that is , where the meanings represented by each symbol are shown in Table 1. During the training process, the agent observes its current state, selects an action according to the current policy, and will change accordingly after being executed in the environment, and then gives the environmental state information at the next moment, thereby outputting a new state; in addition, the DDPG-action is determined. The specific action space A is the decision-making quantity of the model and is the next action obtained by the agent according to the state at the current moment. In this application, the actions are to adjust the operating frequency of the motor, the opening of the wellhead choke and each ICD. The action at time t is , which can be expressed as , as Figure 5 shown, after the agent makes an action, the electric pump simulation of the separate production well gives the environmental variables after the execution of the action. Figure 5 is a schematic diagram of a part of the reinforcement learning process (the separate production electric pump well simulation, that is, the electric submersible pump stratified oil production well simulation model has a simulation interface for interacting with the agent (decision feedback)); finally, the DDPG-reward is determined. In reinforcement learning, the reward comes from the environment. Facing different tasks, the reward function needs to be carefully designed according to the task characteristics and environmental state. The exploration and decision-making process of the electric pump well agent is as Figure 6 shown. Figure 6 is a schematic diagram of the agent exploration process. In this application, the agent (AGENT) is used to determine actions based on the state and perform rewards and punishments (including rewards and punishments) based on the feedback of the environment after the actions are regulated. The goal of the agent exploration is to ensure the highest electric pump efficiency under the premise of ensuring the allocated production of each production layer.

[0079] In a specific embodiment, referring to Figure 7 shown, it is a schematic diagram of a regulation process; first, the reservoir address parameters (oil well liquid production capacity) and target parameters (ideal production and pump efficiency) are determined in advance, and the current state information corresponding to the current environment is determined. According to the deep deterministic policy gradient algorithm, the regulation actions are determined to train the regulation optimization model, and the reward and punishment values for each training are determined based on the reward and punishment function, so as to improve the regulation actions based on the reward and punishment values in the subsequent process. After the regulation is completed based on the regulation actions, the new pump efficiency and production are determined. If the pump efficiency and production meet the requirements, the regulated state information at this time is the optimal. If not, it jumps to the step of determining the regulation actions according to the deep deterministic policy gradient algorithm to train the regulation optimization model.

[0080] In summary, the object of the agent's learning in the DDPG algorithm is the mapping function of the environment-action. It does not require discretization of the control actions, can handle continuous action and state spaces, and is compared with the classical intelligent algorithm - Particle Swarm Optimization (PSO). It has strong optimization and generalization capabilities, verifies the superiority of the DDPG algorithm for optimizing production control parameters, and has great application potential in solving complex problems.

[0081] Correspondingly, the embodiment of the present application also discloses a control device for an ESP layered oil production well simulation model. Refer to Figure 8 As shown, the device includes:

[0082] An information acquisition module 11, configured to acquire the current state information corresponding to the current environment where the ESP layered oil production well simulation model is located;

[0083] A first determination module 12, configured to determine a control action according to the deep deterministic policy gradient algorithm and based on the current state information; the control action is to control a preset parameter; the preset parameter is one or more of the motor operating frequency of the ESP layered oil production well, the wellhead choke diameter, and the opening degree of each layer inflow control device;

[0084] A control module 13, configured to control the ESP layered oil production well simulation model based on the control action;

[0085] A second determination module 14, configured to determine the working performance corresponding to the model after control;

[0086] A judgment module 15, configured to judge whether the working performance meets a preset condition;

[0087] A jump module 16, configured to, if not satisfied, jump to the step of acquiring the current state information corresponding to the current environment where the ESP layered oil production well simulation model is located until the working performance meets the preset condition.

[0088] Optionally, the control device for the ESP layered oil production well simulation model further includes:

[0089] A reward and punishment module 17, configured to determine the reward and punishment value corresponding to the action of controlling the preset parameter based on the reward and punishment function corresponding to the agent and according to the working performance, so as to improve the determination process of the control action based on the reward and punishment value subsequently.

[0090] Wherein, for the more specific working processes of the above various modules, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not described herein again.

[0091] It can be seen that the present application obtains the current state information corresponding to the current environment of the ESP stratified oil production well simulation model, determines a control action according to the deep deterministic policy gradient algorithm and based on the current state information, and controls the ESP stratified oil production well simulation model based on the control action; the control action is to adjust preset parameters; the preset parameters are one or more of the motor operating frequency of the ESP stratified oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device; determines the working performance corresponding to the adjusted model; determines whether the working performance meets a preset condition; if not, jumps to the step of obtaining the current state information corresponding to the current environment of the ESP stratified oil production well simulation model until the working performance meets the preset condition. Thus, the present application realizes reinforcement learning through the deep deterministic policy gradient algorithm to determine the parameter adjustment process, and obtains the final optimal result at a relatively fast speed, improving the speed and accuracy; the present application changes the working performance by adjusting one or more of the motor operating frequency of the ESP stratified oil production well, the wellhead choke diameter, and the opening degree of each layer of inflow control device, so as to achieve the optimal working performance. In summary, the present application controls the ESP stratified oil production well simulation model to improve the working performance of the model.

[0092] Further, an embodiment of the present application also provides an electronic device. Figure 9 It is a structural diagram of an electronic device 20 shown according to an exemplary embodiment, and the content in the figure cannot be considered as any limitation on the scope of use of the present application.

[0093] Figure 9 It is a structural schematic diagram of an electronic device 20 provided by an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the control method of the ESP stratified oil production well simulation model disclosed in any of the foregoing embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0094] In this embodiment, the power supply 26 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 25 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows is any communication protocol applicable to the technical solution of the present application, and no specific limitation is imposed on it here; the input / output interface 24 is used to obtain external input data or output data to the outside, and its specific interface type can be selected according to specific application needs, and no specific limitation is made here.

[0095] In addition, as a carrier for storing resources, the memory 22 can be a read-only memory, a random access memory, a magnetic disk, an optical disk, etc. The resources stored thereon can include a computer program 221, and the storage method can be transient storage or permanent storage. Among them, in addition to the computer program that can be used to complete the regulation method of the electric submersible pump stratified oil production well simulation model executed by the electronic device 20 disclosed in any of the foregoing embodiments, the computer program 221 can further include computer programs that can be used to complete other specific tasks.

[0096] Furthermore, an embodiment of the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the regulation method of the electric submersible pump stratified oil production well simulation model disclosed above is implemented.

[0097] Regarding the specific steps of this method, reference can be made to the corresponding content disclosed in the foregoing embodiments, and details will not be elaborated here.

[0098] The various embodiments in this application are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts between the various embodiments, reference can be made to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and reference can be made to the description in the method part for related parts.

[0099] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0100] The steps of the method or algorithm described in combination with the embodiments disclosed in this article can be directly implemented by hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0101] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0102] The above has introduced in detail a method, device, equipment, and storage medium for regulating an electrical submersible pump stratified oil production well simulation model provided by the present application. Specific examples are used in this text to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A control method for a simulation model of an electric submersible pump stratified oil production well, characterized in that: Applied to intelligent agents, including: Obtain current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model; Determine a control action according to a deep deterministic policy gradient algorithm and based on the current state information; the control action is to control preset parameters; the preset parameters are one or more of the motor operating frequency of the electric submersible pump stratified oil production well, the wellhead nozzle diameter and the opening of each layer inflow control device; Regulating the electric submersible pump stratified oil production well simulation model based on the regulation action, and determining the corresponding working performance of the regulated model; Determining whether the working performance meets the preset conditions; If not, jump to the step of obtaining the current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model until the working performance meets the preset condition.

2. The control method of the electric submersible pump stratified oil production well simulation model according to claim 1, characterized in that: The determining of the working performance corresponding to the adjusted model includes: Determine the ESP oil production and ESP pump efficiency corresponding to the model after regulation.

3. The control method of the electric submersible pump stratified oil production well simulation model according to claim 2, characterized in that: The determining whether the working performance meets the preset conditions includes: Determine whether the oil production of the electric submersible pump is within a preset production range, and determine whether the pump efficiency of the electric submersible pump is the optimal pump efficiency corresponding to the regulated environment; If so, it is determined that the preset condition is met, and if not, it is determined that the preset condition is not met.

4. The control method of the electric submersible pump stratified oil production well simulation model according to claim 1, characterized in that: The current status information includes the motor operating frequency of the electric submersible pump stratified oil production well, the wellhead oil nozzle diameter, the opening of the inflow control device of each layer, the electric submersible pump oil production, the electric submersible pump efficiency, the pump inlet pressure, the pump outlet pressure, the oil pressure, and the pressure behind the nozzle.

5. The control method of the electric submersible pump stratified oil production well simulation model according to claim 1, characterized in that: The determining the control action according to the deep deterministic policy gradient algorithm and based on the current state information includes: Determine and adjust the opening of the inflow control device of each layer according to the deep deterministic policy gradient algorithm and based on the current state information to change the inflow capacity of the oil well; and / or, determining to adjust the wellhead nozzle diameter according to a deep deterministic policy gradient algorithm and based on the current state information to change the pipeline characteristics; And / or, determining to adjust the operating frequency of the motor according to a deep deterministic policy gradient algorithm and based on the current state information.

6. The control method of the electric submersible pump stratified oil production well simulation model according to any one of claims 1 to 5, characterized in that: After determining the working performance corresponding to the adjusted model, the method further includes: Based on the reward and punishment function corresponding to the intelligent agent, the reward and punishment value corresponding to the action of regulating the preset parameters is determined according to the work performance, so as to subsequently improve the determination process of the regulating action based on the reward and punishment value.

7. A control device for a simulation model of an electric submersible pump stratified oil production well, characterized in that: Applied to intelligent agents, including: An information acquisition module is used to obtain current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model; A first determination module is used to determine a control action according to a deep deterministic policy gradient algorithm and based on the current state information; the control action is to control a preset parameter; the preset parameter is one or more of a motor operating frequency of an electric submersible pump stratified oil production well, a wellhead nozzle diameter, and an opening of an inflow control device for each layer; A control module, for controlling the electric submersible pump stratified oil production well simulation model based on the control action; The second determination module is used to determine the working performance corresponding to the adjusted model; A judgment module, used to judge whether the working performance meets the preset conditions; The jump module is used to jump to the step of obtaining the current state information corresponding to the current environment of the electric submersible pump stratified oil production well simulation model if the condition is not met, until the working performance meets the preset condition.

8. The control device for the simulation model of an electric submersible pump stratified oil production well according to claim 7, characterized in that: Also includes: The reward and punishment module is used to determine the reward and punishment value corresponding to the action of regulating the preset parameters based on the reward and punishment function corresponding to the intelligent agent and according to the work performance, so as to subsequently improve the determination process of the regulation action based on the reward and punishment value.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor is used to execute the computer program to implement the control method of the electric submersible pump stratified oil production well simulation model as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: Used to store computer programs; wherein, when the computer program is executed by a processor, the control method of the electric submersible pump stratified oil production well simulation model as described in any one of claims 1 to 6 is implemented.

Images