Production well production operation parameter optimization decision-making method based on particle swarm

Abstract

The invention discloses a production well production operation parameter optimization decision-making method based on a particle swarm, and the method comprises the following steps: determining control parameters which affect a layered liquid production profile of an electric submersible pump according to an oil well multi-layer separate production process; historical production data of an oil producing well are obtained, and well layering liquid producing profiles under the action of different control parameters are counted; performing data preprocessing on the obtained historical data to complete cleaning and correction of the original data; based on an XGBoost algorithm, an electric pump well fluid production profile prediction model is established; and based on a particle swarm optimization algorithm, the intelligent regulation and control method for the layered liquid production profile of the electric submersible pump is established. The problems that in the oil extraction process, the recovery ratio is difficult to control, and inter-well interference is difficult to control are solved, the pump efficiency of the oil extraction pump is improved, it means that underground oil well liquid is conveyed to the ground more efficiently, and oil field extraction is more economical and efficient. By improving the pump efficiency of the oil extraction pump, the energy waste and the production cost can be reduced, the productivity and the oil extraction efficiency are improved, and the profit and the value of an oil field are finally improved.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas production in petroleum engineering, and in particular relates to a particle swarm-based method for optimizing decision-making of production operation parameters of oil wells. Background Art

[0002] In actual production, due to the large differences between different oil layers, the oil production capacity of non-main layers has not been effectively utilized, so the total oil production is far lower than expected. In actual mining, the high-pressure layer will inhibit the production of the low-pressure layer, reducing the oil production of the low-pressure layer and the total production. In order to give full play to the oil production capacity of each layer, it is necessary to make reasonable decisions on the production pressure difference and the lifting parameters of the electric pump to allocate production and injection, so as to change the production pressure difference, which is mainly achieved by changing the opening of the production distributors of each layer in the well.

[0003] Therefore, at this stage, there is an urgent need for an optimization decision-making method for oil well production operation parameters based on particle swarm. Under the condition of achieving the production allocation target, multiple factors such as the opening of each layer's production distributor, the nozzle opening, and the frequency of the electric pump are comprehensively considered to maximize the pump efficiency, thereby improving equipment utilization efficiency, reducing equipment loss, and extending the service life of the equipment. Summary of the invention

[0004] The problem to be solved by the present invention is to provide a particle swarm-based method for optimizing the decision-making of production operation parameters of oil wells. This method enables production allocation between different formations with different production rates, and allows the pump efficiency to be as close to the optimal pump efficiency as possible within the normal working range, which is of great significance for guiding the work of on-site technicians.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a particle swarm-based oil well production operation parameter optimization decision method, comprising the following steps:

[0006] S1: Determine the control parameters that affect the stratified production profile of the electric submersible pump according to the multi-layer production process of the oil well;

[0007] S2: Obtain historical production data of oil wells and calculate the well layered production profiles under different control parameters;

[0008] S3: Perform data preprocessing on the acquired historical data to complete the cleaning and correction of the original data;

[0009] S4: Based on the XGBoost algorithm, a prediction model for the fluid production profile of electric pump wells was established;

[0010] S5: Based on the particle swarm optimization algorithm, an intelligent control method for the stratified production profile of an electric submersible pump is established.

[0011] Furthermore, in S1, the control parameters include pump frequency, dispenser opening and nozzle opening.

[0012] Furthermore, in S2, historical data of production wells are statistically analyzed to establish a mapping data set between well stratified fluid production and surface and downhole control parameters, P = [P 1 , P 2 , ... P t ], P 1 , P 2 , ... P t The liquid production at each moment is represented in sequence, C = [C 1 , C 2 , ... C t ],C t =[f t ,x t ,y t ],C t is the control parameter at time t, where f t is the pump frequency, x t is the opening degree of the distributor, y t The oil nozzle opening.

[0013] Furthermore, in S3, the historical data includes valid data and abnormal data.

[0014] Furthermore, when data is missing, outlier detection, data normalization, and interpolation are used to obtain f(x i ) to process the existing data, the calculation formula is as follows:

[0015]

[0016] Furthermore, in S4, the stratified liquid production is used as the objective function obj, the pump frequency, the production distributor opening, and the nozzle opening are used as the model input parameters, and the historical data set of the production wells is used to train and verify the model.

[0017] Further, the S5 comprises the following steps:

[0018] S51: constructing a functional relation between the input to the target in the particle swarm algorithm, with the stratified liquid production as the target function obj;

[0019] S52: Establishing a fitness function based on stratified production allocation error;

[0020] S53: Setting model constraints and boundary conditions;

[0021] S54: The model is solved after multiple iterations.

[0022] Furthermore, the present invention also provides a device for running the above data processing method.

[0023] Furthermore, the present invention also provides a device, comprising a memory, a processor, and an algorithm stored in the memory and executable on the processor, wherein the processor implements the above-mentioned data processing method when executing the computer program.

[0024] Furthermore, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer algorithm, and the computer algorithm implements the above-mentioned data processing when executed by a processor.

[0025] The advantages and positive effects of the present invention are:

[0026] 1. While achieving the production allocation target, the present invention comprehensively considers multiple factors such as the opening of each layer of the production allocation device, the nozzle opening, the electric pump frequency, etc., so as to maximize the pump efficiency, thereby improving the equipment utilization efficiency, reducing equipment loss, and extending the service life of the equipment.

[0027] 2. The test results of the present invention show that the algorithm model has high accuracy, good stability, fast running speed, good practicality and stability. It makes up for the shortcomings of the existing technology and proposes a feasible solution to further solve the problem of difficult to control recovery rate and well interference in the oil production process. It improves the pump efficiency of the oil production pump, which means that the underground oil well liquid is more efficiently transported to the ground, making the exploitation of the oil field more economical and efficient. By improving the pump efficiency of the oil production pump, energy waste and production costs can be reduced, while productivity and oil production efficiency can be improved, thereby ultimately increasing the profit and value of the oil field. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0030] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0031] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0032] The embodiments of the present invention are further described below in conjunction with the accompanying drawings:

[0033] like Figure 1 As shown, a particle swarm-based oil well production operation parameter optimization decision method includes the following steps:

[0034] S1: Determine the control parameters that affect the stratified production profile of the electric submersible pump according to the multi-layer production process of the oil well. Specifically, the control parameters include pump frequency, production distributor opening and nozzle opening.

[0035] S2: Obtain the historical production data of the oil wells and calculate the well layered production profiles under different control parameters. Specifically, statistically analyze the historical data of the production wells and establish a mapping data set between the well layered production and the ground and downhole control parameters. 1 , P 2 , ... P t ], P 1 , P 2 , ... P t The liquid production at each moment is represented in sequence, C = [C 1 , C 2 , ... C t ],C t =[f t ,x t ,y t ],C t is the control parameter at time t, where f t is the pump frequency, x t is the opening degree of the distributor, y t The oil nozzle opening.

[0036] S3: Perform data preprocessing on the acquired historical data to clean and correct the original data. Specifically, the measured data contains both real and valid data and abnormal data, and may be missing data. It is necessary to use outlier detection, data normalization, and interpolation methods f(x i ) to process the existing data, the calculation formula is as follows:

[0037]

[0038] S4: Based on the XGBoost (decision tree ensemble algorithm under the Gradient Boosting framework) algorithm, a prediction model for the liquid production profile of electric pump wells is established. Specifically, the stratified liquid production is used as the objective function obj, the pump frequency, the opening degree of the distributor, and the nozzle opening degree are used as the model input parameters, and the historical data set of the production wells is used to train and verify the model.

[0039] S5: Based on the particle swarm optimization algorithm, an intelligent control method for the stratified production profile of an electric submersible pump is established. Specifically, S5 includes the following steps:

[0040] S51: construct a functional relationship between the input and the target in the particle swarm algorithm, with the stratified liquid production as the target function obj.

[0041] S52: Establish a fitness function based on stratified production allocation error.

[0042] S53: Set model constraints and boundary conditions.

[0043] S54: The model is solved after multiple iterations.

[0044] The present invention is specifically described below by taking the actual application in the oil production well of A oil field as an example:

[0045] The well is divided into three layers for production allocation, so the control parameters have three production allocation device openings. After the production well runs stably with fixed parameters, the production liquid volume is counted, as shown in Table 1.

[0046] Table 1 Corresponding output under different control parameters

[0047]

[0048] The established electric pump well production profile prediction model is used for real-time prediction. The pump frequency, the opening of the three production distributors and the nozzle opening of the production well are obtained in real time, and the input value is used to train the model to predict the liquid production, as shown in Table 2.

[0049] Table 2. Some prediction values ​​of XGBoost

[0050]

[0051] The target output of the three layers is set to 240m 3 / d, 240m 3 / d, 240m 3 / d, after optimization by the established intelligent control method of ESP layered production profile based on particle swarm optimization algorithm, the optimal combination of frequency, three-layer distributor opening, and nozzle diameter is 50Hz, the three-layer distributor opening is 40.5%, 56.5%, 89.8%, and the nozzle diameter is 14.5mm. As shown in Table 3, the production allocation error meets the production requirements.

[0052] Table 3 Test data

[0053]

[0054] The advantages and positive effects of the present invention are:

[0055] 1. While achieving the production allocation target, the present invention comprehensively considers multiple factors such as the opening of each layer of the production allocation device, the nozzle opening, the electric pump frequency, etc., so as to maximize the pump efficiency, thereby improving the equipment utilization efficiency, reducing equipment loss, and extending the service life of the equipment.

[0056] 2. The test results of the present invention show that the algorithm model has high accuracy, good stability, fast running speed, good practicality and stability. It makes up for the shortcomings of the existing technology and proposes a feasible solution to further solve the problem of difficult to control recovery rate and well interference in the oil production process. It improves the pump efficiency of the oil production pump, which means that the underground oil well liquid is more efficiently transported to the ground, making the exploitation of the oil field more economical and efficient. By improving the pump efficiency of the oil production pump, energy waste and production costs can be reduced, while productivity and oil production efficiency can be improved, thereby ultimately increasing the profit and value of the oil field.

[0057] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A particle swarm-based method for optimizing the decision-making of production and operation parameters of oil wells, characterized in that: The following steps are included: S1: Determine the control parameters that affect the stratified production profile of the electric submersible pump according to the multi-layer production process of the oil well; S2: Obtain historical production data of oil wells and calculate the well layered production profiles under different control parameters; S3: Perform data preprocessing on the acquired historical data to complete the cleaning and correction of the original data; S4: Based on the XGBoost algorithm, a prediction model for the fluid production profile of electric pump wells was established; S5: Based on the particle swarm optimization algorithm, an intelligent control method for the stratified production profile of an electric submersible pump is established.

2. The particle swarm-based method for optimizing the production and operation parameters of oil wells according to claim 1, characterized in that: In S1, the control parameters include pump frequency, dispenser opening and nozzle opening.

3. The particle swarm-based method for optimizing the production and operation parameters of oil wells according to claim 1 or 2, characterized in that: In S2, historical data of production wells are statistically analyzed to establish a mapping data set between well stratified fluid production and surface and downhole control parameters, P = [P1, P2, ... P t ], P1, P2, ... P t represents the liquid production at each moment, C = [C1, C2, ... C t ],C t =[f t ,x t ,y t ],C t is the control parameter at time t, where f t is the pump frequency, x t is the opening degree of the distributor, y t The oil nozzle opening.

4. The particle swarm-based method for optimizing the production and operation parameters of oil wells according to claim 1 or 2, characterized in that: In S3, the historical data includes valid data and abnormal data.

5. The particle swarm-based method for optimizing the decision-making of production and operation parameters of oil wells according to claim 4, characterized in that: When data is missing, outlier detection, data normalization, and interpolation methods are used. i ) to process the existing data, the calculation formula is as follows:

6. The particle swarm-based method for optimizing the production and operation parameters of oil wells according to claim 1 or 2, characterized in that: In S4, the stratified liquid production is used as the objective function obj, the pump frequency, the production distributor opening, and the nozzle opening are used as the model input parameters, and the historical data set of the production wells is used to train and verify the model.

7. The particle swarm-based method for optimizing the production and operation parameters of oil wells according to claim 1 or 2, characterized in that: The S5 comprises the following steps: S51: constructing a functional relation between the input to the target in the particle swarm algorithm, with the stratified liquid production as the target function obj; S52: Establishing a fitness function based on stratified production allocation error; S53: Setting model constraints and boundary conditions; S54: The model is solved after multiple iterations.

8. A device, characterized in that: Run the data processing method according to any one of claims 1 to 7.

9. A device comprising a memory, a processor, and an algorithm stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the data processing method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer algorithm, characterized in that: When the computer algorithm is executed by a processor, the data processing according to any one of claims 1 to 7 is implemented.

Images