Intelligent intermittent opening method and system for rod-pumped well

Abstract

The invention discloses an intelligent intermittent opening method and system for a rod-pumped well. The method comprises the following steps: calculating the working frequency of closing the oil pumping unit by using a stroke sampling method and a relative deviation method, and setting the oil well oil pressure of opening the oil pumping unit; influence factors of the working frequency of the oil pumping unit are obtained; constructing an oil pumping unit working frequency prediction model, constructing a data set based on the influence factors of the oil pumping unit working frequency, and obtaining a working frequency prediction value of the oil pumping unit by using the trained oil pumping unit working frequency prediction model; and controlling the oil pumping unit to intelligently turn on based on the working frequency when the oil pumping unit is turned off, the oil well oil pressure when the oil pumping unit is turned on and the working frequency predicted value of the oil pumping unit. The problem that the switching time of a rod-pumped well is not accurate enough can be solved, and the problems that the pumping capacity of a pumping unit is larger than the actual liquid supply capacity of an oil well, so that the pump efficiency is poor, electric energy is wasted, and motor equipment is damaged can be solved.

Description

Technical Field

[0001] The invention relates to the field of intelligent control of oil pumping units, and in particular to an intelligent spacing method and system for oil pumping wells. Background Art

[0002] Traditional oil wells are often operated in a continuous manner. Although this method can continuously produce crude oil, it may also lead to energy waste and increased equipment wear. In order to overcome these problems, the intelligent splicing method and system for oil wells came into being. This method integrates advanced sensors, controllers and communication technologies to achieve real-time monitoring and intelligent control of the operating status of the oil well. According to the actual production situation and geological conditions of the oil well, the intelligent splicing system can automatically adjust the operating parameters and working hours of the oil well to achieve the best mining effect and energy utilization efficiency.

[0003] The existing intelligent intermittent method of pumping wells sets a threshold value during the downstroke and uses a controller to shut down the pumping unit. That is, when dry pumping occurs, the actual current will drop below this value, and the controller will shut down the pumping unit. In addition, the existing method also detects the average current of the motor, identifies the occurrence of dry pumping from the decrease of the actual average current, and then uses the controller to shut down the pumping unit. However, the actual motor current is affected by the counterweight of the pumping unit, which is not something that can be achieved with such a simple criterion, which will lead to inaccurate switching timing of the pumping well. And when the pumping unit is in working state, if its operating frequency is always at a fixed frequency, the pumping capacity of the pumping unit will be greater than the actual fluid supply capacity of the oil well, resulting in poor pump efficiency and long ineffective working time, which not only wastes a lot of electric energy, but also increases the damage probability and maintenance cost of the pumping equipment. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method and system for intelligent spacing of pumping wells.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: A method for intelligently spacing out a pumping well comprises the following steps: S1. Collect historical power consumption data and operating condition data of the oil pumping unit, calculate the operating frequency of the oil pumping unit shutdown by using the stroke sampling method and the relative deviation method based on the historical power consumption data and operating condition data of the oil pumping unit, and set the oil pressure of the oil well when the oil pumping unit is turned on; S2. Collecting multi-index production data of the oil well, and obtaining influencing factors of the operating frequency of the pumping unit according to the multi-index production data of the oil well; S3, constructing a pumping unit operating frequency prediction model, constructing a data set based on the influencing factors of the pumping unit operating frequency in step S2, using the data set to train the pumping unit operating frequency prediction model, and using the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value; S4. Based on the working frequency of the pumping unit when it is turned off in step S1, the oil pressure of the oil well when the pumping unit is turned on, and the predicted value of the working frequency of the pumping unit taken in step S3, the pumping unit is controlled to be intelligently switched on and off.

[0006] Further, in step S1, based on the historical power consumption data of the oil pumping unit and the operating condition data of the oil pumping unit, the operating frequency of the oil pumping unit shutdown is calculated by using the stroke sampling method and the relative deviation method, including the following steps: A1. Based on the historical power consumption data and operating condition data of the oil pumping unit, the standard value of the operating frequency of the oil pumping unit dry pumping cycle is determined by using the stroke sampling method; A2, based on the historical power consumption data and working condition data of the oil pumping unit, and according to the standard value of the working frequency of the oil pumping unit during the dry pumping cycle in step A1, the stroke difference and the difference change rate of the oil pumping unit during the dry pumping cycle are calculated using the relative deviation method; A3. Based on the historical power consumption data and operating condition data of the oil pumping units, a membership matrix of the stroke difference and a membership matrix of the difference change rate are constructed; A4. Calculate the shut-down operating frequency of the oil pumping unit according to the stroke difference and difference change rate of the oil pumping unit dry pumping cycle in step A2 and the membership matrix of the stroke difference and the membership matrix of the difference change rate in step A3.

[0007] Further, in step A4, the operating frequency of the pumping unit shutdown is calculated based on the stroke difference and difference change rate of the pumping unit dry pumping cycle in step A2 and the membership matrix of the stroke difference and the membership matrix of the difference change rate in step A3, which is expressed as:

[0008] in: is the working frequency of the oil pump shut down, is the correction factor for the pumping unit shutdown frequency, is the stroke difference of the dry pumping cycle of the pumping unit, is the membership matrix of stroke difference, is the difference change rate of the dry pumping cycle of the pumping unit, is the membership matrix of the difference change rate.

[0009] Furthermore, in step S2, the factors affecting the operating frequency of the oil pump include the page height, casing pressure, pump efficiency and flow rate of the oil well.

[0010] Further, in step S3, the pumping unit operating frequency prediction model includes a prediction sub-model and an evaluation sub-model.

[0011] Further, in step S3, the pumping unit operating frequency prediction model is trained using the data set, including the following steps: B1. Input the historical data of the data set into the prediction sub-model to obtain the training prediction value of the pumping unit operating frequency; B2, judging whether the error between the training prediction value of the operating frequency of the oil pumping unit in step B1 and the historical true value of the operating frequency of the oil pumping unit is greater than the maximum error value; if so, proceeding to step B3, otherwise returning to step B2; B3. Use the evaluation sub-model to update the prediction sub-model until the set number of training iterations is reached.

[0012] Further, step B1 includes the following steps: B11. Input the historical data of the data set into the input layer, and use the input layer and hidden layer to get the output value of the hidden layer, expressed as:

[0013] in: is the hidden layer output value, The neurons in the hidden layer The activation function, is the neuron number in the input layer, is the total number of neurons in the input layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the threshold of the input layer and the hidden layer, is the neuron number in the hidden layer, is the total number of neurons in the hidden layer; B12. According to the hidden layer output value in step B11, the hidden layer and the output layer are used to obtain the training prediction value of the pumping unit operating frequency, which is expressed as:

[0014] in: is the predicted value for the pumping unit operating frequency training, is the weight between the hidden layer and the output layer, is the threshold between the hidden layer and the output layer, is the neuron number in the output layer, is the total number of neurons in the output layer.

[0015] Further, step B3 includes the following steps: B31. Use the evaluation sub-model to update the weights and thresholds between the hidden layer and the output layer, expressed as:

[0016]

[0017] in: is the updated weight between the hidden layer and the output layer, To evaluate the learning rate of the sub-model, is the hidden layer output value, Train the predicted value for the pumping unit operating frequency and the historical true value of the pumping unit operating frequency The error between is the threshold between the updated hidden layer and the output layer, is the threshold between the hidden layer and the output layer B32. Use the evaluation sub-model to update the weights and thresholds between the input layer and the hidden layer, expressed as:

[0018]

[0019] in: is the updated weight between the input layer and the hidden layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the neuron number in the output layer, is the total number of neurons in the output layer, is the weight between the hidden layer and the output layer, is the updated threshold of the input layer and hidden layer, is the threshold of the input layer and the hidden layer.

[0020] Furthermore, an improved weight training module is added between the hidden layer and the output layer using the evaluation sub-model. The data processing formula of the improved weight training module is expressed as:

[0021] in: is the output value of the hidden layer after the improved weight training module, is the first bias term coefficient, is the Hadamard product symbol, is the hyperbolic tangent function, is the second bias term coefficient.

[0022] An intelligent splicing system for pumping wells using the above method comprises a pumping start-stop calculation unit, a pumping operation frequency analysis unit, a pumping operation frequency prediction unit and a pumping intelligent splicing control unit.

[0023] The pumping unit start-stop calculation unit is used to collect historical pumping unit power consumption data and pumping unit operating condition data, calculate the pumping unit shutdown operating frequency based on the historical pumping unit power consumption data and pumping unit operating condition data, and set the oil well oil pressure for the pumping unit to start; The pumping unit operating frequency analysis unit is used to collect multi-index production data of the oil well, and obtain the influencing factors of the pumping unit operating frequency according to the multi-index production data of the oil well; The pumping unit operating frequency prediction unit is used to construct a pumping unit operating frequency prediction model, construct a data set based on the influencing factors of the pumping unit operating frequency, use the data set to train the pumping unit operating frequency prediction model, and use the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value; The intelligent intermittent switching control unit of the oil pump is used to control the intelligent intermittent switching of the oil pump according to the working frequency of the oil pump being turned off, the oil pressure of the oil well when the oil pump is turned on, and the predicted value of the working frequency of the oil pump.

[0024] The present invention has the following beneficial effects: (1) The present invention collects historical power consumption data and operating condition data of the oil pumping unit, and based on the historical power consumption data and operating condition data of the oil pumping unit, uses the stroke sampling method and the relative deviation method to calculate the operating frequency of the oil pumping unit shutdown, thereby overcoming the problem of inaccurate timing of the oil pumping unit well switch-on and switch-off caused by simply setting a threshold value at the downstroke to shut down the oil pumping unit; (2) The present invention obtains the influencing factors of the operating frequency of the oil pump and constructs an operating frequency prediction model for the oil pump. A data set is constructed based on the influencing factors of the operating frequency of the oil pump. The operating frequency prediction model of the oil pump is trained using the data set. The operating frequency prediction model of the oil pump is trained to obtain the predicted value of the operating frequency of the oil pump. The operating frequency of the oil pump can be adjusted accordingly according to the actual working conditions to avoid the problem that the extraction capacity of the oil pump is greater than the actual fluid supply capacity of the oil well, thereby causing poor pump efficiency, waste of electric energy, and damage to motor equipment. (3) The present invention proposes an intelligent intermittent system for pumping wells using the above method, comprising a pumping start-stop calculation unit, a pumping operation frequency analysis unit, a pumping operation frequency prediction unit and a pumping intelligent intermittent control unit; the pumping start-stop calculation unit can collect historical pumping power consumption data and pumping operation condition data, calculate the pumping shutdown operation frequency based on the historical pumping power consumption data and pumping operation condition data, and set the oil pressure of the oil well when the pumping is turned on; the pumping operation frequency analysis unit can collect multi-index production data of the oil well, and predict the operation frequency of the oil well according to the oil well's The multi-index production data is used to obtain the influencing factors of the operating frequency of the oil pump; the operating frequency prediction unit of the oil pump can construct an operating frequency prediction model of the oil pump, construct a data set based on the influencing factors of the operating frequency of the oil pump, use the data set to train the operating frequency prediction model of the oil pump, and use the trained operating frequency prediction model of the oil pump to obtain the predicted value of the operating frequency of the oil pump; the intelligent switching control unit of the oil pump can control the intelligent switching of the oil pump according to the operating frequency of the oil pump when it is closed, the oil pressure of the oil well when the oil pump is opened, and the predicted value of the operating frequency of the oil pump. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of a process flow of an intelligent interleaving method for a pumping well; Figure 2 The figure is a schematic diagram of the structure of an intelligent splicing system for oil pumping wells. DETAILED DESCRIPTION

[0026] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0027] like Figure 1 As shown, a method for intelligent spacing of pumping wells includes the following steps: S1. Collect historical power consumption data and operating data of the oil pumping units. Based on the historical power consumption data and operating data of the oil pumping units, use the stroke sampling method and the relative deviation method to calculate the operating frequency of the oil pumping unit shutdown, and set the oil pressure of the oil well when the oil pumping unit is turned on.

[0028] In an optional embodiment of the present invention, the present invention is provided with a pumping unit start-stop calculation unit. The pumping unit start-stop calculation unit collects historical pumping unit power consumption data and pumping unit operating data, and based on the historical pumping unit power consumption data and pumping unit operating data, uses the stroke sampling method and the relative deviation method to calculate the pumping unit shutdown operating frequency, and sets the oil pressure of the oil well when the pumping unit is turned on. The present invention uses the stroke sampling method and the relative deviation method to calculate the pumping unit shutdown operating frequency based on the historical pumping unit power consumption data and pumping unit operating data, and includes the following steps: A1. Based on the historical power consumption data and operating condition data of the oil pumping unit, the stroke sampling method is used to determine the standard value of the operating frequency of the oil pumping unit dry pumping cycle, which is expressed as:

[0029] in: is the standard value of the working frequency of the oil pumping unit during dry pumping cycle, is the mean operating frequency of the first stroke sampling point in the dry pumping cycle of the pumping unit, is the mean operating frequency of the second stroke sampling point in the dry pumping cycle of the pumping unit, is the mean operating frequency of the third stroke sampling point in the dry pumping cycle of the pumping unit, The first period of the dry pumping cycle of the pumping unit The average working frequency of the stroke sampling points, is the number of stroke sampling points in the dry pumping cycle of the pumping unit.

[0030] A2. Based on the historical power consumption data and working condition data of the oil pumping unit, and according to the standard value of the working frequency of the oil pumping unit during the dry pumping cycle in step A1, the stroke difference and the difference change rate of the oil pumping unit during the dry pumping cycle are calculated using the relative deviation method, which is expressed as:

[0031]

[0032]

[0033]

[0034] in: is the stroke difference of the dry pumping cycle of the pumping unit, is the stroke difference of the first stroke sampling point in the dry pumping cycle of the pumping unit, is the stroke difference of the second stroke sampling point in the dry pumping cycle of the pumping unit, is the stroke difference of the third stroke sampling point in the dry pumping cycle of the pumping unit, The first period of the dry pumping cycle of the pumping unit The stroke difference of the stroke sampling points, The first period of the dry pumping cycle of the pumping unit The stroke difference of each stroke sampling point, The first period of the dry pumping cycle of the pumping unit The working frequency of each stroke sampling point, The first period of the dry pumping cycle of the pumping unit The standard value of the working frequency of each stroke sampling point, is the difference change rate of the dry pumping cycle of the pumping unit, is the difference change rate of the first stroke sampling point in the dry pumping cycle of the pumping unit, is the difference change rate of the second stroke sampling point in the dry pumping cycle of the pumping unit, is the difference change rate of the third stroke sampling point in the dry pumping cycle of the pumping unit, The first period of the dry pumping cycle of the pumping unit The difference change rate of each stroke sampling point, The first period of the dry pumping cycle of the pumping unit The difference change rate of each stroke sampling point.

[0035] A3. Based on the historical power consumption data and operating condition data of the oil pumping units, the membership matrix of the stroke difference and the membership matrix of the difference change rate are constructed, which are expressed as:

[0036]

[0037] in: is the membership matrix of stroke difference, is the stroke difference membership of the first stroke sampling point in the dry pumping cycle of the pumping unit, is the stroke difference membership of the second stroke sampling point in the dry pumping cycle of the pumping unit, is the stroke difference membership of the third stroke sampling point in the dry pumping cycle of the pumping unit, The first period of the dry pumping cycle of the pumping unit The stroke difference membership of each stroke sampling point, The first period of the dry pumping cycle of the pumping unit The stroke difference membership of each stroke sampling point, is the membership matrix of the difference change rate, is the membership degree of the difference change rate of the first stroke sampling point in the dry pumping cycle of the pumping unit, is the membership degree of the difference change rate of the third stroke sampling point in the dry pumping cycle of the pumping unit, is the membership degree of the difference change rate of the third stroke sampling point in the dry pumping cycle of the pumping unit, The first period of the dry pumping cycle of the pumping unit The membership degree of the difference change rate of each stroke sampling point is: The first period of the dry pumping cycle of the pumping unit The membership of the difference change rate of each stroke sampling point.

[0038] The present invention is based on historical oil pumping unit power consumption data and oil pumping unit operating condition data, and utilizes a membership function method to calculate the stroke difference membership and the difference change rate membership.

[0039] A4. According to the stroke difference and difference change rate of the dry pumping cycle of the pumping unit in step A2 and the membership matrix of the stroke difference and the membership matrix of the difference change rate in step A3, the working frequency of the pumping unit shutdown is calculated, which is expressed as:

[0040] in: is the working frequency of the oil pump shut down, Correction factor for the operating frequency of the pump shut down.

[0041] S2. Collect multi-index production data of the oil well, and obtain influencing factors of the operating frequency of the oil pump based on the multi-index production data of the oil well.

[0042] In an optional embodiment of the present invention, the present invention is provided with a pumping unit operating frequency analysis unit. The pumping unit operating frequency analysis unit collects multi-index production data of the oil well, and uses a regression analysis method to obtain the influencing factors of the pumping unit operating frequency based on the multi-index production data of the oil well. The influencing factors of the pumping unit operating frequency include the page height, casing pressure, pump efficiency and flow rate of the oil well.

[0043] S3, constructing a pumping unit operating frequency prediction model, constructing a data set based on the influencing factors of the pumping unit operating frequency in step S2, using the data set to train the pumping unit operating frequency prediction model, and using the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value.

[0044] In an optional embodiment of the present invention, the present invention is provided with a pumping unit operating frequency prediction unit. The pumping unit operating frequency prediction unit constructs a pumping unit operating frequency prediction model, constructs a data set based on the influencing factors of the pumping unit operating frequency, trains the pumping unit operating frequency prediction model using the data set, and obtains the pumping unit operating frequency prediction value using the trained pumping unit operating frequency prediction model. The pumping unit operating frequency prediction model includes a prediction sub-model and an evaluation sub-model.

[0045] The present invention uses a data set to train a pumping unit operating frequency prediction model, comprising the following steps: B1. Input the historical data of the data set into the prediction sub-model to obtain the training prediction value of the pumping unit operating frequency.

[0046] Step B1 includes the following steps: B11. Input the historical data of the data set into the input layer, and use the input layer and hidden layer to get the output value of the hidden layer, expressed as:

[0047] in: is the hidden layer output value, The neurons in the hidden layer The activation function, is the neuron number in the input layer, is the total number of neurons in the input layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the threshold of the input layer and the hidden layer, is the neuron number in the hidden layer, is the total number of neurons in the hidden layer.

[0048] B12. According to the hidden layer output value in step B11, the hidden layer and the output layer are used to obtain the training prediction value of the pumping unit operating frequency, which is expressed as:

[0049] in: is the predicted value for the pumping unit operating frequency training, is the weight between the hidden layer and the output layer, is the threshold between the hidden layer and the output layer, is the neuron number in the output layer, is the total number of neurons in the output layer.

[0050] B2. Determine whether the error between the training prediction value of the operating frequency of the oil pump in step B1 and the historical true value of the operating frequency of the oil pump is greater than the maximum error; if so, proceed to step B3, otherwise return to step B2.

[0051] B3. Use the evaluation sub-model to update the prediction sub-model until the set number of training iterations is reached.

[0052] Step B3 includes the following steps: B31. Use the evaluation sub-model to update the weights and thresholds between the hidden layer and the output layer, expressed as:

[0053]

[0054] in: is the updated weight between the hidden layer and the output layer, To evaluate the learning rate of the sub-model, is the hidden layer output value, Train the predicted value for the pumping unit operating frequency and the historical true value of the pumping unit operating frequency The error between is the threshold between the updated hidden layer and the output layer, is the threshold between the hidden layer and the output layer.

[0055] The present invention uses the evaluation sub-model to add an improved weight training module between the hidden layer and the output layer. The data processing formula of the improved weight training module is expressed as:

[0056] in: is the output value of the hidden layer after the improved weight training module, is the first bias term coefficient, is the Hadamard product symbol, is the hyperbolic tangent function, is the second bias term coefficient.

[0057] The present invention adds an improved weight training module between the hidden layer and the output layer, which can selectively control the flow of hidden layer output values ​​under different linear transformations, thereby accelerating the convergence speed of the weights between the hidden layer and the output layer and improving the training efficiency of the prediction sub-model.

[0058] B32. Use the evaluation sub-model to update the weights and thresholds between the input layer and the hidden layer, expressed as:

[0059]

[0060] in: is the updated weight between the input layer and the hidden layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the neuron number in the output layer, is the total number of neurons in the output layer, is the weight between the hidden layer and the output layer, is the updated threshold of the input layer and hidden layer, is the threshold of the input layer and the hidden layer.

[0061] S4. Based on the working frequency of the pumping unit when it is turned off in step S1, the oil pressure of the oil well when the pumping unit is turned on, and the predicted value of the working frequency of the pumping unit taken in step S3, the pumping unit is controlled to be intelligently switched on and off.

[0062] In an optional embodiment of the present invention, the present invention is provided with an intelligent intermittent control unit for the oil pumping unit, which controls the oil pumping unit to be intelligently intermittently switched based on the working frequency of the oil pumping unit when it is turned off, the oil pressure of the oil well when the oil pumping unit is turned on, and the predicted value of the working frequency of the oil pumping unit.

[0063] like Figure 2 As shown, an intelligent intermittent system for pumping wells using the above method includes a pumping start-stop calculation unit, a pumping operation frequency analysis unit, a pumping operation frequency prediction unit and a pumping intelligent intermittent control unit.

[0064] In an optional embodiment of the present invention, the pump start and stop calculation unit is used to collect historical pump power consumption data and pump operating condition data, calculate the pump shutdown operating frequency based on the historical pump power consumption data and pump operating condition data, and set the oil well oil pressure when the pump is turned on.

[0065] In an optional embodiment of the present invention, the pumping unit operating frequency analysis unit is used to collect multi-index production data of the oil well, and obtain the influencing factors of the pumping unit operating frequency based on the multi-index production data of the oil well.

[0066] In an optional embodiment of the present invention, the pumping unit operating frequency prediction unit is used to construct a pumping unit operating frequency prediction model, construct a data set based on the influencing factors of the pumping unit operating frequency, use the data set to train the pumping unit operating frequency prediction model, and use the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value.

[0067] In an optional embodiment of the present invention, the intelligent intermittent switching control unit of the oil pump is used to control the intelligent intermittent switching of the oil pump according to the working frequency of the oil pump being shut down, the oil pressure of the oil well when the oil pump is turned on, and the predicted value of the working frequency of the oil pump.

[0068] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0071] The present invention uses specific embodiments 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 those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0072] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for intelligent spacing of pumping wells, characterized in that: The following steps are involved: S1. Collect historical power consumption data and operating condition data of the oil pumping unit, calculate the operating frequency of the oil pumping unit shutdown by using the stroke sampling method and the relative deviation method based on the historical power consumption data and operating condition data of the oil pumping unit, and set the oil pressure of the oil well when the oil pumping unit is turned on; S2. Collecting multi-index production data of the oil well, and obtaining influencing factors of the operating frequency of the pumping unit according to the multi-index production data of the oil well; S3, constructing a pumping unit operating frequency prediction model, constructing a data set based on the influencing factors of the pumping unit operating frequency in step S2, using the data set to train the pumping unit operating frequency prediction model, and using the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value; S4. Based on the working frequency of the pumping unit when it is turned off in step S1, the oil pressure of the oil well when the pumping unit is turned on, and the predicted value of the working frequency of the pumping unit taken in step S3, the pumping unit is controlled to be intelligently switched on and off.

2. The intelligent spacing method for pumping wells according to claim 1 is characterized in that: In step S1, based on historical power consumption data of the oil pumping unit and the operating condition data of the oil pumping unit, the operating frequency of the oil pumping unit shutdown is calculated using the stroke sampling method and the relative deviation method, including the following steps: A1. Based on the historical power consumption data and operating condition data of the oil pumping unit, the standard value of the operating frequency of the oil pumping unit dry pumping cycle is determined by using the stroke sampling method; A2, based on the historical power consumption data and working condition data of the oil pumping unit, and according to the standard value of the working frequency of the oil pumping unit during the dry pumping cycle in step A1, the stroke difference and the difference change rate of the oil pumping unit during the dry pumping cycle are calculated using the relative deviation method; A3. Based on the historical power consumption data and operating condition data of the oil pumping units, a membership matrix of the stroke difference and a membership matrix of the difference change rate are constructed; A4. Calculate the shut-down operating frequency of the oil pumping unit according to the stroke difference and difference change rate of the oil pumping unit dry pumping cycle in step A2 and the membership matrix of the stroke difference and the membership matrix of the difference change rate in step A3.

3. The intelligent spacing method for pumping wells according to claim 2 is characterized in that: In step A4, the operating frequency of the pumping unit shutdown is calculated based on the stroke difference and difference change rate of the pumping unit dry pumping cycle in step A2 and the membership matrix of the stroke difference and the membership matrix of the difference change rate in step A3, which is expressed as: in: is the working frequency of the oil pump shut down, is the correction factor for the pumping unit shutdown frequency, is the stroke difference of the dry pumping cycle of the pumping unit, is the membership matrix of stroke difference, is the difference change rate of the dry pumping cycle of the pumping unit, is the membership matrix of the difference change rate.

4. The intelligent spacing method for pumping wells according to claim 1 is characterized in that: In step S2, the factors affecting the operating frequency of the pumping unit include the page height, casing pressure, pump efficiency and flow rate of the oil well.

5. The intelligent spacing method for pumping wells according to claim 1 is characterized in that: In step S3, the pumping unit operating frequency prediction model includes a prediction sub-model and an evaluation sub-model.

6. The intelligent spacing method for pumping wells according to claim 5 is characterized in that: In step S3, the pumping unit operating frequency prediction model is trained using the data set, including the following steps: B1. Input the historical data of the data set into the prediction sub-model to obtain the training prediction value of the pumping unit operating frequency; B2, judging whether the error between the training prediction value of the operating frequency of the oil pumping unit in step B1 and the historical true value of the operating frequency of the oil pumping unit is greater than the maximum error value; if so, proceeding to step B3, otherwise returning to step B2; B3. Use the evaluation sub-model to update the prediction sub-model until the set number of training iterations is reached.

7. The intelligent spacing method for pumping wells according to claim 6 is characterized in that: Step B1 includes the following steps: B11. Input the historical data of the data set into the input layer, and use the input layer and hidden layer to get the output value of the hidden layer, expressed as: in: is the hidden layer output value, The neurons in the hidden layer The activation function, is the neuron number in the input layer, is the total number of neurons in the input layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the threshold of the input layer and the hidden layer, is the neuron number in the hidden layer, is the total number of neurons in the hidden layer; B12. According to the hidden layer output value in step B11, the hidden layer and the output layer are used to obtain the training prediction value of the pumping unit operating frequency, which is expressed as: in: is the predicted value for the pumping unit operating frequency training, is the weight between the hidden layer and the output layer, is the threshold between the hidden layer and the output layer, is the neuron number in the output layer, is the total number of neurons in the output layer.

8. The intelligent spacing method for pumping wells according to claim 6 is characterized in that: Step B3 includes the following steps: B31. Use the evaluation sub-model to update the weights and thresholds between the hidden layer and the output layer, expressed as: in: is the updated weight between the hidden layer and the output layer, To evaluate the learning rate of the sub-model, is the hidden layer output value, Train the predicted value for the pumping unit operating frequency and the historical true value of the pumping unit operating frequency The error between is the threshold between the updated hidden layer and the output layer, is the threshold between the hidden layer and the output layer B32. Use the evaluation sub-model to update the weights and thresholds between the input layer and the hidden layer, expressed as: in: is the updated weight between the input layer and the hidden layer, is the weight between the input layer and the hidden layer, For the The input value of a neuron, is the neuron number in the output layer, is the total number of neurons in the output layer, is the weight between the hidden layer and the output layer, is the updated threshold of the input layer and hidden layer, is the threshold of the input layer and the hidden layer.

9. The intelligent spacing method for pumping wells according to claim 8 is characterized in that: An improved weight training module is added between the hidden layer and the output layer using the evaluation sub-model. The data processing formula of the improved weight training module is expressed as: in: is the output value of the hidden layer after the improved weight training module, is the first bias term coefficient, is the Hadamard product symbol, is the hyperbolic tangent function, is the second bias term coefficient.

10. An intelligent oil pumping well interleaving system using the method described in any one of claims 1 to 9, characterized in that: Pumping unit start-stop calculation unit, pumping unit operating frequency analysis unit, pumping unit operating frequency prediction unit and pumping unit intelligent switching control unit; The pumping unit start-stop calculation unit is used to collect historical pumping unit power consumption data and pumping unit operating condition data, calculate the pumping unit shutdown operating frequency based on the historical pumping unit power consumption data and pumping unit operating condition data, and set the oil well oil pressure for the pumping unit to start; The pumping unit operating frequency analysis unit is used to collect multi-index production data of the oil well, and obtain the influencing factors of the pumping unit operating frequency according to the multi-index production data of the oil well; The pumping unit operating frequency prediction unit is used to construct a pumping unit operating frequency prediction model, construct a data set based on the influencing factors of the pumping unit operating frequency, use the data set to train the pumping unit operating frequency prediction model, and use the trained pumping unit operating frequency prediction model to obtain the pumping unit operating frequency prediction value; The intelligent intermittent switching control unit of the oil pump is used to control the intelligent intermittent switching of the oil pump according to the working frequency of the oil pump being turned off, the oil pressure of the oil well when the oil pump is turned on, and the predicted value of the working frequency of the oil pump.

Images