Frequency and voltage coordination optimization method for beam-pumping system based on indicator diagram

Abstract

The present application relates to a frequency and voltage coordination optimization method for beam pumping system based on dynamometer card, which converts the traditional dynamometer card into a P-T dynamometer card of load and time, and obtains the mapping relationship between the P-T dynamometer card and the electric diagram by comparing the P-T dynamometer card and the electric diagram in the same time domain and the same period, relying on the mathematical relationship between the equivalent load torque of the polished rod and the equivalent driving torque of the motor. The method converts the electric diagram features into P-T dynamometer card features, takes the mapping feature points of the P-T dynamometer card as input, relies on deep learning to output the optimal P-T dynamometer card load, and finally obtains the corresponding optimal frequency and optimal voltage according to the mapping model processing data, to establish a flexible intelligent coordination optimization model for the frequency and voltage of the beam pumping system based on the dynamometer card. The present application avoids the application of electric diagram acquisition equipment, improves the energy saving effect and the anti-theft performance, and can ensure the real-time flexible intelligent coordination optimization of the frequency and voltage of the beam pumping system.

Description

Technical Field

[0001] This invention relates to the field of oil and gas extraction technology, and more particularly to a flexible intelligent coordination optimization method for frequency and voltage of a beam pumping system based on a dynamometer diagram. Background Technology

[0002] The walking beam pumping unit is an artificial lifting device used in oil extraction. Due to its simple structure, reliable performance, and low price, it occupies a major market share. The walking beam pumping unit mainly consists of a pump, sucker rod, polished rod, four-bar linkage, gearbox, belt, and electric motor. The electric motor drives the crank to rotate through the speed reduction and torque amplification effect of the belt and gearbox. The four-bar linkage converts the rotational motion into the linear motion of the rod column. Based on the direction of rod column movement, the pumping unit's operation is divided into upstroke and downstroke. In the upstroke, the liquid located in the tubing and above the traveling valve is lifted, and when the fixed valve opens, the liquid in the casing is drawn into the pump barrel. In the downstroke, when the traveling valve opens, the fluid in the pump barrel is discharged, and the liquid load above the traveling valve is transferred to the fixed valve, unloading the sucker rod column. This "up-suction, down-discharge" pumping method, combined with the motion characteristics of the crank-rocker mechanism and the combined effect of the counterweight, results in a "double-hump" bidirectional fluctuation in the surface load of the walking beam pumping system. Periodic fluctuating loads not only affect the balance and reliability of mechanical systems, but also increase the required rated power of motors to meet peak torque demands. This causes motors to operate in low-load areas most of the time, resulting in low power utilization and serious energy waste. In particular, the negative torque generated in local periods drives the motor to generate electricity in reverse, reducing energy transmission efficiency and polluting the power grid, causing unstable power supply and energy waste.

[0003] Existing flexible frequency conversion voltage regulation solutions, such as the patent application number 202211716134.4, "A Flexible Intelligent Coordination and Optimization Method for Frequency and Voltage of a Beam Pumping System Based on Power Diagrams," involve real-time acquisition of power diagrams and real-time intelligent coordination and optimization of the frequency and voltage of the beam pumping system. However, in oilfields, power diagram acquisition equipment is typically placed in ground-level control cabinets, making it vulnerable to theft and unable to guarantee real-time power diagram acquisition. Consequently, the effective implementation of the power diagram-based flexible intelligent coordination and optimization method for frequency and voltage of the beam pumping system cannot be guaranteed. Summary of the Invention

[0004] In view of this, the present invention provides a flexible intelligent coordination optimization method for frequency and voltage of a beam pumping system based on dynamometer diagrams. By deeply analyzing the time-varying correspondence between dynamometer diagrams and flexible frequency conversion voltage regulation, and combining deep learning to establish a flexible intelligent coordination optimization model for frequency and voltage based on dynamometer diagrams, the method achieves online real-time intelligent coordinated flexible frequency conversion voltage regulation of oil wells. The dynamometer is installed on the suspension cable and swings up and down with the up and down strokes, which is relatively safe, has a long service life, and is widely used. It avoids the application of dynamometer diagram acquisition equipment, improves energy saving effect and anti-theft performance, and can ensure real-time flexible intelligent coordination optimization of frequency and voltage of the beam pumping system.

[0005] Therefore, the present invention provides the following technical solution:

[0006] This invention provides a flexible intelligent coordination optimization method for frequency and voltage in a beam pumping system based on dynamometer diagrams, comprising:

[0007] Obtain historical dynamometer diagrams and electrical power diagrams for the same cycle, and establish a mapping model between PT dynamometer diagrams and electrical power diagrams by combining the relationship between the equivalent load torque of the polished rod and the drive torque of the motor.

[0008] Dynamometers are used to automatically collect dynamometer diagram data of beam pumping systems in real time.

[0009] Based on the characteristic points mapped by the PT indicator diagram as the characteristic parameters of the oil pumping system, a flexible intelligent coordination optimization model for frequency and voltage based on the indicator diagram is established.

[0010] Using arbitrary fluctuating load and real-time dynamometer diagram mapping characteristics as input, and based on the frequency and voltage flexible intelligent coordination optimization model of the dynamometer diagram, the optimal frequency and optimal voltage of the beam pumping system are predicted and determined in real time.

[0011] The optimal load corresponding to the optimal frequency and optimal voltage of the beam pumping system is obtained by using the mapping model between the PT indicator diagram and the electrical dynamometer diagram.

[0012] Furthermore, a frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram is established, including:

[0013] An electromechanical coupling dynamic model of a flexible variable frequency voltage regulating beam pumping system is established to analyze the system's dynamic characteristics, energy-saving mechanism, and coupling relationship between flexible frequency conversion and flexible voltage regulation. A flexible synchronous coordination optimization model for frequency and voltage is designed.

[0014] The characteristic points mapped by the PT indicator diagram were selected as the characteristic parameters of the pumping system, and the data classification criteria were determined based on the balance state of the beam pumping unit well.

[0015] Using the frequency and voltage corresponding to the load in the dynamometer diagram output by the flexible synchronous coordination optimization model as the source domain dataset and the target domain dataset, a BP neural network is used to intelligently classify the input dynamometer diagram based on the data classification criteria.

[0016] Preliminary regularization processing is performed on the classified data. A neural network capable of processing time series data is used to establish the mapping relationship between the PT dynamometer diagram and the optimal frequency and optimal voltage. The motor input power, the first derivative of the motor input power, and the second derivative of the motor input power are selected as network input parameters to establish a flexible intelligent coordination optimization model for frequency and voltage based on the dynamometer diagram.

[0017] Furthermore, a mapping model between the PT indicator diagram and the electrical power diagram is established, including:

[0018] Collect historical indicator diagrams and electrical power diagrams for the same cycle;

[0019] Convert the dynamometer diagram into a PT dynamometer diagram with load on the vertical axis and time on the horizontal axis;

[0020] By comparing the PT indicator diagram and the electrical power diagram under the same cycle, the mapping relationship between the PT indicator diagram and the electrical power diagram can be obtained;

[0021] Based on the mapping relationship between the PT indicator diagram and the electrical power diagram, the peaks, troughs, and points where the load is 0 on the electrical power diagram are determined to correspond to the mapping feature points on the PT indicator diagram. Based on the mathematical relationship between the bare rod load and the motor load, the correspondence between the load and frequency voltage on the indicator diagram is obtained. These are used as training data pairs to establish a mapping model between the PT indicator diagram and the electrical power diagram.

[0022] Furthermore, the relationship between the equivalent load torque of the polished rod and the motor drive torque is obtained, including:

[0023] Assuming all transmission components in the ground system are rigid and neglecting the backlash between transmission pairs; taking the output shaft of the induction motor as an equivalent component, the mass and moment of inertia of each transmission component in the ground system from the motor output shaft to the polished rod are equivalently represented to the motor output shaft, thus establishing a single-degree-of-freedom mechanical model of the ground transmission system's motion laws:

[0024] Among them, M ed The equivalent driving torque of the ground system is expressed in N·m (m). ef The equivalent load torque of the ground system is expressed in N·m and J. e This is the equivalent moment of inertia at the motor output shaft, representing the mass and moment of inertia of all moving components from the motor output shaft to the polished rod. The unit is kg·m. 2 J i Let be the moment of inertia of the i-th component in the ground transmission system, expressed in kg·m. 2 θm The rotor angle of the electric motor, ω, is measured in rad. m This is the angular velocity of the motor rotor, measured in rad / s.

[0025] Write the moment balance equation at the connection point D between the walking beam support and the walking beam to obtain the axial force Fp of the connecting rod;

[0026] By applying the force equilibrium equations to the connecting rod, we obtain the tangential force F acting on the connecting rod in the direction of crank motion. t ;

[0027] Write the torque balance equations for the crank rotation center, and then calculate Fp and F t Substituting the values, we obtain the equivalent load torque of the crankshaft as follows:

[0028]

[0029] Among them, L 1b L is the distance from the center of mass of the crank counterweight to the center of rotation of the crank, in meters (m). R L1 is the length of the crank, in meters; L2 is the distance from the crank's center of mass to its center of rotation, in meters; G1 is the crank's weight, in nitrogen (N); G 1b θ is the crank counterweight (in N); τ is the crank hysteresis angle (in rad); θ1 is the angle between the crank and the vertical direction (in rad); θ2 is the angle between the crank and the base rod (in rad); θ3 is the angle between the connecting rod and the base rod (in rad); θ4 is the angle between the walking beam and the base rod (in rad); L C F is the length of the rear arm of the walking beam, in meters (m). PRL The load on the bare rod is in N; L A G3 is the length of the walking beam's forearm, in meters (m); G3 is the weight of the walking beam, in kilometres (N); L3 is the distance from the walking beam's center of mass to its rotation center, in meters (m); α is the angle between the base rod and the vertical direction, in rad; L 3b G is the length from the center of mass of the balance weight of the walking beam to the center of rotation of the walking beam, in meters (m). 3b τ is the self-weight of the balance beam, in N; Y η is the lag angle of the balance weight of the walking beam, in rad; C η represents the mechanical transmission efficiency of bearing C. D G2 is the mechanical transmission efficiency of bearing D; G2 is the weight of the connecting rod in N; k2 is the energy flow coefficient; η B is the mechanical transmission efficiency of bearing B; k3 is the energy flow coefficient.

[0030] Furthermore, an electromechanical coupling dynamic model of the flexible variable frequency pressure regulating beam pumping system is established, including:

[0031] Based on the transient equivalent circuit of the induction motor, and considering the influence of time-varying frequency, time-varying voltage and time-varying electromagnetic parameters on the mechanical characteristics of the motor, a dynamic simulation model of a flexible variable frequency voltage regulating induction motor is established.

[0032] Considering the influence of time-varying friction of each transmission pair on the torsional vibration of each component of the ground transmission system, a multibody dynamics simulation model of the ground system is established.

[0033] Taking into account the mutual coupling effects of motor rotation, surface system transmission, downhole rod string longitudinal vibration, pump reciprocating motion, and oil flow during flexible frequency conversion and pressure regulation drive, an electromechanical coupling dynamic simulation model of the flexible frequency conversion and pressure regulation beam pumping system is established.

[0034] Furthermore, the BP neural network includes: an input layer, a hidden layer, and an output layer;

[0035] A backpropagation (BP) neural network is used to intelligently classify the input dynamometer diagram, including:

[0036] The characteristic parameters of the indicator diagram are input from the input layer. After the signal is processed by each hidden layer, it is finally output from the output layer in three categories: overbalanced, balanced, and underbalanced.

[0037] Furthermore, the training of the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram includes:

[0038] The optimal flexible frequency and optimal flexible voltage are combined with the dynamometer diagram to form a training data pair;

[0039] Based on the training data, the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram is optimized repeatedly to obtain the optimal neuron weight coefficients.

[0040] The optimizations include:

[0041] Calculate the error between the optimized flexible frequency and optimized flexible voltage output by the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram and the optimal flexible frequency and optimal voltage;

[0042] The error signal is backpropagated from the last layer to the next layer to obtain the error learning signal for each layer, and then the weights of the neurons in each layer are corrected based on the error learning signal.

[0043] Furthermore, it also includes: simulating test frequency and voltage, and correcting the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram based on the prediction effect.

[0044] The present invention also provides a computer-readable storage medium comprising a stored program, wherein, when the program is executed, the described method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram is performed.

[0045] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the aforementioned method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram through the computer program.

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] This invention transforms traditional indicator diagrams into load-time PT indicator diagrams. By relying on the mathematical relationship between the equivalent load torque of the smooth rod and the equivalent drive torque of the motor, the PT indicator diagram and the electrical dynamometer diagram are compared in the same time domain and the same period to obtain the mapping relationship between the two. This method converts electrical dynamometer diagram features into PT indicator diagram features, uses the mapped feature points of the PT indicator diagram as input, and leverages deep learning to output the optimal PT indicator diagram load. Finally, it processes the data according to the mapping model to obtain the corresponding optimal frequency and voltage, establishing a flexible intelligent coordination optimization model for frequency and voltage in a beam pumping system based on the indicator diagram. This achieves "high load, low frequency, high voltage; low load, high frequency, low voltage" in the pumping system, "peak shaving and valley filling" of load torque, reducing load fluctuations, and improving energy efficiency and anti-theft performance. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a roadmap for the research on flexible intelligent coordination and optimization of frequency and voltage based on dynamometer diagrams in this invention.

[0050] Figure 2 This is a mechanical model of the motion law of the ground system in this embodiment of the invention;

[0051] Figure 3 This is a schematic diagram of the motion of the four-bar linkage in an embodiment of the present invention;

[0052] Figure 4 This is a schematic diagram of the force distribution on the four-bar linkage in an embodiment of the present invention. Detailed Implementation

[0053] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0054] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0055] In oilfields, dynamometer data acquisition equipment is typically housed in ground-level control cabinets, making it vulnerable to theft. In contrast, dynamometers, mounted on suspension devices, swing up and down with the stroke, offering greater safety, longer lifespan, and wider application. If the optimal coordination between frequency and voltage can be determined in real-time based on changes in the dynamometer's shape, the time-consuming technical bottleneck of relying on dynamic models for optimization can be overcome, and anti-theft performance can be improved.

[0056] The bare rod (ground) dynamometer diagram is a graphical representation of the load at the suspension point of the pumping unit as its displacement changes. It is a closed curve measured by the dynamometer within one pumping cycle of the pumping unit. This invention, relying on deep learning, addresses the periodic fluctuations in the dynamometer diagram of the pumping system by using time-varying frequency and voltage within the cycle as optimization variables. This allows the pumping system to predict and determine the optimal frequency and voltage in real time, achieving the goal of reducing load fluctuations and energy consumption in the pumping system in real time. It solves the problem that existing technologies, which rely on dynamic models and optimization algorithms, are time-consuming and thus reduce energy-saving effects.

[0057] like Figure 1 As shown, this invention provides a flexible intelligent coordination optimization method for frequency and voltage in a beam pumping system based on a dynamometer diagram, comprising the following steps:

[0058] S1. Based on fully considering the actual motion law of the suspension point and combining the motion law of the ground transmission system of the beam pumping unit, the relationship between the equivalent load torque of the polished rod and the motor drive torque is obtained, that is, the mathematical relationship between the polished rod load and the motor load.

[0059] The relationship between the equivalent load torque of the guide rod and the drive torque of the motor, i.e., the mathematical relationship between the guide rod load and the motor load, specifically includes:

[0060] S11. Assuming all transmission components of the ground system are rigid and neglecting the backlash of each transmission pair, and taking the output shaft of the induction motor as an equivalent component, the mass and moment of inertia of each transmission component of the ground system from the motor output shaft to the polished rod are equivalently represented to the motor output shaft, establishing a single-degree-of-freedom mechanical model of the ground transmission system's motion law (e.g., Figure 2 As shown):

[0061]

[0062]

[0063] Among them, M ed M is the equivalent driving torque of the ground system (N·m). ef J is the equivalent load torque of the ground system (N·m). e The equivalent moment of inertia (kg·m) at the motor output shaft is the conversion of the mass and moment of inertia of all moving components from the motor output shaft to the polished rod. 2 ), J i Let i be the rotational inertia of the i-th component of the ground transmission system (kg·m) 2 ), θ m ω is the rotor angle of the electric motor (rad). m ω is the angular velocity of the motor rotor (rad / s). i Let m be the angular velocity (rad / s) of the i-th component of the ground transmission system. i Let v be the mass (kg) of the i-th component of the ground transmission system. i Let be the velocity of the center of mass of the i-th component in the ground transmission system (m / s); and let M be the equivalent driving torque of the ground system. ed It is determined by the electromagnetic torque of the induction motor.

[0064] S12. The equivalent load torque of the ground system is the load from the polished rod, converted through a four-bar linkage and transmitted to the motor output shaft via a belt reducer. Based on the assumption that all transmission components of the ground system are rigid, the equivalent load torque of the ground system can be expressed by the equivalent load torque of the crankshaft, calculated using the following formula:

[0065]

[0066] Among them, M ec i is the equivalent load torque of the crankshaft (N·m); mg η is the transmission ratio from the motor output shaft to the crankshaft. mgk1 is the transmission efficiency from the motor output shaft to the crankshaft; k1 is the energy flow coefficient, which is the value when the equivalent load torque M of the crankshaft is... ec When >0, k1 = -1; when the crankshaft equivalent load torque M ec When k ≤ 0, k1 = 1;

[0067] Based on the working principle of the four-bar linkage of a beam pumping unit, draw a schematic diagram of the four-bar linkage's motion, such as... Figure 3 As shown, the bottom dead center of the polished rod is taken as the starting position of the pumping system.

[0068] Considering the external loads and the weight of each component of the four-bar linkage, draw a free-body diagram of the four-bar linkage, such as... Figure 4 As shown.

[0069] By applying the moment balance equations at the connection point D between the walking beam support and the walking beam, the axial force of the connecting rod is obtained as follows:

[0070]

[0071]

[0072] Among them, F PRL For the rod load (N), F P L is the axial force (N) on the connecting rod. A L is the length (m) of the walking beam's forearm. C L is the length of the rear arm of the walking beam (m), L3 is the distance from the center of mass of the walking beam to the center of rotation of the walking beam (m), L 3b G is the length (m) from the center of mass of the balance weight of the walking beam to the center of rotation of the walking beam, G3 is the weight of the walking beam (N), G 3b Let θ be the self-weight of the walking beam (N), β be the angle between the connecting rod and the walking beam (rad), θ3 be the angle between the connecting rod and the base rod (rad), θ4 be the angle between the walking beam and the base rod (rad), α be the angle between the base rod and the vertical direction (rad), and τ be the angle between the base rod and the vertical direction. Y η is the lag angle (rad) of the balance weight of the walking beam. C For the mechanical transmission efficiency of bearing C, η D Let k2 be the mechanical transmission efficiency of bearing D, and k2 be the energy flow coefficient. When the speed of the polished rod is v... a When k > 0, k2 = -1, and when the velocity of the rod v a When k ≤ 0, k2 = 1.

[0073] S13. Applying the force balance equations to the connecting rod, we obtain the following tangential force acting on the connecting rod in the direction of crank motion:

[0074]

[0075] Among them, F tG1 is the tangential force (N) exerted by the connecting rod on the crank's direction of motion; G2 is the connecting rod's weight (N); θ1 is the angle between the crank and the vertical direction (rad); θ2 is the angle between the crank and the base rod (rad); ηB is the mechanical transmission efficiency of bearing B; k3 is the energy flow coefficient, when... When, k2 = -1; when At that time, k2 = 1.

[0076] S14. Write the torque balance equation for the crank rotation center, and then use equation F P Sum of F t Substituting the values, we obtain the equivalent load torque of the crankshaft as follows:

[0077]

[0078] Among them, L 1b L is the distance (m) from the center of mass of the crank counterweight to the center of rotation of the crank. R L1 is the length of the crank (m), L1 is the distance from the crank's center of mass to its rotation center (m), G1 is the crank's weight (N), and G... 1b Let τ be the crank counterweight (N), and τ be the crank counterweight lag angle (rad).

[0079] S2. Collect a large number of historical dynamometer diagrams and electrical power diagrams under the same cycle, and transform the traditional dynamometer diagram into a PT dynamometer diagram with load as the vertical axis and time as the horizontal axis. Compare the PT dynamometer diagram and electrical power diagram under the same cycle to obtain the mapping relationship between the PT dynamometer diagram and electrical power diagram. Derive the mapping feature points on the PT dynamometer diagram corresponding to the peak (maximum value in the neighborhood), trough (minimum value in the neighborhood) of the load in the electrical power diagram and the point where the load is 0. Based on the mathematical relationship between the bare rod load and the motor load, obtain the correspondence between the load in the dynamometer diagram and the frequency voltage. Use these as training data pairs to establish a mapping model between the PT dynamometer diagram and electrical power diagram.

[0080] S3. By establishing an electromechanical coupling dynamic model of a flexible variable frequency voltage regulating beam pumping system, the dynamic characteristics of the system, the energy-saving mechanism, and the coupling relationship between flexible frequency conversion and flexible voltage regulation are analyzed to guide the design of a flexible synchronous coordination optimization model for frequency and voltage.

[0081] The establishment of an electromechanical coupling dynamic model for a flexible variable frequency pressure regulating beam pumping system specifically includes:

[0082] S31. Based on the transient equivalent circuit of the induction motor, considering the influence of time-varying frequency, time-varying voltage and time-varying electromagnetic parameters on the mechanical characteristics of the motor, a dynamic simulation model of a flexible variable frequency voltage regulation induction motor is established.

[0083] S32. Considering the influence of time-varying friction of each transmission pair on the torsional vibration of each component of the ground transmission system, establish a multibody dynamics simulation model of the ground system.

[0084] S33. Taking into account the mutual coupling effects of motor rotation, surface system transmission, downhole rod string longitudinal vibration, oil pump reciprocating motion and oil flow during flexible frequency conversion pressure regulation drive, an electromechanical coupling dynamic simulation model of flexible frequency conversion pressure regulation beam pumping system is established.

[0085] S4. Real-time automatic acquisition and storage of dynamometer card data from the beam pumping system. The dynamometer card data is then normalized.

[0086] S5. Select the PT indicator diagram mapping feature points as the characteristic parameters of the pumping system, and determine the data classification standard based on the balance state (underbalanced, balanced, overbalanced) of the beam pumping well, i.e. the ratio of the time taken for the upper and lower strokes.

[0087] Among them, the balance state of a beam pumping unit well includes underbalanced state, balanced state and overbalanced state.

[0088] The data classification criteria are: to test the time t corresponding to the upstroke and downstroke of the pumping unit respectively. 上 and t 下 If t 上 >t 下 If t 上 <t 下 If t 上 =t 下 Then there is a balance.

[0089] S6. Using the frequency and voltage corresponding to the load in the dynamometer diagram output by the flexible synchronous coordination optimization model as the source domain dataset and the target domain dataset, a BP neural network is used to intelligently classify the input samples based on the data classification criteria.

[0090] A backpropagation (BP) neural network consists of an input layer, hidden layers, and an output layer. The BP neural network intelligently classifies input samples, specifically including:

[0091] The characteristic parameters of the dynamometer diagram are input from the input layer. After the signal is processed through each hidden layer, it is finally output from the output layer in three categories: overbalanced, balanced, and underbalanced. Classification facilitates data processing, and what is needed is balanced data. The unbalanced data is classified, their characteristics are identified, and based on these characteristics, the unbalanced data is optimized to be balanced.

[0092] S7. Perform preliminary regularization on the classified data, establish the mapping relationship between the PT dynamometer diagram and the optimal frequency and optimal voltage using a neural network that can effectively process time series data, select the motor input power, the first derivative of the motor input power and the second derivative of the motor input power as network input parameters, and select and optimize the data assimilation algorithm of the neuron weights according to the network efficiency, and establish a frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram.

[0093] Regularization helps improve generalization performance and prevents overfitting of classification data.

[0094] Among them, neural networks are those that can effectively process time series data, such as Long Short-Term Memory Networks (LSTM), Recurrent Neural Networks (RNN), etc.

[0095] The model training process is as follows: Using the PT indicator diagram as the source domain dataset and the optimal flexible frequency and optimal flexible voltage as the target domain dataset, the optimal flexible frequency and optimal flexible voltage output from the dynamic simulation model of the flexible variable frequency and voltage regulating induction motor are combined with the PT indicator diagram under constant frequency and constant voltage conditions obtained from the electromechanical coupling dynamic simulation model of the system to form a training data pair. Based on this training data pair, the model is repeatedly optimized to obtain the optimal neuron weight coefficients. More specifically, for the error between the actual output (optimized flexible frequency and optimized flexible voltage) and the expected output (optimized flexible frequency and optimal flexible voltage) of the model, the error signal is backpropagated layer by layer from the last layer to obtain the error learning signal of each layer. Then, the weights of the neurons in each layer are corrected based on the error learning signal. The above process is repeated continuously, with the weights constantly adjusted during this process. This process continues until the network output error is reduced to below a pre-set threshold.

[0096] The data assimilation algorithm for optimizing neuron weights in network efficiency screening is as follows:

[0097] The neural network is trained on multiple datasets to select and optimize neuron weight coefficients. Test data can also update neuron weight coefficients, meaning that neuron weight coefficients are optimized in each working process.

[0098] S8. When the model is input with any fluctuating load and dynamometer diagram mapping characteristics, the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram can predict and determine the optimal frequency and optimal voltage of the beam pumping system in real time according to its characteristics, obtain the corresponding optimal load of the dynamometer diagram, realize the pumping system of "high load, low frequency and high pressure, low load, high frequency and low pressure", and "peak shaving and valley filling" of load torque.

[0099] S9. Conduct simulation tests on frequency and voltage, predict the results, and further revise the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram.

[0100] The method described in the above embodiments transforms the traditional indicator diagram into a load-time PT indicator diagram. By relying on the mathematical relationship between the equivalent load torque of the polished rod and the equivalent drive torque of the motor, the PT indicator diagram and the electrical dynamometer diagram are compared in the same time domain and the same period to obtain the mapping relationship between the PT indicator diagram and the electrical dynamometer diagram. This method converts the electrical dynamometer diagram features into PT indicator diagram features, uses the mapping feature points of the PT indicator diagram as input, and outputs the optimal PT indicator diagram load based on deep learning. Finally, it processes the data according to the mapping model to obtain the corresponding optimal frequency and optimal voltage, establishing a flexible intelligent coordination optimization model for frequency and voltage of the beam pumping system based on the indicator diagram. When any fluctuating load characteristics are input, this optimization system can predict and determine the optimal load of the indicator diagram in real time based on its characteristics, thereby determining the optimal frequency and optimal voltage of the system. This achieves "high load, low frequency, high voltage; low load, high frequency, low voltage" for the pumping system, "peak shaving and valley filling" of the load torque, and realizing the goal of reducing load fluctuations and saving energy in the pumping system in real time.

[0101] In another embodiment, the present invention also provides a computer-readable storage medium comprising a stored program, wherein, when the program is executed, the above-described method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram is performed.

[0102] In another embodiment, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for flexible intelligent coordination and optimization of frequency and voltage of a beam pumping system based on a dynamometer diagram through the computer program.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A flexible intelligent coordination optimization method for frequency and voltage in a beam pumping system based on dynamometer diagrams, characterized in that, include: Obtain historical dynamometer diagrams and electrical power diagrams for the same cycle, and establish a mapping model between PT dynamometer diagrams and electrical power diagrams by combining the relationship between the equivalent load torque of the polished rod and the drive torque of the motor. Dynamometers are used to automatically collect dynamometer diagram data of beam pumping systems in real time. Based on the characteristic points mapped by the PT indicator diagram as the characteristic parameters of the oil pumping system, a flexible intelligent coordination optimization model for frequency and voltage based on the indicator diagram is established. Using arbitrary fluctuating load and real-time dynamometer diagram mapping characteristics as input, and based on the frequency and voltage flexible intelligent coordination optimization model of the dynamometer diagram, the optimal frequency and optimal voltage of the beam pumping system are predicted and determined in real time. The optimal load corresponding to the optimal frequency and optimal voltage of the beam pumping system is obtained by using the mapping model between the PT indicator diagram and the electrical power diagram. Establish a frequency and voltage flexible intelligent coordination optimization model based on dynamometer diagrams, including: An electromechanical coupling dynamic model of a flexible variable frequency voltage regulating beam pumping system is established to analyze the system's dynamic characteristics, energy-saving mechanism, and coupling relationship between flexible frequency conversion and flexible voltage regulation. A flexible synchronous coordination optimization model for frequency and voltage is designed. The characteristic points mapped by the PT indicator diagram were selected as the characteristic parameters of the pumping system, and the data classification criteria were determined based on the balance state of the beam pumping unit well. Using the frequency and voltage corresponding to the load in the dynamometer diagram output by the flexible synchronous coordination optimization model as the source domain dataset and the target domain dataset, a BP neural network is used to intelligently classify the input dynamometer diagram based on the data classification criteria. Preliminary regularization processing is performed on the classified data. A neural network capable of processing time series data is used to establish the mapping relationship between the PT dynamometer diagram and the optimal frequency and optimal voltage. The motor input power, the first derivative of the motor input power, and the second derivative of the motor input power are selected as network input parameters to establish a flexible intelligent coordination optimization model for frequency and voltage based on the dynamometer diagram. Establish a mapping model between the PT indicator diagram and the electrical power diagram, including: Collect historical indicator diagrams and electrical power diagrams for the same cycle; Convert the dynamometer diagram into a PT dynamometer diagram with load on the vertical axis and time on the horizontal axis; By comparing the PT indicator diagram and the electrical power diagram under the same cycle, the mapping relationship between the PT indicator diagram and the electrical power diagram can be obtained; Based on the mapping relationship between the PT indicator diagram and the electrical power diagram, the peaks, troughs, and points where the load is 0 on the electrical power diagram are determined to correspond to the mapping feature points on the PT indicator diagram. Based on the mathematical relationship between the bare rod load and the motor load, the correspondence between the load and frequency voltage on the indicator diagram is obtained. These are used as training data pairs to establish a mapping model between the PT indicator diagram and the electrical power diagram.

2. The method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram as described in claim 1, characterized in that, To obtain the relationship between the equivalent load torque of the polished rod and the drive torque of the motor, including: Assuming all transmission components in the ground system are rigid and neglecting the backlash between transmission pairs; taking the output shaft of the induction motor as an equivalent component, the mass and moment of inertia of each transmission component in the ground system from the motor output shaft to the polished rod are equivalently represented to the motor output shaft, thus establishing a single-degree-of-freedom mechanical model of the ground transmission system's motion laws: ; in, This represents the equivalent driving torque of the ground system, expressed in N·m. The equivalent load torque of the ground system is expressed in N·m. This is the equivalent moment of inertia at the motor output shaft, representing the mass and moment of inertia of all moving components from the motor output shaft to the polished rod. The unit is kg·m. 2 , This refers to the rotor angle of the electric motor, measured in rad. By applying the moment balance equations at the connection point D between the walking beam support and the walking beam, the axial force of the connecting rod can be obtained. ; By applying the force equilibrium equations to the connecting rod, we obtain the tangential force acting on the connecting rod in the direction of crank motion. ; Write the torque balance equations for the crank rotation center, and then... and Substituting the values, we obtain the equivalent load torque of the crankshaft as follows: ； in, This is the distance from the center of mass of the crank counterweight to the center of rotation of the crank, in meters (m). This is the length of the crank, in meters (m). This is the distance from the crank's center of mass to the crank's center of rotation, in meters (m). The weight of the crankshaft is in N (newtons). The crank counterweight is its own weight, in N; The crankshaft hysteresis angle is measured in rad. The angle between the crank and the vertical direction, measured in rad; The angle between the crank and the base rod, measured in rad. The angle between the connecting rod and the base rod is expressed in rad. The angle between the walking beam and the base member is expressed in rad. This is the length of the rear arm of the walking beam, in meters (m). The load on the bare rod is in N; This is the length of the walkway's forearm, in meters (m). The weight of the moving beam is in N; This is the distance from the centroid of the walking beam to the center of rotation of the walking beam, in meters (m). α The angle between the base rod and the vertical direction is expressed in rad. This is the length from the center of mass of the balance weight of the walking beam to the center of rotation of the walking beam, in meters (m). The self-weight of the balance beam is expressed in N; The lag angle of the balance weight of the walking beam is expressed in rad. Let C be the mechanical transmission efficiency of bearing C; Let D be the mechanical transmission efficiency of bearing D; This is the weight of the connecting rod, expressed in N. Energy flow direction coefficient; The mechanical transmission efficiency of bearing B; k3 This is the energy flow direction coefficient.

3. The method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram as described in claim 1, characterized in that, Establish an electromechanical coupling dynamic model for a flexible variable frequency pressure regulating beam pumping system, including: Based on the transient equivalent circuit of the induction motor, and considering the influence of time-varying frequency, time-varying voltage and time-varying electromagnetic parameters on the mechanical characteristics of the motor, a dynamic simulation model of a flexible variable frequency voltage regulating induction motor is established. Considering the influence of time-varying friction of each transmission pair on the torsional vibration of each component of the ground transmission system, a multibody dynamics simulation model of the ground system is established. Taking into account the mutual coupling effects of motor rotation, surface system transmission, downhole rod string longitudinal vibration, pump reciprocating motion, and oil flow during flexible frequency conversion and pressure regulation drive, an electromechanical coupling dynamic simulation model of the flexible frequency conversion and pressure regulation beam pumping system is established.

4. The method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on a dynamometer diagram as described in claim 1, characterized in that, The BP neural network includes: an input layer, a hidden layer, and an output layer; A backpropagation (BP) neural network is used to intelligently classify the input dynamometer diagram, including: The characteristic parameters of the indicator diagram are input from the input layer. After the signal is processed by each hidden layer, it is finally output from the output layer in three categories: overbalanced, balanced, and underbalanced.

5. The method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on dynamometer diagrams according to claim 1, characterized in that, The training of the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram includes: The optimal flexible frequency and optimal flexible voltage are combined with the dynamometer diagram to form a training data pair; Based on the training data, the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram is optimized repeatedly to obtain the optimal neuron weight coefficients. The optimizations include: Calculate the error between the optimized flexible frequency and optimized flexible voltage output by the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram and the optimal flexible frequency and optimal voltage; The error signal is backpropagated from the last layer to the next layer to obtain the error learning signal for each layer, and then the weights of the neurons in each layer are corrected based on the error learning signal.

6. The method for flexible intelligent coordination and optimization of frequency and voltage in a beam pumping system based on dynamometer diagrams according to claim 1, characterized in that, Also includes: Simulation tests were conducted on frequency and voltage, and the frequency and voltage flexible intelligent coordination optimization model based on the dynamometer diagram was corrected based on the predicted results.

7. A computer-readable storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, it performs the frequency and voltage flexible intelligent coordination optimization method for a beam pumping system based on a dynamometer diagram as described in any one of claims 1 to 6.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the frequency and voltage flexible intelligent coordination optimization method for a beam pumping system based on a dynamometer diagram, as described in any one of claims 1 to 6, through the computer program.

Images