Hybrid drive system for a pumping unit and control method thereof

Abstract

The application belongs to the technical field of pumping unit driving system control, and provides a hybrid driving system for a pumping unit and a control method thereof. The system comprises a wind energy-hydraulic conversion subsystem, a hydraulic secondary balance power subsystem and a sensor detection and control subsystem. The application fuses unstable wind energy with grid electric energy and recoverable hydraulic energy / kinetic energy in the system through hydraulic transmission, adopts an energy management strategy based on model predictive control (MPC), realizes dynamic matching and collaborative work between wind energy, grid electric energy and recoverable energy and the up-and-down stroke demand of the pumping unit, effectively solves the problem that the prior art fails to deeply couple unstable wind energy resources into an industrial mechanical driving system, and thus improves the depth and efficiency of comprehensive utilization of energy of the pumping unit driving system.

Description

Technical Field

[0001] This invention relates to the field of control technology for pumping unit drive systems, and particularly to a hybrid drive system for pumping units and its control method. Background Technology

[0002] Since the beginning of the 21st century, the drawbacks of existing drive technologies used in industrial production have become increasingly apparent, falling far short of the current goals of refined and energy-efficient industrial production. Currently, in mechanical systems with cyclical load characteristics, such as oilfield pumping units, direct motor drive is often used to meet peak power demands, leading to long-term inefficient operation – akin to "overpowering a small vehicle." Especially during load braking or reversal, the motor enters a "reverse generator" state, with the generated energy typically consumed through braking resistors, resulting in energy waste. Existing technologies attempt to improve this by introducing frequency converters or flywheel energy storage, but these systems are complex and costly. Furthermore, existing technologies rarely deeply couple unstable clean energy sources (wind power) with the aforementioned industrial drive scenarios, making it difficult to achieve efficient, stable access and comprehensive utilization of green energy.

[0003] To address the aforementioned issues, existing technologies have proposed some improvements, but solutions for energy recovery and clean energy utilization still have shortcomings: In the area of ​​wind energy hydraulic utilization, patents such as CN202110038571.4 (Compact Electro-hydraulic Wind Turbine Generator Assembly), CN201010106574.9 (A Hydraulic Controlled Frequency Stabilized Wind Turbine Generator), and CN201810649273.7 (A Multi-stage Hydraulic Wind Turbine Generator) primarily utilize the hydraulic energy converted from wind energy to drive a generator, with the system output being electrical energy. These solutions fail to address how to directly and efficiently couple the unstable hydraulic power of wind energy, in the form of hydraulic or mechanical energy, into an industrial mechanical system that is primarily powered by mechanical rotation.

[0004] In the area of ​​hydraulic energy recovery, as described in documents such as "A Composite Energy Recovery System for Pure Electric Excavator Boom and Stick," the braking energy or gravitational potential energy within the system is effectively recovered and reused through hydraulic accumulators and secondary components (pumps / motors). However, the energy source of such solutions depends entirely on the operation of the main unit itself, which is an "internal circulation" energy saving method, lacking a "source source" approach to actively reduce the energy consumption of the main power grid by introducing external clean energy from the source.

[0005] In summary, existing technologies only focus on wind power hydraulic generation without deeply coupling it with mechanical power systems, or are limited to internal energy recovery without a means of generating new energy sources. Currently, there is no comprehensive technical solution that can deeply couple unstable wind energy resources with a hydraulic balance power system with secondary regulation functions through hydraulic transmission and conversion, thereby simultaneously achieving both "generating new energy sources" (wind energy utilization) and "saving energy" (energy recovery). Summary of the Invention

[0006] Therefore, one object of the present invention is to provide a hybrid drive system and control method for an oil pumping unit to solve the problems mentioned in the background art and overcome the shortcomings of the prior art.

[0007] To achieve the above objectives, in one aspect, the present invention provides a hybrid drive system for an oil pumping unit, comprising: The wind power-hydraulic conversion subsystem includes a wind turbine and a first variable displacement hydraulic pump connected to a hydraulic oil tank, the first variable displacement hydraulic pump being connected to the wind turbine; The hydraulic secondary balancing power subsystem includes secondary components and accumulators connected to the hydraulic oil tank; The sensor detection and control subsystem includes a controller, and a pressure transmitter and a torque-speed sensor that are communicatively connected to the controller; the controller is communicatively connected to the wind turbine and the first variable displacement hydraulic pump.

[0008] Preferably, the wind power-hydraulic conversion subsystem further includes a one-way valve and a first relief valve, wherein the first relief valve is connected in parallel with the first variable displacement hydraulic pump, and the one-way valve is connected in series with the first relief valve and the first variable displacement hydraulic pump.

[0009] Preferably, the hydraulic secondary balancing power subsystem includes a servo valve, a second relief valve, a third relief valve, a fourth relief valve connected to the oil tank, and a two-position four-way solenoid directional valve. The servo valve is connected to a double-acting, double-rod hydraulic cylinder, which is connected to the secondary component. The servo valve is communicatively connected to the controller. The second relief valve is connected in parallel to a second variable displacement hydraulic pump, which is connected to a first drive motor. A shuttle valve is connected in series with the second relief valve and the second variable displacement hydraulic pump. The third relief valve is connected in parallel with the secondary component. The secondary component, the third relief valve, and the fourth relief valve are all connected to the two-position four-way solenoid directional valve. The two-position four-way solenoid directional valve is communicatively connected to the controller. A filter is provided between the secondary component and the hydraulic oil tank. The secondary element is connected to the load via a transmission assembly. The transmission assembly includes a driving wheel and a driven wheel, which are connected by a transmission belt. The driving wheel is connected to a second drive motor, and an overrunning clutch is provided between the driven wheel and the load.

[0010] Preferably, the pressure transmitter is located in the hydraulic circuit between the first variable displacement hydraulic pump and the shuttle valve, and the torque speed sensor includes a first torque speed sensor and a second torque speed sensor. The first torque speed sensor is located between the driven wheel and the secondary element, and the second torque speed sensor is located between the second drive motor and the driving wheel.

[0011] On the other hand, the present invention provides a hybrid drive control method for an oil pumping unit, the method employing an energy management strategy based on model predictive control, including: Real-time acquisition of wind speed and wind turbine rotation speed signals, accumulator pressure signals, secondary component rotation speed and torque signals, and second drive motor torque and rotation speed signals; Based on the wind turbine's rotational speed signal, the target displacement is determined by maximum power point tracking. Based on the accumulator's real-time pressure signal, the load pressure state (SOP) characterizing the accumulator's energy storage state is calculated. Based on historical data, the wind speed change trend and load demand change within a certain time window are predicted using a long short-term memory neural network. A state-space prediction model is constructed based on a pre-built subsystem model. Construct a multi-objective optimization function with the objectives of stabilizing the output torque of the drive motor, maintaining the load state SOP within the target range, and minimizing the deviation between the actual wind power and the optimal value. The subsystem model includes: a wind-liquid power supply end model, a simplified steady-state model of the energy storage device, a secondary component model, and a drive motor model; Within each control cycle, based on the currently acquired system state and the predicted wind speed and load demand change sequences, the optimization function with constraints is solved in a rolling manner to obtain the optimal control sequence for each actuator control quantity in the future control time domain. The first control command in the optimal control sequence is issued to the corresponding actuator, and each actuator acts to coordinate wind energy, grid power and recovered energy to jointly drive the load and recover energy.

[0012] As a preferred embodiment, the wind-liquid power supply end model is as follows: ; ; ; ; in, air density, Where is the radius of the wind turbine. The wind turbine rotation speed, Wind speed The wind energy utilization coefficient is the tip speed ratio. The function, arrive It is the wind turbine design constant. The pitch angle of the wind turbine blade; The simplified steady-state model of the accumulator is as follows: ; in, This is the minimum operating pressure of the accumulator. For pressure The volume of gas below, For the highest work pressure and have ; The secondary component model is as follows: ; ; in, To transmit torque, The displacement when a secondary component is used as a motor or pump. For the pressure difference between inlet and outlet, Indicates the flow rate of the secondary component. and A positive superscript indicates the mechanical efficiency of the motor, while a negative superscript indicates the mechanical efficiency of the pump. Rotational speed; The drive motor model is as follows: ; ; ; in, The output torque of the drive motor, To drive the motor speed, The motor torque constant is For the motor armature resistance, Let be the back electromotive force constant of the motor. This is the voltage at the motor terminals.

[0013] Preferably, the formula for calculating the stress state (SOP) is as follows: ; in, For the real-time pressure of the accumulator, and These are the upper limit of the permissible safe operating pressure of the accumulator and the pre-charge pressure, respectively.

[0014] Preferably, the multi-objective optimization function is as follows: in, Let k be the future control sequence to be solved, k be the current time, and i be the prediction step size index. For time-domain prediction, the first term aims to stabilize the motor output torque. Track its normal operating values The second objective is to maintain the accumulator's state of charge (SOP) within the ideal range. Nearby, the third term aims to minimize the deviation between the actual wind power and the optimal value, so that the actual wind power... Approaching the optimal value , For multi-objective optimization functions, This is the weighting coefficient for the motor torque tracking error. This is the weighting coefficient for the SOP tracking error under load conditions. This is the weighting coefficient for wind power tracking error.

[0015] Preferably, the constraints are as follows: , ; ; ; ; in, This refers to the motor's output power. For the power of secondary components, Power consumed by the load The motor torque constant is This represents the motor current.

[0016] Preferably, the actuator includes a first variable displacement hydraulic pump, a servo valve, secondary components, a second drive motor, and a two-position four-way solenoid directional valve.

[0017] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows: This invention integrates unstable wind energy with grid power and recyclable hydraulic / kinetic energy within the system via hydraulic transmission. The method employs a model predictive control (MPC)-based energy management strategy, achieving dynamic matching and coordinated operation between wind energy, grid power, and recovered energy with the up-and-down stroke requirements of the pumping unit. This effectively solves the problem of existing technologies failing to deeply couple unstable wind energy resources into industrial machinery drive systems, thereby improving the depth and efficiency of comprehensive energy utilization in the pumping unit drive system.

[0018] This invention maximizes the capture of random wind energy through a maximum power point tracking strategy, directly converting it into hydraulic energy for driving or storage, thus achieving "source generation" and proactively reducing energy consumption from the external power grid. Simultaneously, through a hydraulic secondary balancing power subsystem and intelligent control, it efficiently recovers and stores the kinetic and potential energy during the pumping unit's descent, braking, or reversal, achieving "energy saving." Addressing the long-term inefficiency and wasted braking energy caused by "over-powered" systems in existing technologies, this invention significantly improves the overall energy efficiency of the system.

[0019] This invention utilizes multi-objective optimization through model predictive control to fully leverage unstable wind energy and recover energy while prioritizing the stability of the drive motor's output torque. This effectively suppresses mechanical shocks and vibrations caused by periodic commutation of the pumping unit and wind energy fluctuations. Simultaneously, the accumulator's peak-shaving and valley-filling effect smooths the instantaneous power demand on the grid side, reducing the impact on the oilfield's regional power grid.

[0020] This invention introduces a long short-term memory (LSTM) neural network to predict short-term wind speed and pumping unit load demand, and combines it with model predictive control for feedforward optimization. This enables the system to sense changes in wind conditions and load trends in advance, and actively adjust energy distribution and storage strategies, thereby improving the system's adaptability under real-world unstable operating conditions.

[0021] This invention defines and applies "State of Charge (SOP)" as the core indicator for energy management of accumulators, enabling the controller to make real-time decisions on charging and discharging priorities based on the SOP. This ensures that the accumulator is always in a highly efficient and safe operating range. Through quantitative energy state management, the system's operating status is optimized and component lifespan is extended.

[0022] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1This is a schematic diagram of the hybrid drive system for an oil pumping unit according to the present invention; Figure 2 This is a schematic flowchart of the hybrid drive control method for an oil pumping unit according to the present invention.

[0024] The components are as follows: 1. Wind turbine; 2. Gearbox; 3. First variable displacement hydraulic pump; 4. Hydraulic oil tank; 5. Check valve; 6. Accumulator; 7. Pressure transmitter; 8. Hydraulic oil tank; 9. Shuttle valve; 10. Second relief valve; 11. First drive motor; 12. Second variable displacement hydraulic pump; 13. Hydraulic oil tank; 14. Controller; 15. Servo valve; 16. Double-acting double-rod hydraulic cylinder; 17. Hydraulic oil tank; 18. Filter; 19. Secondary components; 20. First torque and speed sensor; 21. Driven wheel; 22. Drive belt; 23. Drive wheel; 24. Second torque and speed sensor; 25. Second drive motor; 26. Overrunning clutch; 27. Load; 28. Hydraulic oil tank; 29. ​​Third relief valve; 30. Fourth relief valve; 31. Two-position four-way solenoid directional valve; 32. Hydraulic oil tank; 33. First relief valve. Detailed Implementation

[0025] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0026] Firstly, such as Figure 1 As shown, the hybrid drive system for an oil pumping unit provided by the present invention includes: The wind power-hydraulic conversion subsystem includes a wind turbine 1 and a first variable displacement hydraulic pump 3 connected to a hydraulic oil tank. The first variable displacement hydraulic pump 3 is connected to the wind turbine 1. The hydraulic secondary balancing power subsystem includes a secondary element 19 connected to a hydraulic oil tank and an accumulator 6. The sensor detection and control subsystem includes a controller 14, and a pressure transmitter 7 and a torque and speed sensor that are communicatively connected to the controller 14; the controller 14 is communicatively connected to the wind turbine 1 and the first variable displacement hydraulic pump 3.

[0027] The wind energy-hydraulic conversion subsystem, the hydraulic secondary balance power subsystem, and the sensor detection and control subsystem are connected in pairs.

[0028] The wind power-hydraulic conversion subsystem also includes a one-way valve 5 and a first relief valve 33. The first relief valve 33 is connected in parallel with the first variable displacement hydraulic pump 3, and the one-way valve 5 is connected in series with the first relief valve 33 and the first variable displacement hydraulic pump 3. The hydraulic secondary balancing power subsystem includes a servo valve 15, a second relief valve 10 connected to the oil tank, a third relief valve 29, a fourth relief valve 30, and a two-position four-way solenoid directional valve 31. The servo valve 15 is connected to a double-acting double-rod hydraulic cylinder 16, which is connected to the secondary element 19. The servo valve 15 is communicatively connected to the controller 14 and is connected to the hydraulic oil tank.

[0029] The second relief valve 10 is connected in parallel with a second variable displacement hydraulic pump 12, which is connected to a first drive motor 11. The second relief valve 10 and the second variable displacement hydraulic pump 12 are connected in series with a shuttle valve 9, which is connected to the servo valve 15.

[0030] The third relief valve 29 is connected in parallel with the secondary element 19. The secondary element 19, the third relief valve 29, and the fourth relief valve 30 are all connected to the two-position four-way solenoid directional valve 31. The two-position four-way solenoid directional valve 31 is communicatively connected to the controller 14 and is connected to the hydraulic oil tank. A filter 18 is provided between the secondary element 19 and the hydraulic oil tank. The secondary element 19 is connected to the load 27 via a transmission assembly. The transmission assembly includes a drive wheel 23 and a driven wheel 21. The drive wheel 23 and the driven wheel 21 are connected by a transmission belt 22. The drive wheel 23 is connected to a second drive motor 25. An overrunning clutch 26 is provided between the driven wheel 21 and the load 27.

[0031] The pressure transmitter 7 is located in the hydraulic circuit between the first variable displacement hydraulic pump 3 and the shuttle valve 9. The torque and speed sensor includes a first torque and speed sensor 20 and a second torque and speed sensor 24. The first torque and speed sensor 20 is located between the driven wheel 21 and the secondary element 19, and the second torque and speed sensor 24 is located between the second drive motor 25 and the driving wheel 23.

[0032] like Figure 1 As shown, the hybrid drive system of the present invention mainly covers three subsystem levels: a wind turbine 1, a gearbox (optional) 2, a first variable displacement hydraulic pump 3 connected to the oil tank, a one-way valve 5, and a first relief valve 33 connected to the oil tank, which constitute a wind energy-hydraulic conversion subsystem, wherein the first variable displacement hydraulic pump 3 and the first relief valve 33 are connected in parallel, and the two are connected in series with the one-way valve 5. The hydraulic secondary balancing power subsystem consists of an accumulator 6, a shuttle valve 9, a second relief valve 10 connected to the hydraulic oil tank, a second variable displacement hydraulic pump 12 driven by a first drive motor 11, a servo valve 15, a double-acting double-rod hydraulic cylinder 16, a filter 18, a secondary element 19, a third relief valve 29 and a fourth relief valve 30 connected to the hydraulic oil tank, a two-position four-way solenoid directional valve 31, a driven pulley 21, a transmission belt 22, a drive pulley 23, a second drive motor 25, and an overrunning clutch 26. The second relief valve 10 and the second variable displacement hydraulic pump 12 are connected in parallel, and both are connected in series with the shuttle valve 9. The third relief valve 29 and the secondary element 19 are connected in parallel on the same branch, and both, along with the fourth relief valve 30, are connected to the two-position four-way solenoid directional valve 31. The servo valve 15 is directly connected to the double-acting double-rod hydraulic cylinder 16, and the double-acting double-rod hydraulic cylinder 16 is mechanically connected to the secondary element 19. These three components form a simple mechanical-hydraulic servo mechanism. The controller 14, pressure transmitter 7, first torque-speed sensor 20, second torque-speed sensor 24, and their connection to the dotted line constitute a sensing and control subsystem. The second drive motor 25 of the hybrid drive system of the pumping unit described in this invention is connected to the load 27. The remaining reference numerals 4, 8, 13, 17, 28, and 32 indicate structures representing hydraulic oil tanks. It is understood that... Figure 1 Solid lines indicate hydraulic circuit connections, dashed lines indicate communication connections, and double solid lines indicate mechanical connections.

[0033] The working principle of the system of this invention is as follows: Upon system startup, the second drive motor 25 reverses, and the overrunning clutch 26 transmits power to the secondary element 19 via a transmission device / assembly consisting of the driving pulley 23, the transmission belt 22, and the driven pulley 21. At this time, the secondary element 19 charges hydraulic oil into the accumulator 6, completing the initial charging of the system. The controller 14 makes the decision based on the oil supply status of the first variable displacement hydraulic pump 3 driven by the wind turbine 1. If the oil supply is sufficient to meet the initial requirements of the accumulator, the second drive motor 25 will not reverse.

[0034] When the load is fully operational, the driving power for the load 27 is provided by the second drive motor 25 and the secondary element 19, and the power of the secondary element 19 is jointly provided by the accumulator 6 and the first variable displacement hydraulic pump 3 driven by the wind turbine 1. Especially when the load has periodic motion characteristics, the second drive motor 25 will change its rotation direction together with the load, so that the secondary element 19 is in the "pump" working state, converting the kinetic energy of the reverse motion into hydraulic energy and storing it in the accumulator 6, realizing power / energy recovery for the forward motion of the next cycle, and the power burden on the second drive motor 25 can be reduced by the secondary element 19 being in the "motor" working state. In addition, the controller 14 can comprehensively adjust the displacement of the first variable displacement hydraulic pump 3, the throttling status of the servo valve 15, and the working state of the two-position four-way solenoid directional valve 30 through the digital signals of the pressure transmitter 7, the first speed-torque transmitter 20, and the second speed-torque transmitter 24.

[0035] Given that the hybrid drive system contains multiple types of energy supply sources, in order to maximize the use of clean energy, improve the system's self-powered performance, and achieve stable output of the motor's load torque, a feedback principle is considered, which is mainly based on the optimization method of the controller (main control strategy) uniformly allocating control to each sub-energy supply source, thereby realizing the system's "top-down" control path.

[0036] Based on this, the second aspect, such as Figure 2 As shown, this invention provides a hybrid drive control method for oil pumping units, applicable to the aforementioned hybrid drive system for oil pumping units. The method employs an energy management strategy based on model predictive control (MPC), and includes: S1: Real-time acquisition of wind speed and wind turbine rotation speed signals, accumulator pressure signals, secondary component rotation speed and torque signals, and second drive motor torque and rotation speed signals.

[0037] It is understood that the signals collected by this invention are not limited to these. For example, temperature signals can be collected by a temperature sensor, flow rate signals can be collected by a flow rate sensor, voltage and current signals can be collected by a voltage and current sensor, etc. It is possible for those skilled in the art to set up multiple sensors in the system to collect corresponding signals for system control according to actual needs.

[0038] S2: Based on the rotational speed signal of the wind turbine, the target displacement is determined based on maximum power point tracking. Based on the real-time pressure signal of the accumulator, the load pressure state SOP, which characterizes the energy storage state of the accumulator, is calculated. Based on historical data, the wind speed change trend and load demand change within a certain time window are predicted through a long short-term memory neural network.

[0039] Specifically, the wind energy maximization capture sub-strategy based on maximum power point tracking (MPPT) detects wind speed. (or wind turbine speed) ) and hydraulic pump output pressure Calculate the optimal torque setting value at the current wind turbine speed. This value corresponds to the maximum power point at the current wind speed. Then, the hydraulic pump displacement is adjusted. This enables the hydraulic pump to track the load torque. Due to the torque and displacement of the hydraulic pump and pressure It is directly proportional, therefore the target displacement can be calculated using the following formula: ; in, This represents the maximum wind energy utilization factor, which is 0.593. This indicates the optimal tip speed ratio.

[0040] Let the optimal torque constant be For any given wind turbine must.

[0041] Specifically, the "State of Load" (SOP) is introduced as an energy management indicator for the accumulator: To quantify and manage the energy reserves in the energy storage device 6, this invention defines a "State of Pressure (SOP)": ; in, For the real-time pressure of the accumulator, and The SOP (State Operation Point) is the upper limit of the system's allowed safe operating pressure and the pre-charge pressure. As a critical state input controller, it determines whether to prioritize using energy from the accumulator or charging it to maintain the system's energy buffering capacity. The specific decision-making process is as follows: High-energy state thresholds are defined through prior computer simulations of the actual system or through experience gained from early trial runs. and low energy state threshold The specific values ​​are determined, and the optimal operating value of the system is also determined. .

[0042] Based on the range of the real-time SOP value, the controller 14 executes the corresponding energy management strategy. Its logical judgment process and system actions are as follows: When detected When the load 27 is braking or reversing, the secondary element 19 is prioritized for charging the accumulator 6. At this time, the controller sets the power command output of the secondary element 19 in motor mode to zero or the minimum value, limiting or prohibiting the release of energy from the accumulator 6. When the load 27 is braking or reversing, the secondary element 19 is forced to operate in pump mode to store the recovered energy into the accumulator 6 as much as possible.

[0043] In addition, if If the energy supply is insufficient and the system allows, the second drive motor 25 is reversed (via the overrunning clutch 26) to drive the secondary element 19 as a pump to charge the accumulator. Even if wind energy is sufficient, ensuring the charging of the accumulator 6 is prioritized, which may temporarily deviate from the MPPT optimal curve to match the charging demand.

[0044] When detected When the energy stored in the accumulator 6 is released, the controller 14 sets the power command for the secondary element 19 to recover energy in pump mode to zero. Simultaneously, when the load 27 requires power, the controller prioritizes controlling the secondary element 19 to operate as a motor, increasing its displacement to release energy from the accumulator 6 to bear part of the load, thereby reducing the power output of the second drive motor 25. Furthermore, if wind energy continues to be input and the accumulator 6 is full, the controller 14 will directly command the first variable displacement hydraulic pump 3 to reduce its displacement and deviate from the MPPT point, achieving "wind limiting" or "wind curtailment" to protect system safety.

[0045] When detected At that time, controller 14 will... Set as , this The input to the optimization function of the MPC master policy (i.e., the multi-objective optimization function) The optimization function is solved for the item.

[0046] Other, The settings primarily consider system security, efficiency, and dynamic response margin. The settings primarily consider the system's power buffering capacity, response stability, and emergency requirements. In practical applications, reasonable values ​​can be derived based on engineering experience, taking into account specific system pressure levels, accumulator volume, load cycle characteristics, and safety factor requirements. The core of this invention lies in the decision-making method itself based on dual thresholds for partitioning decisions, rather than specific threshold values. Therefore, the specific determination process and values ​​of the thresholds are not elaborated upon.

[0047] Specifically, the load or wind speed prediction sub-strategy based on LSTM network: By introducing a prediction module to construct a long short-term memory (LSTM) neural network, it can be implemented according to the sensors configured in the system: by installing wind speed sensors, the LSTM network is trained to learn the time series characteristics of historical wind speed data, which is used to predict the wind speed change trend in the future short time window; by using the periodic change properties of load movement, the LSTM network is trained to learn historical load torque or movement cycle data, which is used to predict the load demand change in the future short term.

[0048] S3: Construct a state-space prediction model based on a pre-built subsystem model.

[0049] The subsystem model includes: a wind-liquid power supply model, a simplified steady-state model of the energy storage device, a secondary component model, and a drive motor model.

[0050] Specifically, establish based on wind speed (or wind turbine speed) ) and the first variable displacement hydraulic pump with a displacement of 3 The mapping relationship between wind and liquid forms a "wind-liquid" energy supply model, which is used to characterize the power captured by wind energy. : ; in, Let R be the air density and R be the rotor radius. The hydraulic pump outputs hydraulic power. Its displacement outlet pressure and speed Related. Also. The wind energy utilization coefficient is the tip speed ratio. A function, which is expressed by the following empirical formula: ; ; in arrive These are constants that depend on the specific design of the wind turbine. For a given wind turbine, these constants are fixed. Meanwhile, this invention considers the wind turbine pitch angle... It is a fixed value, and its changes are not considered.

[0051] Specifically, a simplified steady-state model of the accumulator is established, neglecting heat exchange, and the stored hydraulic energy... With gas pressure and volume Relevant. The model is expressed as follows: ; in, This indicates the minimum operating pressure of the accumulator. Indicates pressure The volume of the gas below (i.e., the maximum usable gas volume of the accumulator) Indicates the highest work pressure and has The total volume of the accumulator is usually known during calculations. and precharge pressure Then Substituting into the formula for polytropic gas processes ( (where the constant is) can be used to obtain .

[0052] Specifically, a secondary element model is established. When secondary element 19 is used as a pump / motor, its displacement when used as a motor or pump is determined. Inlet and outlet pressure difference With transmission torque Between rotational speeds The steady-state relationship is expressed by the model as follows: ; ; in, Indicates the flow rate of the secondary component. and A positive superscript indicates the mechanical (volumetric) efficiency of the motor, while a negative superscript indicates the mechanical (volumetric) efficiency of the pump. and The value ranges from 0 to 1, excluding 0 and 1.

[0053] Specifically, a drive motor model is established, and the electrical parameters of the second drive motor 25 and its output torque are determined. Rotation speed The dynamic relationship between them. This model is expressed as follows: ; ; ; in, Represents the motor torque constant. Indicates the armature resistance of the motor. This represents the back electromotive force constant of the motor. This indicates the voltage at the motor terminals.

[0054] The established subsystem models are discretized and combined into an overall state-space prediction model of the system, which is used to predict the evolution of the system state (such as SOP, motor torque, etc.) under future control sequences.

[0055] S4: Construct a multi-objective optimization function with the objectives of stabilizing the output torque of the drive motor, maintaining the load state SOP within the target range, and minimizing the deviation between the actual wind power and the optimal value.

[0056] Specifically, the multi-objective optimization function is designed as follows: in, Let k be the future control sequence to be solved, k be the current time, and i be the prediction step size index (from 1 to k). ); For prediction in the time domain; the first term aims to stabilize the motor output torque. Track its normal operating values The reference normal value can be determined by load demand; the second item aims to maintain the accumulator's SOP within the ideal range. Nearby; the third term aims to minimize the deviation between the actual wind power and the optimal value, so that the actual wind power... Approaching the optimal value . This is the weighting coefficient for the motor torque tracking error. This is the weighting coefficient for SOP tracking error. This is the weighting coefficient for wind power tracking error.

[0057] The constraints applied during the calculation of this multi-objective function include, but are not limited to: Actuator limiting: , ; Accumulator pressure range: ; Power / torque balance equation: ; ; in This refers to the motor's output power. The power of the secondary components (positive for output, negative for input). Power consumed by the load The motor torque constant is This represents the motor current.

[0058] S5: Within each control cycle, based on the currently acquired system state and the predicted wind speed and load demand change sequences, the optimization function with constraints is solved in a rolling manner to obtain the optimal control sequence for each actuator control quantity in the future control time domain.

[0059] S6: The first control command in the optimal control sequence is sent to the corresponding actuator. Each actuator then coordinates the use of wind energy, grid power, and recovered energy to jointly drive the load and recover energy.

[0060] MPC solves the constrained optimization problem online to obtain the optimal control sequence, but only the first step of the control input is sent to each actuator (first variable displacement hydraulic pump 3, servo valve 15, secondary element 19, second drive motor 25, and two-position four-way solenoid directional valve 31). At the next sampling time, the system collects new state measurements (pressure, torque, etc.), combines them with the updated prediction sequence from LSTM, and repeats the optimization calculation to achieve closed-loop rolling optimization control.

[0061] In one embodiment, the system cycle control process is as follows: After the system starts up, the controller continues to run the above-mentioned integrated energy management strategy: Initialize the detection accumulator SOP; if it is lower than... Then the drive motor is started to reverse, and the accumulator is filled with liquid to the set threshold through the secondary components; Real-time acquisition of wind speed / load, pressure, torque, and speed signals; The MPPT sub-strategy calculates the optimal pump displacement recommendation value based on the current wind speed; The LSTM prediction module outputs future wind speed / load sequences; The MPC master strategy is based on the current system state, the predicted sequence, and the optimization objective, and it solves the problem in a rolling manner and issues comprehensive control commands. Each actuator operates, driving the load and recovering energy, and the system enters the next control cycle.

[0062] Compared with the prior art, the present invention has the following advantages: 1. This invention integrates unstable wind energy with grid power and recyclable hydraulic / kinetic energy within the system via hydraulic transmission. Through a model predictive control (MPC)-based energy management strategy, it achieves dynamic matching and coordinated operation between wind energy, grid power, and recovered energy, aligning with the up-and-down stroke requirements of the pumping unit. This improves the depth and efficiency of comprehensive energy utilization in the pumping unit drive system.

[0063] 2. This invention achieves both "open source" and "cost reduction", resulting in a significant improvement in overall energy efficiency: Open source: By using a maximum power point tracking strategy, it maximizes the capture of random wind energy and directly converts it into hydraulic energy to drive the pumping unit load or for storage, thus actively reducing the power consumption of the external power grid. Energy saving: Through the hydraulic secondary balance power subsystem and intelligent control, the kinetic and potential energy of the pumping unit during downward movement, braking or reversing is efficiently recovered and stored in the accumulator. It is then released again during energy-demanding periods such as the load stroke, effectively reducing energy waste within the system.

[0064] 3. This invention utilizes multi-objective optimization through model predictive control to fully leverage unstable wind energy and recover energy while prioritizing the stability of the drive motor's output torque. This effectively suppresses mechanical shocks and vibrations caused by periodic commutation of the pumping unit and wind energy fluctuations. Simultaneously, the accumulator's peak-shaving and valley-filling effect smooths the instantaneous power demand on the grid side, reducing the impact on the oilfield's regional power grid.

[0065] 4. This invention introduces a long short-term memory neural network to predict short-term wind speed and pumping unit load demand, and combines this with model predictive control for feedforward optimization. This enables the system to anticipate changes in wind conditions and load trends, proactively adjust energy allocation and storage strategies, and improve the system's adaptability under real-world field conditions.

[0066] 5. This invention defines and applies "State of Charge (SOP)" as the core indicator for energy management of the accumulator. The controller makes real-time decisions on charging and discharging priorities based on the SOP, ensuring that the accumulator is always in a highly efficient and safe operating range, thus extending the component's lifespan.

[0067] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0068] It will be readily understood by those skilled in the art that this invention includes any combination of the inventive description and specific embodiments outlined in the foregoing specification, as well as the various parts shown in the accompanying drawings. Due to space limitations and for the sake of brevity, not all of these combinations have been described in detail. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0069] Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A hybrid drive system for an oil pumping unit, characterized in that, include: The wind power-hydraulic conversion subsystem includes a wind turbine and a first variable displacement hydraulic pump connected to a hydraulic oil tank, the first variable displacement hydraulic pump being connected to the wind turbine; The hydraulic secondary balancing power subsystem includes secondary components and accumulators connected to the hydraulic oil tank; The sensor detection and control subsystem includes a controller, and a pressure transmitter and a torque-speed sensor that are communicatively connected to the controller; the controller is communicatively connected to the wind turbine and the first variable displacement hydraulic pump. The hydraulic secondary balancing power subsystem includes a servo valve, a second relief valve, a third relief valve, a fourth relief valve connected to the hydraulic oil tank, and a two-position four-way solenoid directional valve. The servo valve is connected to a double-acting, double-rod hydraulic cylinder, which is connected to the secondary component. The servo valve is communicatively connected to the controller. The second relief valve is connected in parallel to a second variable displacement hydraulic pump, which is connected to a first drive motor. A shuttle valve is connected in series with the second relief valve and the second variable displacement hydraulic pump. The third relief valve is connected in parallel with the secondary component. The secondary component, the third relief valve, and the fourth relief valve are all connected to the two-position four-way solenoid directional valve. The two-position four-way solenoid directional valve is communicatively connected to the controller. A filter is provided between the secondary component and the hydraulic oil tank. The secondary element is connected to the load via a transmission assembly. The transmission assembly includes a driving wheel and a driven wheel, which are connected by a transmission belt. The driving wheel is connected to a second drive motor, and an overrunning clutch is provided between the driven wheel and the load. The pressure transmitter is located in the hydraulic circuit between the first variable displacement hydraulic pump and the shuttle valve. The torque and speed sensor includes a first torque and speed sensor and a second torque and speed sensor. The first torque and speed sensor is located between the driven wheel and the secondary element, and the second torque and speed sensor is located between the second drive motor and the driving wheel.

2. The hybrid drive system for an oil pumping unit as described in claim 1, characterized in that, The wind power-hydraulic conversion subsystem also includes a one-way valve and a first relief valve. The first relief valve is connected in parallel with the first variable displacement hydraulic pump, and the one-way valve is connected in series with the first relief valve and the first variable displacement hydraulic pump.

3. A hybrid drive control method for an oil pumping unit, applicable to the hybrid drive system described in claim 1 or 2, characterized in that, The method employs an energy management strategy based on model predictive control, including: Real-time acquisition of wind speed and wind turbine rotation speed signals, accumulator pressure signals, secondary component rotation speed and torque signals, and second drive motor torque and rotation speed signals; Based on the wind turbine's rotational speed signal, the target displacement is determined by maximum power point tracking. Based on the accumulator's real-time pressure signal, the load pressure state (SOP) characterizing the accumulator's energy storage state is calculated. Based on historical data, the wind speed change trend and load demand change within a certain time window are predicted using a long short-term memory neural network. A state-space prediction model is constructed based on a pre-built subsystem model. Construct a multi-objective optimization function with the objectives of stabilizing the output torque of the second drive motor, maintaining the load pressure state SOP within the target range, and minimizing the deviation between the actual wind power and the optimal value; The subsystem model includes: a wind-liquid power supply end model, a simplified steady-state model of the energy storage device, a secondary component model, and a second drive motor model; Within each control cycle, based on the currently acquired system state and the predicted wind speed and load demand change sequences, the optimization function with constraints is solved in a rolling manner to obtain the optimal control sequence for each actuator control quantity in the future control time domain. The first control command in the optimal control sequence is issued to the corresponding actuator, and each actuator acts to coordinate wind energy, grid power and recovered energy to jointly drive the load and recover energy.

4. The hybrid drive control method for an oil pumping unit as described in claim 3, characterized in that, The wind-liquid power supply end model is as follows: ； ； ； ； in, air density, Where is the radius of the wind turbine. The wind turbine rotation speed, For wind speed, The wind energy utilization coefficient is the tip speed ratio. The function, arrive It is the wind turbine design constant. The pitch angle of the wind turbine blade; The simplified steady-state model of the accumulator is as follows: ； in, This is the minimum operating pressure of the accumulator. For pressure The volume of gas below, For the highest work pressure and have ; The secondary component model is as follows: ； ； in, To transmit torque, The displacement when a secondary component is used as a motor or pump. For the pressure difference between inlet and outlet, Indicates the flow rate of the secondary component. and A positive superscript indicates the mechanical efficiency of the motor, while a negative superscript indicates the mechanical efficiency of the pump. Rotational speed; The second drive motor model is as follows: ； ； ； in, This is the output torque of the second drive motor. The speed of the second drive motor. The torque constant of the motor. For the motor armature resistance, Let be the back electromotive force constant of the motor. This is the voltage at the motor terminals.

5. The hybrid drive control method for an oil pumping unit as described in claim 3, characterized in that, The formula for calculating the State of Load (SOP) is as follows: ； in, For the real-time pressure of the accumulator, and These are the upper limit of the permissible safe operating pressure of the accumulator and the pre-charge pressure, respectively.

6. The hybrid drive control method for an oil pumping unit as described in claim 3, characterized in that, The multi-objective optimization function is as follows: in, Let k be the future control sequence to be solved, k be the current time, and i be the prediction step size index. For time-domain prediction, the first term aims to stabilize the motor output torque. Track its normal operating values The second objective is to maintain the accumulator's state of charge (SOP) within the ideal range. Nearby, the third term aims to minimize the deviation between the actual wind power and the optimal value, so that the actual wind power... Approaching the optimal value , For multi-objective optimization functions, This is the weighting coefficient for the motor torque tracking error. This is the weighting coefficient for the SOP tracking error under load conditions. This is the weighting coefficient for wind power tracking error.

7. The hybrid drive control method for an oil pumping unit as described in claim 3, characterized in that, The constraints are as follows: in, This refers to the motor's output power. For the power of secondary components, Power consumed by the load The torque constant of the motor. This represents the motor current.

8. The hybrid drive control method for an oil pumping unit as described in claim 3, characterized in that, The actuator includes a first variable displacement hydraulic pump, a servo valve, secondary components, a second drive motor, and a two-position four-way solenoid directional valve.

Images