Control processing method of hydraulic system, computer device and storage medium

Abstract

The application discloses a control processing method of a hydraulic system, a computer device and a storage medium. The method comprises the following steps: acquiring a required flow and an actual flow of the hydraulic system; obtaining a first control parameter by using a mechanism model according to the required flow; acquiring a flow error of the required flow and the actual flow; inputting the first control parameter and the flow error into a data model to obtain a control compensation parameter; determining a second control parameter according to the first control parameter and the control compensation parameter, and controlling the hydraulic system to output a corresponding target flow by using the second control parameter. In this way, the accuracy of the control parameter is improved, so that the hydraulic system can more accurately output a flow matching the actual requirement, and the control performance and the control reliability of the hydraulic system are improved.

Description

Technical Field

[0001] This application relates to the field of hydraulic technology, and in particular to a control and processing method for a hydraulic system, a computer device, and a storage medium. Background Technology

[0002] Hydraulic systems are widely used in many fields such as engineering machinery, aerospace, metallurgical equipment, and shipbuilding due to their advantages such as high power density, fast response speed, and good control precision. With the continuous improvement of industrial automation, the requirements for the control performance of hydraulic systems are increasing, especially under complex working conditions, where control precision has become one of the important indicators for measuring the performance of hydraulic systems.

[0003] In the control process of a hydraulic system, the system output flow rate exhibits complex relationships among multiple parameters. The coupling characteristics between these parameters cause the hydraulic system to exhibit strong nonlinear characteristics. Different types of hydraulic components, different structural designs, and different operating conditions can all lead to significant differences in system characteristics.

[0004] Traditional hydraulic system control methods are mainly divided into two categories: one is PID control and its improved methods. These methods have a simple structure, but the control effect is difficult to guarantee when faced with the nonlinear and time-varying characteristics of hydraulic systems; the other is model-based control methods. However, pure mechanistic models are difficult to accurately describe complex nonlinear characteristics, and pure data-driven models are obviously insufficient to cope with actual operating conditions.

[0005] Therefore, how to improve the control accuracy of hydraulic systems, thereby enhancing their control performance and reliability, has become an urgent technical problem to be solved. Summary of the Invention

[0006] The purpose of this application is to provide a control processing method, computer equipment, and storage medium for a hydraulic system, which can improve the accuracy and reliability of hydraulic system control.

[0007] To achieve the above objectives: In a first aspect, embodiments of this application provide a control processing method for a hydraulic system, comprising: acquiring the required flow rate and the actual flow rate of the hydraulic system; obtaining a first control parameter based on the required flow rate using a mechanism model; acquiring the flow error between the required flow rate and the actual flow rate; inputting the first control parameter and the flow error into a data model to obtain a compensation parameter; determining a second control parameter based on the first control parameter and the control compensation parameter, and using the second control parameter to control the hydraulic system to output a corresponding target flow rate.

[0008] In one embodiment, the method further includes: acquiring training sample sets under various different operating conditions, the training sample sets including first control parameter samples, flow error samples between demand flow samples and actual flow samples, and corresponding optimal control compensation parameter samples; using the first control parameter samples and flow error samples as inputs and the optimal control compensation parameters as outputs, training a preset machine learning model to obtain a data model.

[0009] In one embodiment, the operating conditions include at least two operating conditions. Under multiple different operating conditions, training sample sets are obtained respectively, including: obtaining corresponding training sample subsets under multiple operating condition values ​​of the same operating condition. The training sample sets include training sample subsets corresponding to at least two operating conditions.

[0010] In one embodiment, the method further includes: at least two operating conditions, including at least two of oil temperature, engine speed and load pressure, and obtaining corresponding training sample subsets under multiple operating condition values, including at least one of the following: obtaining a first training sample subset under multiple oil temperature values; obtaining a second training sample subset under multiple engine speed values; and obtaining a third training sample subset under multiple load pressure values.

[0011] In one embodiment, obtaining the optimal compensation parameter sample under each operating condition includes: under the target operating condition, adjusting the test control compensation parameter to adjust the test second control parameter so that the percentage of the test flow error between the test demand flow and the test actual flow converges to a preset threshold range, and the test second control parameter and the test actual flow have a corresponding relationship; recording the test first control parameter under the target operating condition, the test flow error between the test demand flow and the test actual flow, and the corresponding relationship of the test control compensation parameter; and determining the test control parameter as the optimal compensation parameter sample under the target operating condition.

[0012] In one embodiment, the preset threshold range is a percentage greater than or equal to a first threshold and less than or equal to a second threshold. Adjusting the test control compensation parameter by adjusting the test control compensation parameter includes: adjusting the test control compensation parameter in a step-wise manner to adjust the test control compensation parameter; wherein, adjusting the test control compensation parameter in a step-wise manner includes: when the percentage of the test flow error between the test demand flow and the test actual flow is less than the first threshold, increasing the test control compensation parameter by increasing a fixed step value, so that the adjusted test control parameter is the test first control parameter plus the test compensation parameter; when the percentage of the test flow error between the test demand flow and the test actual flow is greater than the second threshold, decreasing the test control compensation parameter by decreasing a fixed step value, so that the adjusted test control parameter is the test first control parameter minus the test compensation parameter; when the percentage of the test flow error between the test demand flow and the test actual flow is between the first threshold and the second threshold, maintaining the test control compensation parameter.

[0013] In one embodiment, obtaining the optimal compensation parameter sample under each operating condition includes: calculating the optimal compensation parameter sample based on the second test control parameter corresponding to the percentage convergence of the test flow error between the test demand flow and the test actual flow within a preset threshold range, and the first test control parameter.

[0014] In one embodiment, the mechanism model includes a first mechanism model and a second mechanism model. Based on the demand flow rate, the mechanism model is used to obtain a first control parameter, which includes: determining a current first flow rate influence parameter based on the demand flow rate using the first mechanism model; determining a target second flow rate influence parameter based on the current first flow rate influence parameter using the second mechanism model; and determining the first control parameter of the hydraulic system based on the target second flow rate influence parameter.

[0015] Secondly, embodiments of this application provide a computer device, including: a processor and a memory storing a computer program, wherein when the processor runs the computer program, the control processing method of the hydraulic system described above is implemented.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the control processing method of the hydraulic system described above.

[0017] The hydraulic system control processing method, computer equipment, and storage medium provided in this application embodiment acquire the required flow rate and actual flow rate of the hydraulic system; obtain a first control parameter based on the required flow rate using a mechanism model; acquire the flow error between the required flow rate and the actual flow rate; input the first control parameter and the flow error into a data model to obtain a control compensation parameter; determine a second control parameter based on the first control parameter and the control compensation parameter; and use the second control parameter to control the hydraulic system to output the corresponding target flow rate. In this way, the embodiments of this application obtain control compensation parameters by inputting the first control parameter and the flow error into a pre-trained data model. On the one hand, by inputting the first control parameter obtained from the mechanistic model and the flow error into the data model together, the mechanistic model and the data model are organically integrated. This utilizes the accurate description of the physical characteristics of the system by the mechanistic model and leverages the compensation capability of the data model for complex nonlinear factors, thereby improving the control accuracy of the hydraulic system. On the other hand, by using the flow error as one of the inputs to the data model, the generation process of the compensation parameter can perceive the magnitude and direction of the current control deviation in real time, forming a closed-loop feedback compensation mechanism, which effectively suppresses the impact of external interference and parameter changes on the control effect. In summary, by compensating and correcting the first control parameter with the compensation parameter, the accuracy of control can be further improved, thereby improving the matching degree between the output flow and the actual demand flow, and further enhancing the control performance and reliability of the hydraulic system. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the control and processing method of the hydraulic system provided in the embodiments of this application.

[0019] Figure 2 Another schematic flowchart of the control method for the hydraulic system provided in the embodiments of this application.

[0020] Figure 3 This is a schematic diagram of a scenario for the data model provided in an embodiment of this application.

[0021] Figure 4 This is a schematic diagram illustrating the linear relationship between valve core displacement and command current in a hydraulic system provided in this application embodiment.

[0022] Figure 5 A flowchart illustrating the method for constructing a data model provided in this application embodiment.

[0023] Figure 6 This is a schematic diagram illustrating the linear relationship between valve core displacement and command current in a hydraulic system provided in this application embodiment.

[0024] Figure 7 This is a schematic diagram illustrating the relationship between the throttling area of ​​the valve core and the displacement of the valve core in a hydraulic system provided in this application embodiment.

[0025] Figure 8 A schematic diagram illustrating the specific flow of the control and processing method for the hydraulic system provided in this application embodiment.

[0026] Figure 9 This is a schematic diagram of the structure of a control and processing device for a hydraulic system provided in an embodiment of this application.

[0027] Figure 10 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the scope of the application.

[0029] The following is an explanation of the terms that may be involved in this embodiment: A hydraulic system is a technical system that uses a liquid (usually hydraulic oil) as the working medium. It converts mechanical energy into pressure energy through a hydraulic pump, and then converts that pressure energy back into mechanical energy through actuators such as hydraulic cylinders and hydraulic motors, thereby achieving power transmission and control. Typically, a hydraulic system includes: power components, such as a hydraulic pump, which converts mechanical energy into liquid pressure energy; actuators, such as hydraulic cylinders and hydraulic motors, which convert liquid pressure energy into mechanical energy; control components, such as hydraulic valves, which control the direction, pressure, and flow rate of the fluid; and auxiliary components, such as oil tanks, pipelines, filters, and accumulators. For example, hydraulic systems can be applied to construction machinery, such as excavators and cranes; aerospace applications, such as landing gear and steering surface control; and also to metallurgical equipment and shipbuilding industries.

[0030] Mechanistic model: Also known as a white-box model or physical model, it is a model established based on the fundamental physical and chemical principles and laws of the system, and described by mathematical equations to describe the inherent laws of the system. This model is based on at least one of the following: physical laws, fluid mechanics principles, and system structural parameters. The physical laws include, but are not limited to, at least one of Newton's laws, conservation of energy, and conservation of mass; the fluid mechanics principles include the thin-walled orifice flow formula and Bernoulli's equation; and the system structural parameters include the relationship between valve core displacement and flow area, and spring stiffness.

[0031] Data models, also known as black-box models, are models built based on input-output data. They learn the behavioral patterns of a system through methods such as machine learning and statistical analysis, without relying on a deep understanding of the system's internal physical mechanisms. The construction of data models typically requires a large amount of training data and machine learning algorithms, such as neural networks, support vector machines, or decision trees. These data models can fit complex nonlinear relationships, are relatively simple to model, and offer high accuracy.

[0032] Residual compensation model: This is a special type of data model used to learn and compensate for the prediction error (i.e., residual) of the main model (such as a mechanistic model), thereby improving the accuracy of the overall model by superimposing compensation.

[0033] See Figure 1This application provides a control processing method for a hydraulic system. This method can be executed by a hydraulic system processing device provided in this application. The control processing device can be implemented in software and / or hardware, such as a computer, server, embedded controller, or programmable logic controller. For example, the hydraulic system control processing method provided in this application can run in the vehicle controller of construction machinery, a dedicated hydraulic system controller, or a cloud control platform. The technical solution provided in this application can be applied to various scenarios requiring precise control of the hydraulic system's output flow, including but not limited to: construction machinery, industrial hydraulics, mobile hydraulics, and aerospace. In the field of construction machinery, hydraulic systems are used in applications such as cranes (e.g., boom extension / retraction, telescopic boom extension / retraction, slewing platform rotation, etc.) requiring uniform speed control; excavators (e.g., boom lifting / lowering, stick extension / retraction, bucket tilting, etc.); concrete pump trucks (e.g., boom deployment / retraction); and aerial work platforms (e.g., platform lifting / rotation). In the field of industrial hydraulics, applications include injection molding machines (e.g., mold closing, injection, ejection processes); forging equipment (e.g., sliding motion control); and metallurgical equipment (e.g., mill roll gap adjustment). In the field of mobile hydraulics, applications include agricultural machinery (e.g., suspension system lifting / lowering, harvester header adjustment); and mining vehicles (e.g., lifting system control). In the aerospace field, applications include aircraft landing gear retraction / extension, control surface control, and door opening / closing hydraulic actuation systems. In these applications, hydraulic systems often need to maintain stable flow output under complex conditions such as load changes, oil temperature fluctuations, and generator speed changes, requiring high control accuracy and response speed. The control processing method provided in this application can effectively improve the control accuracy of hydraulic systems under complex conditions, meeting the refined control needs of various operational scenarios. The execution entity in this application embodiment can be a control chip integrated in a hydraulic system controller, or a cloud server controlling the hydraulic system via remote communication. The controller can be a dedicated embedded controller, a programmable logic controller (PCL), or an industrial control computer, etc. This embodiment takes the hydraulic system controller as the execution entity for the hydraulic system control method as an example. The hydraulic system control processing method provided in this embodiment includes: Step S11: Obtain the required flow rate and actual flow rate of the hydraulic system.

[0034] Step S12: Based on the demand flow rate, obtain the first control parameter using the mechanism model.

[0035] Step S13: Obtain the traffic error between the required traffic and the actual traffic.

[0036] Step S14: Input the first control parameter and flow error into the data model to obtain the control compensation parameter.

[0037] Step S15: Determine the second control parameter based on the first control parameter and the control compensation parameter, and use the second control parameter to control the hydraulic system to output the corresponding target flow rate.

[0038] In this context, demand flow refers to the expected output flow value of the hydraulic system within the current control cycle. Demand flow reflects the flow supply required for hydraulic actuators, such as hydraulic cylinders and hydraulic motors, to complete their intended actions, and is the input target for hydraulic system control.

[0039] For example, the methods for obtaining demand traffic include, but are not limited to, at least one of the following: The flow rate is directly obtained from the detected operational commands. For example, in construction machinery, operators issue control commands through input devices such as joysticks, handles, or pedals. These command signals, such as handle angles and pedal travel, are directly converted into the required flow rate after calibration. For instance, in the luffing and lowering operation of a crane, the angle at which the operator pushes the handle directly determines the desired luffing speed. Through a preset handle angle-to-flow rate mapping relationship, the current required flow rate can be calculated in real time.

[0040] The required flow rate is calculated based on a preset control strategy. For example, in an automated control scenario, the required flow rate is calculated and generated in real time by the upper-level controller based on a preset motion trajectory, speed curve, or process requirements. For instance, during the mold closing process of an injection molding machine, the controller calculates the required flow rate value for the current stage in real time based on a preset mold closing speed curve.

[0041] External commands are received through a communication interface; for example, in a distributed control system, the required flow can be sent from the main controller to the hydraulic controller via a communication method such as CAN bus or Ethernet.

[0042] In this context, actual flow rate refers to the flow rate actually output by the hydraulic system to the actuator (such as a hydraulic cylinder or hydraulic motor) within the current control cycle. Actual flow rate is a key feedback quantity for measuring the control effect of a hydraulic control system, and together with the demand flow rate, it forms the basis of closed-loop control. It can be understood that the closer the actual flow rate is to the demand flow rate, the better the performance and reliability of the hydraulic control system; conversely, the closer the actual flow rate is to the demand flow rate, the worse the performance and reliability of the hydraulic control system.

[0043] For example, the actual flow rate can be directly measured by a sensor, such as installing a flow sensor at a critical location in the hydraulic system, like the control valve outlet or actuator inlet, to measure the actual flow rate through the pipeline in real time. For example, the flow sensor may include at least one of the following: a turbine flow meter, a gear flow meter, an ultrasonic flow meter, and an electromagnetic flow meter.

[0044] For example, the actual flow rate can also be indirectly calculated based on the motion parameters of the actuator. For instance, for a hydraulic cylinder, the actual flow rate can be calculated based on the piston speed v measured by a displacement sensor and the effective piston area A: Q = v·A. The piston speed can be obtained differentially from the displacement sensor or directly measured by a speed sensor.

[0045] For example, actual flow rate can also be obtained by employing state estimation methods. For instance, observer estimation can be used to estimate actual flow rate in real time based on measurable parameters (such as pressure and displacement) and the system model by constructing a system state observer (such as a Romberg observer).

[0046] For example, the actual traffic can also be the target traffic obtained subsequently.

[0047] For example, the mechanism model includes a first mechanism model and a second mechanism model. In step S12, obtaining the first control parameter based on the demand flow rate using the mechanism model may include: determining the current first flow rate influence parameter based on the demand flow rate through the first mechanism; determining the target second flow rate influence parameter based on the current first flow rate influence parameter through the second mechanism model; and determining the first control parameter of the hydraulic system based on the target second flow rate influence parameter.

[0048] For example, determining the current first flow impact parameter based on the demand flow through the first mechanism model may include: inputting the demand flow and state parameters into the mechanism model to calculate the current first flow impact parameter.

[0049] Among them, state parameters can refer to various physical quantities that reflect the current working state of the hydraulic system, and are the basic inputs for the calculation of the mechanism model.

[0050] For example, the status parameter may include: valve spool pressure drop. It is understood that valve spool pressure drop can refer to the pressure difference between the valve spool inlet and outlet when hydraulic oil passes through the valve spool. For example, the valve spool pressure drop can be obtained by at least one of the following methods: measurement by pressure sensors installed at the valve spool inlet and outlet; or, for some valve port structures, indirect calculation based on system pressure and load pressure.

[0051] For example, depending on the actual control requirements, the state parameters may also include: oil temperature, load pressure, engine speed, and valve core displacement, etc. Among them, oil temperature affects oil viscosity and density, which in turn affects the flow coefficient and flow characteristics; load pressure reflects the load condition of the hydraulic actuator; engine speed is the output flow rate of a hydraulic pump driven by an engine; and valve core displacement is the actual valve core position obtained by a displacement sensor.

[0052] The first flow-affecting parameter can be a physical quantity that reflects the flow capacity of the valve orifice and directly affects the output flow rate of the hydraulic system. Here, the first flow-affecting parameter can be the throttling area under a certain opening of the valve core. It can be understood that the throttling area refers to the effective flow cross-sectional area when hydraulic oil flows through the throttling orifice of the valve core, and this throttling area determines the flow rate through the valve orifice under the same pressure differential. It can be understood that the larger the throttling area, the stronger the flow capacity, and the greater the flow rate under the same pressure differential.

[0053] The first mechanism model can be a mathematical model based on the fundamental principles of fluid mechanics, used to describe the physical relationship between flow rate and related parameters in a hydraulic system. For example, taking the throttling area as the first flow rate influencing parameter and the valve core pressure drop as the state parameter, the first mechanism model can be the ideal thin-walled orifice flow formula (1): (1) in, Q To meet the demand for traffic, C d It is a flow coefficient, which can be an empirical value; for example, it can be a fixed value. This refers to the pressure drop across the valve core, or the valve core pressure drop (i.e., the state parameter mentioned above). ρ For example, the oil density is usually a constant when the oil temperature varies from 20℃ to 80℃; Ac is the throttling area of ​​a certain opening of the valve core (i.e., the current first flow rate influence parameter mentioned above).

[0054] Thus, using the first mechanism model described above, given the known demand flow and state parameters, the current first flow impact parameters can be calculated.

[0055] For example, determining the target second flow influence parameter based on the current first flow influence parameter through a second mechanism model includes: outputting the target second flow influence parameter based on the current first flow influence parameter through a second mechanism model, wherein the second mechanism model includes a mapping relationship constructed from the first flow influence parameter, the second flow influence parameter, and the third flow influence parameter that affects the second flow influence parameter of the hydraulic system.

[0056] The second flow-affecting parameter refers to a physical quantity that is directly related to the first flow-affecting parameter and reflects the positional state of the hydraulic valve spool. Here, the second flow-affecting parameter can be the valve spool displacement.

[0057] It is understandable that valve spool displacement refers to the distance the valve spool moves from its closed position (zero position) to its current operating position. Valve spool displacement determines the degree of valve opening, thus affecting the size of the flow area. In hydraulic valve design, there is a definite correspondence between valve spool displacement and throttling area. It is understandable that for most hydraulic valves, as valve spool displacement increases, the throttling area increases, and the flow capacity is enhanced.

[0058] The third flow-affecting parameter refers to other physical quantities that have a coupled effect on the second flow-affecting parameter (such as valve core displacement). Here, the third flow-affecting parameter includes hydraulic force and throttle inlet pressure.

[0059] Understandably, hydraulic force can be the additional force exerted on the valve core by hydraulic oil flowing through the valve orifice due to changes in flow velocity and direction. Hydraulic force always tends to close the valve orifice, meaning its direction of action is opposite to the valve core's opening direction. The magnitude of the hydraulic force is related to the inlet pressure of the throttle orifice and the valve core displacement.

[0060] It is understandable that the throttle inlet pressure refers to the pressure value when hydraulic oil enters the throttle orifice of the valve core, which is usually the high-pressure side pressure. The throttle inlet pressure not only indirectly affects the valve core displacement through hydraulic force, but also directly participates in flow calculation. Under different throttle inlet pressures, even if the valve core displacement is the same, the magnitude of the hydraulic force is different, so the degree of influence on the valve core displacement varies.

[0061] For example, there is a coupling relationship between the third flow-influencing parameter and the second flow-influencing parameter; that is, the magnitude of the hydraulic force depends on the valve core displacement and the inlet pressure of the throttle port, and the hydraulic force, in turn, affects the actual displacement of the valve core. This application introduces the coupling relationship between the third and second flow-influencing parameters through a second mechanism model, which can yield a more accurate target second flow-influencing parameter compared to the second flow-influencing parameter obtained by simply considering the correspondence between the first and second flow-influencing parameters.

[0062] Understandably, the second mechanism model is a mathematical model that describes the complex mapping relationship between the first flow-affecting parameter (such as throttling area), the second flow-affecting parameter (such as valve core displacement), and the third flow-affecting parameter (such as hydraulic force and throttling port inlet pressure). Unlike traditional methods that treat the relationship between each parameter independently, the second mechanism model achieves multi-parameter coupling characteristics by constructing a unified mapping relationship.

[0063] For example, such as Figure 2 As shown, the data model receives, for example, a command current I. x The flow error is taken as input, and the output is a control compensation parameter. This compensation parameter is related to the command current I. xThe final control parameters are obtained by superimposing them, which are used to drive the control valve, thus forming a closed-loop control circuit.

[0064] In practical applications, the data model can be pre-trained in the following way: under different operating conditions, such as different oil temperatures, generator speeds, and load pressures, tests are conducted. For example, a large number of data samples are collected, including test command current, test flow error, and optimal compensation parameters. The test command current and test flow error are used as input features, and the optimal compensation parameters are used as output labels to construct a training sample set. Machine learning algorithms are then used to train the sample set, enabling the model to learn the mapping relationship between input and output. After training, the model is deployed, for example, in a controller for real-time control.

[0065] Specifically, please refer to Figure 3 ,like Figure 3 As shown, Figure 3 This is a schematic diagram of a data model provided in an embodiment of this application, with the first control parameter being the command current I. x For example, data models such as Kalman filtering or neural networks calculate the command current I based on the difference between the input demand flow and the actual flow, i.e., the flow error mentioned above, and the mechanism model. x That is, the first control parameter output by the mechanism model mentioned above is used to output the compensation current ΔI through the mathematical model, which is the control compensation parameter mentioned above.

[0066] The first control parameter refers to the input information acting on the hydraulic system actuator to drive the valve spool to the target position. For example, the first control parameter may include at least one of the following: command current, pilot pressure, and valve spool drive displacement.

[0067] The target flow rate is the actual flow rate output by the hydraulic system based on the first control parameter.

[0068] For example, the command current can refer to the current signal applied to the actuator, such as an electromagnet or a servo valve torque motor. The electromagnetic force generated by the electromagnet is proportional to the magnitude of the current, driving the valve core to move against resistances such as spring force and hydraulic force. For example, there is a preset linear conversion relationship between the command current and the valve core displacement.

[0069] In some embodiments, determining the first control parameters of the hydraulic system based on the target second flow rate influence parameter includes: The target second flow rate influencing parameter is converted into the first control parameter through a preset linear transformation relationship. The preset linear transformation relationship includes a proportional coefficient and a compensation coefficient. Different hydraulic products have different proportional coefficients and / or compensation coefficients.

[0070] Specifically, taking the valve spool displacement as the second flow-affecting parameter and the commanded current as the first control parameter as an example, the valve spool displacement X and the commanded current... The relationship is essentially an electromechanical (linear or rotary) relationship, generally using a linearized model, which simplifies this part to... The function or model, when batch products differ due to processing errors and characteristic parameters, addresses the impact of batch product inconsistencies on system control accuracy by adjusting the t coefficient. is a proportionality coefficient, reflecting the gain characteristics of, for example, an electromagnet valve core system; t is a compensation coefficient, used to calibrate zero-position offset and consistency differences in batch products.

[0071] Specifically, please refer to Figure 4 For the product corresponding to t1, when the valve core displacement X=1, the command current Ix=0, indicating a significant dead zone. For the product corresponding to t2, the command current Ix and the valve core displacement X have a strictly linear relationship, reducing the dead zone. It can be understood that the dead zone refers to the area where the output signal does not respond or change when the input signal varies within a certain range. In hydraulic systems, the dead zone typically manifests as follows: when the command current increases from zero, the valve core does not move immediately, but only begins to move after the current reaches a certain threshold. For example, a compensation coefficient can be used to address the dead zone problem and the inconsistency between batch products.

[0072] Thus, by calibrating the zero-position offset using a compensation coefficient, the relationship curve between the command current and the valve core displacement is shifted, thereby reducing or even eliminating the dead zone and improving control sensitivity under small signal inputs. Furthermore, by setting personalized proportional coefficients and / or compensation coefficients for different hydraulic products, the problem of inconsistent control caused by processing errors, material differences, and other factors in batch production is solved. Even if different products have different electromagnet characteristics, spring stiffness, and friction characteristics, personalized coefficient calibration ensures that each product achieves a consistent control effect, significantly improving the control consistency of batch products. Simultaneously, by reasonably setting the proportional coefficient, a good linear relationship between the command current and the valve core displacement can be maintained at different operating points, simplifying the design of the control strategy and improving the stability and predictability of the control system.

[0073] In summary, in steps S11-S15 above, control compensation parameters are obtained by inputting the first control parameter and the flow error into the pre-trained data model. On the one hand, by inputting the first control parameter obtained from the mechanistic model and the flow error into the data model together, the mechanistic model and the data model are organically integrated. This utilizes the accurate description of the system's physical characteristics by the mechanistic model and leverages the data model's ability to compensate for complex nonlinear factors, thereby improving the control accuracy of the hydraulic system. On the other hand, by using the flow error as one of the inputs to the data model, the generation process of compensation parameters can perceive the magnitude and direction of the current control deviation in real time, forming a closed-loop feedback compensation mechanism that effectively suppresses the impact of external interference and parameter changes on the control effect. In conclusion, by compensating and correcting the first control parameter with the compensation parameter, the accuracy of control can be further improved, thereby increasing the matching degree between the output flow and the actual demand flow, and further enhancing the control performance and reliability of the hydraulic system.

[0074] In some implementations, the method further includes: Under various operating conditions, training sample sets are obtained respectively. The training sample sets include first control parameter samples, flow error samples between demand flow samples and actual flow samples, and corresponding optimal compensation parameter samples. Using the first control parameter sample and the flow error sample as input, and the optimal control compensation parameter as output, the preset machine learning model is trained to obtain the data model.

[0075] Among them, the first control parameter sample refers to the result calculated by the mechanism model based on the demand flow and state parameters under various operating conditions.

[0076] Among them, the demand flow sample and the actual flow sample can be the demand flow input and the actual flow obtained in the test; while the flow error sample is the difference between the same set of demand flow samples and actual flow samples.

[0077] Among them, the optimal compensation parameter sample refers to the compensation value that, under specific operating conditions, enables the error between the demand flow and the actual flow to converge within a preset threshold range.

[0078] Each sample includes a complete set of input-output pairs, with the input being the first control parameter sample and the flow error sample, and the output being the optimal compensation parameter sample; all samples are combined to form a dataset for model training.

[0079] The machine learning model can employ a variety of algorithms, including but not limited to at least one of the following: deep neural network algorithm, convolutional neural network algorithm, bidirectional long short-term memory neural network algorithm, Kalman filter model algorithm, support vector regression algorithm, and random forest algorithm, etc.

[0080] Thus, in this embodiment of the application, by collecting training samples and training the data model under multiple operating conditions, the data model can learn the compensation rules under different operating conditions, thereby improving the accuracy of the model output and providing a favorable foundation for obtaining more accurate second control parameters in the future.

[0081] In some implementations, the operating conditions include at least two operating conditions. Under multiple operating conditions, training sample sets are obtained respectively, including: obtaining corresponding training sample subsets under multiple operating condition values ​​of the same operating condition, wherein the training sample set includes training sample subsets corresponding to at least two operating conditions.

[0082] Thus, this embodiment of the application obtains training sample subsets under different operating conditions, enabling the training sample set to cover combinations of various operating conditions, significantly improving the generalization ability and adaptability of the data model. By collecting multiple operating condition values ​​(such as multiple oil temperature points, multiple speed points, and multiple load pressure points) under a single operating condition, the model can learn the compensation rules when the same factor changes continuously, further improving the model's accuracy and robustness. The training sample set constructed in this way is more comprehensive and balanced, laying a solid foundation for training a high-quality data model.

[0083] In some implementations, at least two operating conditions, including at least two of oil temperature, engine speed, and load pressure, are used to obtain corresponding training sample subsets under multiple operating condition values, including at least one of the following: The first training sample subset was obtained at multiple oil temperature points. A second training sample subset was obtained at multiple engine speed points. A third training sample subset is obtained at multiple load pressure points.

[0084] Thus, by selecting key operating conditions such as oil temperature, engine speed, and load pressure as sample collection dimensions, this embodiment of the application covers the main factors affecting the control accuracy of the hydraulic system, making the training sample set more representative and targeted. By selecting multiple value points under each operating condition (such as multiple oil temperature points from -20℃ to 80℃, multiple speed points from 700rpm to 1600rpm, and multiple pressure points from 30bar to 350bar), the model can learn the compensation rules when various factors change continuously, effectively improving the adaptability and prediction accuracy of the data model in the entire operating range, and providing a reliable data foundation for high-precision hydraulic control.

[0085] In some implementations, obtaining a sample of optimal compensation parameters for each operating condition includes: Under target operating conditions, the test control compensation parameter is adjusted to adjust the test second control parameter so that the percentage error between the test required flow rate and the test actual flow rate converges to a preset threshold range. The test second control parameter has a corresponding relationship with the test actual flow rate. Record the first control parameter of the test under the target working condition, the test flow error between the test required flow and the actual test flow, and the corresponding relationship between the test control compensation parameters; The test control parameters are determined as the optimal compensation parameter sample under the target operating conditions.

[0086] In this context, the test control compensation parameter can be a fixed value in each adjustment, while the test second control parameter is the second control parameter obtained after one adjustment based on a certain foundational second control parameter sample. The test demand flow refers to the corresponding demand flow sample in each adjustment during the training process; similarly, the test actual flow is the corresponding actual flow sample in each adjustment during the training process.

[0087] Thus, the embodiments of this application adaptively adjust the test control compensation parameters so that the second test control parameter automatically approaches the optimal value until the flow error converges to the preset threshold range. This allows for rapid self-calibration of the optimal compensation parameters with only a small number of samples, significantly improving the training efficiency of the data model and the economy of sample collection.

[0088] In some implementations, the preset threshold range is a percentage greater than or equal to a first threshold and less than or equal to a second threshold. The test second control parameter is adjusted by adjusting the test control compensation parameter, including adjusting the test control compensation parameter in a stepwise manner to adjust the test second control parameter. The step-by-step adjustment of test control compensation parameters includes: When the percentage error between the test demand traffic and the test actual traffic is less than the first threshold, the test control compensation parameter is increased by increasing a fixed step value, so as to adjust the test second control parameter to the test first control parameter plus the test compensation parameter. When the percentage error between the test demand traffic and the test actual traffic is greater than the second threshold, the test control compensation parameter is reduced by decreasing the fixed step value, so that the test second control parameter is adjusted to the test first control parameter minus the test compensation parameter. The test control compensation parameter is maintained when the percentage error between the test demand traffic and the actual test traffic is between the first threshold and the second threshold.

[0089] Thus, this embodiment of the application sets a first threshold and a second threshold to form an error allowable range, and adopts a fixed step size method to adaptively adjust the test control compensation parameters, thereby achieving rapid and automatic optimization of the optimal compensation parameters. Without the need for complex optimization algorithms, the test second control parameters can be quickly converged to the ideal range. Through the closed-loop mechanism of step adjustment and threshold judgment, the stability and reliability of the sample collection process are guaranteed, providing high-quality training samples for subsequent data model training.

[0090] In some implementations, obtaining a sample of optimal compensation parameters for each operating condition includes: The optimal compensation parameter sample is calculated based on the second control parameter and the first control parameter corresponding to the percentage convergence of the test traffic error between the test demand traffic and the test actual traffic within a preset range.

[0091] Specifically, taking a two-input single-output model as an example, factors such as oil temperature, load pressure, and engine speed need to be considered when building the data model. The model training steps are as follows: Within the oil temperature range (e1~en) ℃, based on the data model calculation method, data at different oil temperatures, such as -20℃, 0℃, 20℃, 40℃, 60℃, and 80℃, are obtained through bench or real vehicle testing to obtain the first training sample subset. Within the engine speed range (f1~fn) rpm, based on the data model calculation method in Chapter 1, data at different engine speeds, such as 700rpm, 1000rpm, 1300rpm, and 1600rpm, are obtained through bench or real vehicle testing to obtain the second training sample subset. Within the load pressure range of (g1~gn) bar, based on the data model calculation method in Chapter 1, data from different load pressures, such as 30 bar, 60 bar, 90 bar, 120 bar, ..., 350 bar, are used to obtain the third training sample subset. Other factors that interfere with system load, such as oil viscosity and uneven ground, were also tested and data obtained using the same methods described above, and will not be elaborated further.

[0092] Further, please refer to Figure 5 ,like Figure 5 As shown, the data (i.e., the training sample set) obtained in the aforementioned four steps using AMESim software (or other electro-hydraulic simulation software, etc.) is used to manually set an I... y(e.g., 5mA) To minimize the difference between the required flow rate and the actual flow rate for each data set (e.g., the error percentage between the required flow rate and the actual flow rate, ideally ≤5% and ≥-5%), when the error percentage between the required flow rate and the actual flow rate is less than the preset threshold range (the preset threshold range is generally set to -5% to 5%, and can be modified as needed), if the error percentage is less than -5%, the actual current I = I x +I y If the error percentage is greater than 5%, the actual current I = I x -I y ; Calculate and output the compensation current ΔI. The compensation current ΔI = actual current I - command current I x ; Train a two-input, single-output model using MATLAB software (Deep Neural Networks, DNN) or other software model training tools.

[0093] It should be noted that Iy here is a fixed step size value, such as Iy=5mA.

[0094] Thus, this embodiment calculates the optimal compensation parameter sample by using the difference between the second test control parameter and the first test control parameter corresponding to the convergence of the test flow error percentage to a preset range. This gives the process of obtaining the optimal compensation parameter a clear physical meaning and mathematical relationship. This method does not require complex optimization algorithms; it only needs to record the control parameters in the convergence state to directly calculate the optimal compensation amount. The calculation process is simple and efficient. The accuracy of the sample is ensured by the error convergence judgment, providing high-quality labeled data for subsequent data model training, effectively improving the training effect and prediction accuracy of the model.

[0095] In some implementations, a two-layer mapping relationship is constructed based on a first flow rate influence parameter, a second flow rate influence parameter, and a third flow rate influence parameter that affects the second flow rate influence parameter of the hydraulic system, in order to build a mechanistic model, including: Construct a first mapping relationship between the second flow influence parameter and the first flow influence parameter. This first mapping relationship includes M+1 discrete nodes, and each node corresponds to a set of first flow parameter values ​​and second flow parameter values. The influence of the third flow influence parameter on the second flow influence parameter is discretized, and an M×N dimensional grid with the second flow influence parameter and the third flow influence parameter as the dimensions is constructed as the second mapping relationship, where M is the number of intervals formed by the nodes of the second flow influence parameter, and N is the number of intervals of the third flow influence parameter corresponding to the second flow influence parameter. Based on the first and second mapping relationships, a second mechanism model is constructed.

[0096] For example, taking the valve spool displacement as the second flow-influencing parameter and the throttling area as the first flow-influencing parameter. It should be noted that in hydraulic systems, the relationship between valve spool displacement and throttling area exhibits a complex nonlinear relationship. Understandably, this nonlinear relationship can be determined by the valve port geometry; different types of hydraulic valves and different port designs (such as full-circumference opening, non-full-circumference opening, U-groove, and V-groove) will all lead to differences in the relationship curve between valve spool displacement and throttling area.

[0097] In some implementations, constructing a first mapping relationship between the second flow influence parameter and the first flow influence parameter includes: obtaining M+1 discrete points in the correspondence between the second flow influence parameter and the first flow influence parameter, with each discrete point recording a set of first flow influence parameter values ​​and second flow influence parameter values; establishing relationship curves between adjacent nodes of the M+1 discrete points using piecewise interpolation or spline interpolation to form the first mapping relationship, which is used to obtain the theoretical second flow influence parameter based on the current first flow influence parameter.

[0098] For example, the nonlinear relationship between the throttling area Ac and the valve core displacement Xz can be constructed by using piecewise interpolation and spline curves to establish the relationship Xz = f(Ac).

[0099] Taking piecewise interpolation as an example, the relationship is shown in the following table (1): (1) As shown in the table above, the points are divided into m discrete points, therefore the valve core displacement X z The subscripts are from 1 to m, such as the m-th node, where the flow area is A. cm The valve core displacement is X zm .

[0101] Thus, by acquiring M+1 discrete points and constructing the first mapping relationship using piecewise interpolation or spline interpolation methods, an accurate description of the nonlinear characteristics between, for example, valve core displacement and throttling area is achieved. This significantly improves the accuracy of the mapping relationship while ensuring computational efficiency. Furthermore, by discretizing complex nonlinear relationships and converting them into a lookup table interpolation form that is easy to implement in engineering, it is convenient for real-time calculation in embedded controllers and has good engineering practical value.

[0102] In some implementations, the influence of the third flow rate influence parameter on the second flow rate influence parameter is discretized, and an M×N dimensional grid with the second and third flow rate influence parameters as dimensions is constructed as a second mapping relationship, including: The second flow rate influence parameter is discretized into M+1 boundary points, forming M continuous intervals; The third flow influence parameter is discretized into N+1 boundary points, forming N continuous intervals; For each set of second flow influence parameter intervals and third flow influence parameter intervals, determine the corresponding second flow influence parameter compensation amount; An M×N dimensional grid is constructed, and each grid node stores the compensation amount of the second flow impact parameter under the corresponding interval combination, forming the second mapping relationship. The compensation amount of the second flow impact parameter is used to correct the theoretical second flow impact parameter obtained from the first mapping relationship, so as to obtain the actual second flow impact parameter as the second target flow impact parameter.

[0103] For example, the influence of the third flow influence parameter on the second flow influence parameter is discretized, and an M×N dimensional grid with the second flow influence parameter and the third flow influence parameter as dimensions is constructed as the second mapping relationship, including: The second flow rate influence parameter is discretized into M+1 boundary points d1, d2, ..., d... M , d M+1 And form M continuous intervals (d1~d2), (d2~d3), ..., (d M ~d M+1 ); The third flow rate influence parameter is discretized into N+1 boundary points c1, c2, ..., c N , c N+1 And form N consecutive intervals (c1~c2), (c2~c3), ..., (c M ~c M+1 ); For each set of second flow influence parameter interval i and third flow influence parameter interval j, determine the corresponding second flow influence parameter compensation amount Xkij, where i is an integer greater than or equal to 1 and less than or equal to M, and j is an integer greater than or equal to 1 and less than or equal to N. An M×N dimensional grid is constructed, and each grid node stores the compensation amount of the second flow impact parameter under the corresponding interval combination, forming the second mapping relationship. The compensation amount of the second flow impact parameter is used to correct the theoretical second flow impact parameter obtained from the first mapping relationship, so as to obtain the actual second flow impact parameter as the target second flow impact parameter.

[0104] For example, the third flow rate influence parameter is the hydraulic force F. k With the inlet pressure P of the throttle orifice F Taking the high-pressure port pressure as an example, and using the second flow rate influence parameter as the valve core displacement X... z For example, hydraulic F k With the inlet pressure P of the throttle orifice F Valve core displacement Xz (or control current) They are related and can be expressed through functional relationships such as F. k =f(P) F X z ) or F k =f(P) F , This indicates that the compensation amount for the second flow rate influence parameter can be calculated through the above functional relationship.

[0105] However, in practical applications, hydraulic force always tends to close the valve orifice. Without considering the influence of hydraulic force, the inlet pressure P of the throttle orifice... F The magnitude of the valve core displacement is affected by X. k This can be understood as the compensation amount for the second flow impact mentioned above, X. k With the inlet pressure P of the throttle orifice F and valve core displacement X z The relationship between them can be expressed by the functional relation X. k =g(P) F X z It should be noted that the throttle inlet pressure represents the influence of hydraulic force, which is a change within a hydraulic valve. Based on the correlation between hydraulic force and throttle inlet pressure, the throttle inlet pressure is used, for example, to indirectly represent the hydraulic force.

[0106] For example, X k The relationship between the throttle inlet pressure PF and the valve core displacement Xz can also be represented by constructing a table (2) to characterize the influence of hydraulic force on the valve core displacement X. k : (2) As shown in the table above, the inlet pressure P of the throttle orifice... F Divide into n discrete intervals, then let the valve core displacement be X. zm For example, the inlet pressure P of the throttle orifice F The effects on valve core displacement are as follows: , , ..., .

[0108] To facilitate understanding, the relationship between the first flow rate influence parameter and the second flow rate influence parameter, namely the valve core throttling area and the valve core displacement, can be illustrated by, for example... Figure 6 As shown, when the valve core displacement is at nodes [1, m+1], the relationship between the actual area A of the valve core and the actual displacement X of the valve core satisfies X [1,m+1] =X z[1,m+1] - That is, based on the current first flow influence parameter, the theoretical second flow influence parameter, which does not take into account the third flow influence parameter, is first calculated through the first mapping relationship. Then, the compensation amount of the second flow influence parameter is obtained according to the second mapping relationship. Finally, the theoretical second flow influence parameter is corrected based on the second flow compensation amount to obtain the actual second flow influence parameter, i.e., the target second flow influence parameter.

[0109] In some implementations, based on the current first flow rate impact parameter, a target second flow rate impact parameter is output through a second mechanism model, including: Based on the current first flow impact parameter, obtain the corresponding theoretical second flow impact parameter from the first mapping relationship; Based on the intervals of the theoretical second flow influence parameter and the third flow influence parameter, the corresponding compensation amount of the second flow influence parameter is queried from the second mapping relationship; The target second flow influence parameter is calculated by superimposing the theoretical second flow influence parameter with the compensation amount of the second flow influence parameter.

[0110] For example, based on the current first flow impact parameter, the corresponding theoretical second flow impact parameter can be queried from the first mapping relationship. This can be done by querying the corresponding second flow impact parameter through the lookup table (1) based on the current first flow impact parameter, or by using a functional relationship such as X. z = f(A) c The second flow rate influence parameter was calculated.

[0111] For example, based on the intervals where the theoretical second flow influence parameter and the third flow influence parameter are located, the corresponding compensation amount for the second flow influence parameter can be obtained from the second mapping relationship. This can be done by querying the corresponding compensation amount for the second flow influence parameter through lookup table (2) based on the intervals where the theoretical second flow influence parameter and the third flow influence parameter are located, or by using the functional relationship X. k =g(P) F X z The compensation amount of the second flow influence parameter is calculated to correct the theoretical second flow influence parameter, thereby obtaining the actual second flow influence parameter, which is the target second flow influence parameter mentioned above.

[0112] Thus, by first querying the theoretical second flow rate influencing parameters, then querying the compensation amount, and finally superimposing the calculations, a precise description of the multi-parameter coupling characteristics of the hydraulic system is achieved. By using two equivalent methods, table lookup or function calculation, to obtain the theoretical values ​​and compensation amounts, both the flexibility and adaptability of the calculation are ensured, and multiple implementation paths are provided for different application scenarios. This method transforms complex nonlinear coupling relationships into simple table lookup or function calculation, which significantly reduces the computational burden of real-time control while ensuring calculation accuracy, and is particularly suitable for resource-constrained environments such as embedded controllers.

[0113] In some implementations, the hydraulic system is controlled to output a corresponding target flow rate according to a first control parameter, including: Obtain control compensation parameters; The second control parameter is determined based on the first control parameter and the control compensation parameter; Based on the second control parameter, the target flow rate corresponding to the control hydraulic coefficient is output.

[0114] Among them, the control compensation parameter can refer to the additional input quantity used to correct the first control parameter.

[0115] The control compensation parameters may include, but are not limited to, at least one of the following: Compensation current ΔI: When the first control parameter is the command current, the control compensation parameter is the compensation current; Compensation pressure ΔP: When the first control parameter is the pilot pressure, the control compensation parameter is the compensation pressure; Compensation displacement ΔX: When the first control parameter is the valve core drive displacement, the control compensation parameter is the compensation displacement.

[0116] For example, the source of the control compensation parameter may include at least one of the following: The output of the data model is calculated based on the current operating conditions and flow error using a pre-trained data model, such as a neural network or Kalman filter. Obtained by looking up a table, for example, by looking up the compensation parameters from a pre-built table based on the current working conditions; Function calculations, such as substituting the current operating parameters into the compensation function for calculation.

[0117] Thus, in this embodiment of the application, by controlling the compensation of the compensation parameters, a more accurate second control parameter can be obtained, so as to output a flow rate that matches the actual needs more accurately, thereby further improving the control performance and control reliability of the hydraulic system.

[0118] It should be noted that the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0119] In one specific implementation, taking the training of the residual model using a neural network or Kalman filter algorithm as an example, please refer to [link to relevant documentation]. Figure 7 ,like Figure 7 As shown, the trained data model is superimposed on the mechanistic model. The mechanistic model (i.e., the combination of the first and second mechanistic models mentioned above) outputs, for example, a command current. This command current is then input into the mathematical model. The mathematical model compensates for this command current with a compensation current output based on the current command current, resulting in the compensated command current, which is the second control parameter. This compensated current is then used to control the actual flow output of the control valve. The actual flow can also be used as the input for the error flow between the demand flow and the actual flow, feeding it into the data model. This forms a continuously correcting closed-loop control system, which improves the control performance and reliability of the hydraulic system, making the actual flow increasingly closer to the demand flow.

[0120] In one specific implementation, the method is used as an example in the application of a crane; please refer to [link to relevant documentation]. Figure 8 ,like Figure 8 As shown, taking the crane's uniform angular velocity variable luffing lowering operation as an example, the mechanism model in the figure can be either the first mechanism model or the second mechanism model mentioned above, i.e., constructed based on the thin-walled orifice flow formula and the double-layer meshing method, used to calculate the first control parameter according to the demand flow rate and state parameters; the data model can adopt a residual compensation model, using the first parameter and flow error as input, and outputting control compensation parameters, such as compensation current ΔI, to correct the output of the mechanism model. The output of the fusion model is the final control parameter, i.e., the second control parameter, which serves as the input for subsequent control.

[0121] The PI controller, located after the fusion model, receives the final control parameters output by the fusion model and performs proportional-integral control processing on them. The functions of the PI controller include: proportional control (amplifying or reducing the current error to improve response speed); integral control (eliminating steady-state error to improve control accuracy); and output smoothing (smoothing the final control parameters output by the fusion model to reduce the impact of sudden changes in control input on the system). The output of the PI controller is a regulated control signal that can be directly applied to the amplitude transformer.

[0122] The luffing mechanism is an actuator of the crane used to achieve the pitching movement of the boom. During the luffing lowering operation, the luffing mechanism receives control signals from the PI controller, driving the hydraulic cylinders or hydraulic motors to smoothly lower the boom at the desired angular velocity. The motion state of the luffing mechanism directly affects the crane's operational safety and operator comfort, requiring high control precision and response speed.

[0123] exist Figure 8In the feedback path, this represents the method of obtaining the actual traffic. For example, since directly measuring traffic is costly, an indirect calculation method is used to obtain the actual traffic: The piston displacement of the luffing cylinder is measured in real time using a displacement sensor. Differentiating the displacement signal yields the piston velocity v. Calculate the actual flow rate based on the effective area A of the hydraulic cylinder: Qactual = v×A.

[0124] The calculated actual flow rate is used, on the one hand, to compare with the demand flow rate to form a flow rate error, which is then fed back to the data model for compensation; on the other hand, it serves as a basis for evaluating the control effect.

[0125] Thus, this embodiment of the application applies the fusion method of mechanism model and data model to the actual crane luffing control, and uses it in series with a PI controller. This leverages the advantages of the fusion model in nonlinear compensation and the strengths of the PI controller in steady-state error elimination and smooth control, forming a complementary control architecture. At the same time, the method of indirectly calculating the actual flow rate by cylinder displacement can achieve flow feedback without adding sensors, reducing system costs and improving engineering practicality. Finally, closed-loop control is formed through actual flow feedback, enabling the system to perceive the control effect in real time and make dynamic adjustments, improving the accuracy and stability of luffing control, and achieving smooth lowering of the crane boom at a uniform angular velocity.

[0126] To achieve the above objectives, embodiments of this application also provide a control device for a hydraulic system, which is applied to computer equipment or engineering machinery. Please refer to [link to relevant documentation]. Figure 9 The device includes: The first acquisition module 91 is used to acquire the required flow rate and the actual flow rate of the hydraulic system; The second acquisition module 92 is used to obtain the first control parameter based on the demand flow rate and the mechanism model. The third acquisition module 93 is used to acquire the flow error between the required flow and the actual flow; Input module 94 is used to input the first control parameter and flow error into the data model to obtain control compensation parameters; The determining module 95 is used to determine the second control parameter based on the first control parameter and the control compensation parameter, and to use the second control parameter to control the hydraulic system to output the corresponding target flow.

[0127] In some embodiments, the apparatus further includes: The fourth acquisition module is used to acquire training sample sets under various different operating conditions. The training sample sets include first control parameter samples, flow error samples between demand flow samples and actual flow samples, and corresponding optimal control compensation parameter samples. The training module is used to train a preset machine learning model with the first control parameter sample and the flow error sample as input and the optimal control compensation parameter as output, so as to obtain a data model.

[0128] In some implementations, the operating conditions include at least two operating conditions, and the fourth acquisition module is further used for: Under multiple operating conditions with the same operating conditions, corresponding training sample subsets are obtained respectively. The training sample set includes training sample subsets corresponding to at least two operating conditions.

[0129] In some implementations, at least two operating conditions are included, including at least two of oil temperature, engine speed and load pressure. The fourth acquisition module is also used to acquire at least one of the following: acquire a first training sample subset at multiple oil temperature points; acquire a second training sample subset at multiple engine speed points; and acquire a third training sample subset at multiple load pressure points.

[0130] In some implementations, the fourth acquisition module is also used for: Under target operating conditions, the test control compensation parameter is adjusted to adjust the test second control parameter so that the percentage error between the test demand flow and the test actual flow converges to within the preset threshold range, and the test second control parameter has a corresponding relationship with the test actual flow. Record the first control parameter of the test under the target working condition, the test flow error between the test required flow and the actual test flow, and the corresponding relationship of the test control compensation parameter; The test control parameters are determined as the optimal compensation parameter sample under the target operating conditions.

[0131] In some implementations, the preset threshold range is a percentage greater than or equal to the first threshold and less than or equal to the second threshold. The fourth acquisition module is also used to: adjust the test second control parameter by adjusting the test control compensation parameter in a stepwise manner. The step-by-step adjustment of test control compensation parameters includes: When the percentage error between the test demand traffic and the test actual traffic is less than the first threshold, the test control compensation parameter is increased by increasing the fixed step value, so that the test second control parameter is adjusted to the test first control parameter plus the test compensation parameter. When the percentage error between the test demand traffic and the test actual traffic is greater than the second threshold, the test control compensation parameter is reduced by decreasing the fixed step value, so that the test second control parameter is adjusted to the test first control parameter minus the test compensation parameter. When the percentage error between the test demand traffic and the actual test traffic is between the first threshold and the second threshold, the test control compensation parameters are maintained.

[0132] In some implementations, the fourth acquisition module is further configured to: The optimal compensation parameter sample is calculated based on the second test control parameter corresponding to the percentage convergence of the test traffic error between the test demand traffic and the actual test traffic within a preset threshold range, and the first test control parameter.

[0133] In some embodiments, the second acquisition module 92 is further configured to: Based on the demand flow, the parameters affecting the current first flow are determined using the first mechanism model; Based on the current first flow rate impact parameters, the target second flow rate impact parameters are determined using the second mechanism model. Based on the target second flow rate influence parameter, determine the first control parameter of the hydraulic system.

[0134] It should be noted that the description of the control and processing device for the hydraulic system above is similar to the description of the control and processing method for the hydraulic system above, and the beneficial effects of the same method will not be repeated. For technical details not disclosed in the embodiments of the control and processing device for the hydraulic system in this application, please refer to the description of the embodiments of the control and processing method for the hydraulic system in this application.

[0135] To achieve the above objectives, embodiments of this application also provide an engineering machinery that integrates a hydraulic system controller, i.e., the execution body of the above-described method, and may include the aforementioned hydraulic system control processing device. In some embodiments, the engineering machinery may include: a memory configured to store instructions; and a processor configured to retrieve instructions from the memory and, when executing the instructions, to implement the aforementioned hydraulic system control processing method. In examples, the engineering machinery may, for instance, be a control processing device integrating the aforementioned hydraulic system.

[0136] Based on the same application concept as the foregoing embodiments, this application provides a computer device, such as... Figure 10 As shown, the computer device includes: a processor 610 and a memory 611 storing a computer program; wherein, Figure 10 The processor 610 shown in the diagram does not indicate that there is only one processor 610, but only indicates the positional relationship of the processor 610 relative to other devices. In practical applications, there can be one or more processors 610; similarly, Figure 10The memory 611 shown in the diagram has the same meaning, that is, it is only used to indicate the positional relationship of memory 611 relative to other devices. In practical applications, there can be one or more memories 611. When the processor 610 runs the computer program, it implements the control processing method of the hydraulic system described above.

[0137] The computer device may also include at least one network interface 612. Various components of the computer device are coupled together via a bus system 613. It is understood that the bus system 613 is used to implement communication between these components. In addition to a data bus, the bus system 613 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 10 The general designated all buses as Bus System 613.

[0138] Based on the same application concept as the foregoing embodiments, this embodiment also provides a computer-readable storage medium storing a computer program. The computer-readable storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it can also be various devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc. When the computer program stored in the computer-readable storage medium is run by a processor, it implements the above-described process solution generation method. For the specific steps implemented when the computer program is executed by the processor, please refer to [link to relevant documentation]. Figure 1 The description of the illustrated embodiments will not be repeated here.

[0139] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control and processing method for a hydraulic system, characterized in that, include: Obtain the required flow rate and actual flow rate of the hydraulic system; Based on the demand flow rate, the first control parameter is obtained using the mechanism model; Obtain the traffic error between the required traffic and the actual traffic; The first control parameter and the flow error are input into the data model to obtain the control compensation parameter; Based on the first control parameter and the control compensation parameter, a second control parameter is determined, and the second control parameter is used to control the hydraulic system to output the corresponding target flow rate.

2. The method according to claim 1, characterized in that, The method further includes: Under various operating conditions, training sample sets are obtained respectively. The training sample sets include first control parameter samples, flow error samples between demand flow samples and actual flow samples, and corresponding optimal control compensation parameter samples. Using the first control parameter sample and the flow error sample as input, and the optimal control compensation parameter as output, a preset machine learning model is trained to obtain the data model.

3. The method according to claim 2, characterized in that, The operating conditions include at least two operating conditions, and the acquisition of training sample sets under multiple different operating conditions includes: Under multiple operating conditions with the same operating conditions, corresponding training sample subsets are obtained respectively. The training sample subsets include training sample subsets corresponding to at least two operating conditions.

4. The method according to claim 3, characterized in that, The at least two operating conditions include at least two of oil temperature, engine speed, and load pressure. Obtaining corresponding training sample subsets under multiple operating conditions of the same type includes at least one of the following: The first training sample subset was obtained at multiple oil temperature points. A second training sample subset was obtained at multiple engine speed points. A third training sample subset is obtained at multiple load pressure points.

5. The method according to claim 2, characterized in that, Obtain the optimal compensation parameter sample for each operating condition, including: Under target operating conditions, the test control compensation parameter is adjusted to adjust the test second control parameter so that the percentage error between the test required flow rate and the test actual flow rate converges to a preset threshold range. The test second control parameter has a corresponding relationship with the test actual flow rate. Record the first control parameter of the test under the target working condition, the test flow error between the test required flow and the actual test flow, and the corresponding relationship between the test control compensation parameters; The test control parameters are determined as the optimal compensation parameter sample under the target operating conditions.

6. The method according to claim 5, characterized in that, The preset threshold range is a percentage greater than or equal to the first threshold and less than or equal to the second threshold. The step of adjusting the test control compensation parameter by adjusting the test control compensation parameter includes: adjusting the test control parameter in a step manner to adjust the test control parameter. The step-by-step adjustment of the test control compensation parameters includes: When the percentage error between the test demand flow and the test actual flow is less than the first threshold, the test control compensation parameter is increased by increasing a fixed step value, so that the test second control parameter is adjusted to the test first control parameter plus the test compensation parameter. When the percentage error between the test demand flow and the test actual flow is greater than the second threshold, the test control compensation parameter is reduced by decreasing the fixed step value, so that the test second control parameter is adjusted to the test first control parameter minus the test compensation parameter. The test control compensation parameter is maintained when the percentage error between the test demand traffic and the actual test traffic is between the first threshold and the second threshold.

7. The method according to claim 5, characterized in that, Obtain the optimal compensation parameter sample for each operating condition, including: The optimal compensation parameter sample is calculated based on the second test control parameter corresponding to the percentage convergence of the test traffic error between the test demand traffic and the actual test traffic within a preset threshold range, and the first test control parameter.

8. The method according to claim 1, characterized in that, The mechanism model includes a first mechanism model and a second mechanism model. Obtaining the first control parameter based on the demand flow using the mechanism model includes: Based on the demand flow, the current first flow impact parameters are determined through the first mechanism model; Based on the current first flow rate impact parameter, the target second flow rate impact parameter is determined through the second mechanism model; The first control parameter of the hydraulic system is determined based on the target second flow rate influence parameter.

9. A computer device, characterized in that, include: The processor and the memory storing a computer program implement the control processing method of the hydraulic system according to any one of claims 1 to 8 when the processor runs the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the control processing method of the hydraulic system according to any one of claims 1 to 8.

Images