Constant tension control method for hydraulic pushing mechanism

Abstract

The application relates to the technical field of hydraulic servo control, and discloses a constant tension control method of a hydraulic pushing mechanism, which comprises the following steps: collecting a steady-state pressure signal output by a steady-state pressure sensor, a first dynamic pressure signal output by a first dynamic pressure sensor and a second dynamic pressure signal output by a second dynamic pressure sensor in a hydraulic constant tension control system of the hydraulic pushing mechanism in real time; executing low-frequency steady-state pushing force control and high-frequency dynamic suppression control in parallel; combining disturbance source discrimination and state prediction; and synthesizing a high-frequency compensation signal to drive a high-frequency actuator. Through disturbance source discrimination and state prediction, the application realizes targeted suppression and predictive compensation of disturbances from different sources, effectively suppresses load nonlinearity and high-frequency disturbance of a hydraulic source, improves the precision and dynamic stability of constant tension control, and is suitable for high-temperature industrial pushing equipment.

Description

Technical Field

[0001] This invention relates to the field of hydraulic servo control technology, specifically a constant tension control method for hydraulic pushing mechanisms, applicable to industrial scenarios such as crucible pushing in graphitization furnaces and tension control in coiling machines. Background Technology

[0002] In the production of carbon materials, graphitization furnaces are key process equipment. These machines use a hydraulic pushing mechanism to continuously and stably push a string of crucibles loaded with the material to be processed through a high-temperature heating zone to complete the heat treatment process. To ensure the uniformity and stability of the heating process and to avoid mechanical damage to the crucibles and furnace structure caused by excessive or uneven pushing force, the pushing force provided by the hydraulic pushing mechanism must be kept precisely constant, which is constant tension control.

[0003] Currently, the control of such hydraulic pushing mechanisms mostly adopts a single pressure closed-loop control. A pressure sensor monitors the working pressure of the hydraulic cylinder, and the controller adjusts the opening of the main control valve according to the deviation between the working pressure of the hydraulic cylinder and the target set value, as described in patent CN110107725A, through a proportional-integral-derivative algorithm.

[0004] However, in actual operation, this control method has inherent limitations. The friction between the crucible string and the furnace bottom track has highly nonlinear characteristics, with static friction being much greater than dynamic friction. This makes the system prone to "stick-slip" phenomena. The hydraulic cylinder thrust gradually accumulates in the process of overcoming static friction. Once the crucible string begins to move, the required thrust drops instantaneously due to the sudden change from friction to dynamic friction. Conventional controllers and main control valve systems do not have sufficient response bandwidth to quickly adapt to this pressure change caused by load characteristics, resulting in impact and axial movement of the crucible string.

[0005] In addition, the power source of the hydraulic system itself, such as the hydraulic pump, will also generate high-frequency pressure pulsations in the pipeline. These disturbances from the hydraulic source are directly transmitted to the actuator, which also affects the stability of the thrust. The high-flow main control valve used to regulate the system's base pressure cannot suppress such high-frequency disturbances in its dynamic response performance.

[0006] Therefore, existing single-loop control strategies cannot simultaneously ensure the stability of the base thrust while suppressing transient impacts of stick-slip phenomena and attenuating high-frequency pulsations of the hydraulic source when dealing with complex disturbances from different sources and with varying frequency bands. This makes it difficult to achieve high-precision and high-stability constant tension control in complex disturbance environments, thereby affecting the stability of the entire production process and the quality of the final product. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a constant tension control method for a hydraulic pushing mechanism. By combining parallel low-frequency steady-state control and high-frequency dynamic suppression with disturbance source identification and state prediction, it achieves targeted suppression and predictive compensation for multi-source disturbances, improving control accuracy and dynamic stability. This solves the problem that existing single control strategies cannot simultaneously and effectively suppress pressure surges caused by load nonlinearity and high-frequency pulsations of the hydraulic source itself in complex disturbance environments, leading to insufficient system control accuracy and dynamic stability.

[0008] To solve the above-mentioned technical problems, the present invention provides a constant tension control method for a hydraulic pushing mechanism, comprising the following steps: S1. Signal Acquisition: Real-time acquisition of the steady-state pressure signal output by the steady-state pressure sensor, the first dynamic pressure signal output by the first dynamic pressure sensor, and the second dynamic pressure signal output by the second dynamic pressure sensor in the hydraulic constant tension control system of the hydraulic push mechanism. S2. Parallel control calculation: Parallel execution of low-frequency steady-state thrust control and high-frequency dynamic suppression control; Low-frequency steady-state thrust control: The control signal of the main actuator of the hydraulic constant tension control system is calculated based on the steady-state pressure signal and the preset target pressure. High-frequency dynamic suppression control includes the following steps: Disturbance source identification: Based on the timing relationship between the first dynamic pressure signal and the second dynamic pressure signal, identify the source of pressure fluctuation and output the disturbance source flag bit; State prediction: Based on the steady-state pressure signal and the rate of change of the steady-state pressure signal, predict the slippage event of the hydraulic cylinder of the hydraulic pushing mechanism and output the prediction trigger signal; High-frequency compensation control law synthesis: Based on the state of the disturbance source flag and the predicted trigger signal, a high-frequency actuator control signal is synthesized.

[0009] S3, Composite Drive Execution: The main actuator is driven by the main actuator control signal, and the high-frequency actuator of the hydraulic constant tension control system is driven by the high-frequency actuator control signal. S4. Cyclic Control: Steps S1 to S3 are executed cyclically according to the set control cycle, which is used for the composite closed-loop control of the hydraulic cylinder thrust of the hydraulic pushing mechanism.

[0010] In the core technical solution of this invention, the high-frequency dynamic suppression control achieves the suppression of transient disturbances through the following principle: First, the first and second dynamic pressure signals, which are respectively located on the load side and hydraulic source side of the pipeline, are subjected to high-pass filtering to extract the high-frequency dynamic pressure components.

[0011] Subsequently, the cross-correlation function between the two high-frequency dynamic pressure components is calculated to measure the similarity between the two high-frequency dynamic pressure components under different time delays. By determining the time delay corresponding to the peak value of the cross-correlation function, the propagation direction of the pressure wave can be determined. Positive or negative time delays correspond to the disturbance propagating from the hydraulic source side or from the load side, respectively. By comparing the time delay with a preset dead zone threshold, a flag indicating the source of the disturbance can be output, providing a decision basis for subsequent adaptive control.

[0012] The disturbance source identification step includes: performing high-pass filtering on the first dynamic pressure signal and the second dynamic pressure signal to obtain high-frequency dynamic pressure components; calculating the cross-correlation function between the high-frequency dynamic pressure components and determining the peak delay of the cross-correlation function; comparing the sign of the peak delay with a preset time delay dead zone threshold, outputting the disturbance source flag bit, and identifying the disturbance as originating from either the load side or the hydraulic source side.

[0013] The low-frequency steady-state thrust control steps include: calculating the steady-state error between the target pressure and the steady-state pressure signal, using a proportional-integral-derivative control law, and calculating the main actuator control signal based on the steady-state error. The proportional-integral-derivative control law includes preset proportional gain, integral gain, and derivative gain.

[0014] In order to overcome the impact of slippage events caused by the difference between static and dynamic friction forces under the load on system stability, the occurrence of slippage can be predicted in advance by monitoring the changing characteristics of steady-state pressure signals.

[0015] First, the rate of change of the steady-state pressure signal is calculated. Second, by setting combined conditions, it is determined whether the system has entered a viscous state of stable pressure accumulation. When the rate of change of pressure is continuously greater than a positive threshold within a certain time window, and the change process is smooth and the variance is less than another threshold, the system is determined to be in a viscous state.

[0016] Finally, under the condition that the system is in a viscous state, if the amplitude of the steady-state pressure reaches or exceeds the preset static friction threshold, it is determined that slippage is about to occur, and a predictive trigger signal is immediately generated. This method enables the controller to change from passive response to active prevention, and intervene before the disturbance occurs.

[0017] Based on the above results of disturbance source identification and state prediction, the final control signal output to the high-frequency actuator is composed of three independent control components superimposed: load disturbance suppression signal, source disturbance suppression signal, and prediction feedforward compensation signal.

[0018] The state prediction step includes: calculating the pressure change rate of the steady-state pressure signal; when the pressure change rate is continuously greater than a first positive threshold within a time window and the variance of the pressure change rate is less than a second threshold, determining that the system is in a viscous state; when the system is in a viscous state and the steady-state pressure signal reaches or exceeds a static friction threshold, setting the prediction trigger signal to a trigger state.

[0019] In the high-frequency compensation control law synthesis step, the high-frequency actuator control signal is composed of the superposition of the load disturbance suppression signal, the source disturbance suppression signal, and the prediction feedforward compensation signal.

[0020] The load disturbance suppression signal is calculated based on the reference disturbance signal and the first dynamic compensator. The reference disturbance signal is the high-frequency component of the first dynamic pressure signal. The first dynamic compensator is used to compensate for the dynamic hysteresis of the high-frequency actuator and to counteract it.

[0021] Each component is selectively activated based on the results of discrimination and prediction: Load disturbance suppression signal: It is activated when the flag indicates that the disturbance originates from the load side. The load disturbance suppression signal is calculated based on the pressure disturbance component on the load side and the first dynamic compensator. The first dynamic compensator is used to compensate for the dynamic hysteresis of the high-frequency actuator, so that the first dynamic compensator can generate a pressure wave with the opposite phase to the disturbance wave to achieve cancellation.

[0022] The high-frequency compensation control law synthesis step includes: activating the load disturbance suppression signal when the disturbance source flag indicates that the disturbance originates from the load side; activating the source disturbance suppression signal when the disturbance source flag indicates that the disturbance originates from the hydraulic source side; and activating the prediction feedforward compensation signal when the prediction trigger signal generates a rising edge.

[0023] Source disturbance suppression signal: It is activated when the flag indicates that the disturbance originates from the hydraulic source side. The source disturbance suppression signal is calculated based on the pressure disturbance component at the source and the second dynamic compensator. The second dynamic compensator can be designed as a notch filter for specific frequencies such as hydraulic pump pulsation, and is used to actively absorb disturbances from the source.

[0024] The source disturbance suppression signal is calculated based on the reference disturbance signal and the second dynamic compensator, which is designed to actively absorb pressure pulsations from the hydraulic source.

[0025] The predictive feedforward compensation signal is a preset pressure relief pulse waveform, which has a pre-calibrated amplitude, shape, and duration.

[0026] Predictive feedforward compensation signal: Activated when the predictive trigger signal is set to the trigger state, the predictive feedforward compensation signal is a pressure relief pulse waveform pre-calibrated according to the experiment, used to drive the high-frequency actuator to generate an instantaneous pressure relief action, offsetting the pressure drop that will be caused by slippage in advance.

[0027] The peak delay search is performed within a preset delay range, which is determined based on the distance between the first dynamic pressure sensor and the second dynamic pressure sensor and the speed of sound of the pressure wave propagating in the pipeline.

[0028] This invention provides a constant tension control method for a hydraulic pushing mechanism. It has the following beneficial effects: 1. This invention uses a first dynamic pressure sensor and a second dynamic pressure sensor at different locations, combined with cross-correlation function analysis, to determine online whether the source of pressure disturbance is the load side or the hydraulic source side. Based on this determination result, the corresponding control components can be selectively activated for targeted suppression. This method makes the control action more targeted, improves the efficiency and accuracy of disturbance suppression, and solves the problem that in a complex disturbance environment, the existing single control strategy cannot simultaneously and effectively suppress the pressure change caused by load nonlinearity and the high-frequency pulsation of the hydraulic source itself, thus leading to insufficient system control accuracy and dynamic stability.

[0029] 2. By monitoring the rate of change and fluctuation characteristics of steady-state pressure, this invention can predict slip events caused by nonlinear friction. Before the event occurs, the system actively triggers a preset feedforward compensation signal to offset the impending pressure drop. This predictive compensation mechanism advances the control timing from post-event response to pre-event intervention, thereby effectively avoiding the severe impact on system stability during the stick-slip transition process.

[0030] 3. The present invention adopts a composite control architecture, which combines low-frequency control responsible for steady-state tracking with high-frequency control responsible for transient suppression. The high-frequency control further integrates disturbance source discrimination and state prediction functions, which can adaptively respond to disturbances of different sources and natures, and achieve comprehensive suppression of wide-bandwidth, multi-source disturbances. Ultimately, it improves the overall accuracy and dynamic robustness of constant tension control of hydraulic push mechanism. Attached Figure Description

[0031] Figure 1 This is the overall flowchart of the present invention; Figure 2 This is a flowchart of the high-frequency dynamic suppression control of the present invention; Figure 3 This is a flowchart of the low-frequency steady-state thrust control of the present invention. Detailed Implementation

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Please refer to the appendix. Figure 1 This invention provides a constant tension control method for a hydraulic pushing mechanism, comprising the following steps: S1. Signal Acquisition: Real-time acquisition of the steady-state pressure signal output by the steady-state pressure sensor, the first dynamic pressure signal output by the first dynamic pressure sensor, and the second dynamic pressure signal output by the second dynamic pressure sensor in the hydraulic constant tension control system of the hydraulic push mechanism. S2. Parallel control calculation: Parallel execution of low-frequency steady-state thrust control and high-frequency dynamic suppression control; Low-frequency steady-state thrust control: The control signal of the main actuator of the hydraulic constant tension control system is calculated based on the steady-state pressure signal and the preset target pressure. High-frequency dynamic suppression control includes the following steps: Disturbance source identification: Based on the timing relationship between the first dynamic pressure signal and the second dynamic pressure signal, identify the source of pressure fluctuation and output the disturbance source flag bit; State prediction: Based on the steady-state pressure signal and the rate of change of the steady-state pressure signal, predict the slippage event of the hydraulic cylinder of the hydraulic pushing mechanism and output the prediction trigger signal; High-frequency compensation control law synthesis: Based on the state of the disturbance source flag and the predicted trigger signal, a high-frequency actuator control signal is synthesized.

[0034] S3, Composite Drive Execution: The main actuator is driven by the main actuator control signal, and the high-frequency actuator of the hydraulic constant tension control system is driven by the high-frequency actuator control signal. S4. Cyclic Control: Steps S1 to S3 are executed cyclically according to the set control cycle, which is used for the composite closed-loop control of the hydraulic cylinder thrust of the hydraulic pushing mechanism.

[0035] In one embodiment, the sensor group includes a steady-state pressure sensor, a first dynamic pressure sensor, and a second dynamic pressure sensor. The steady-state pressure sensor is used to measure the average or low-frequency pressure of the system. The steady-state pressure sensor has a large measurement range but a low response frequency. The first and second dynamic pressure sensors are both high-frequency pressure sensors, used to accurately capture the high-frequency fluctuation details of the pressure. In a key layout scheme, implementing this method requires placing the first dynamic pressure sensor close to the hydraulic cylinder load and the second dynamic pressure sensor close to the hydraulic source side of the main actuator, ensuring that the two are separated by a known physical distance along the pipeline axis.

[0036] In the specific execution of this step, the controller uses its built-in multi-channel high-speed analog-to-digital converter to collect the analog voltage outputs of the above-mentioned sensors in real time and synchronously according to a preset high sampling frequency (e.g., 20 kHz), and converts these voltage signals into digital signals for subsequent processing.

[0037] To support subsequent accurate calculations, the constant tension control method of the hydraulic pushing mechanism also includes a step of preprocessing the acquired raw digital signals. In particular, the controller applies digital high-pass filtering to the first dynamic pressure signal and the second dynamic pressure signal to obtain high-frequency dynamic pressure components that do not contain DC and steady-state components.

[0038] The function of a high-pass filter is to filter out slowly changing or constant parts of a signal, retaining only those signal components that represent rapid pressure fluctuations. In one embodiment, the high-pass filter can be a second-order Butterworth high-pass filter. The cutoff frequency of the second-order Butterworth high-pass filter is set according to the typical spectral characteristics of the system disturbance, for example, at 5 Hz, to ensure that dynamic disturbances caused by hydraulic pump pulsation or load stick-slip can be effectively separated. This preprocessing step is the basis for accurate subsequent disturbance source identification.

[0039] Main actuator: As a low-frequency, high-flow-rate regulating element, it is used to regulate the base pressure of the system. In one embodiment, the main actuator is an electro-hydraulic proportional valve or servo valve with a high flow coefficient.

[0040] High-frequency actuator: As a high-frequency, micro-compensation element, it is used to quickly suppress instantaneous pressure fluctuations. In one embodiment, the high-frequency actuator is a piezoelectric injection valve or a high-speed switching valve with a response frequency of up to 1000 Hz.

[0041] Sensor group: Used for comprehensive monitoring of the pressure status in the main oil inlet line. The sensor group includes: Steady-state pressure sensor: Used to measure the average or low-frequency pressure of a system. Steady-state pressure sensors have a large range but a low response frequency.

[0042] The first dynamic pressure sensor and the second dynamic pressure sensor are both high-frequency pressure sensors used to accurately capture the details of high-frequency pressure fluctuations. In a key layout, the first dynamic pressure sensor is positioned close to the load of the hydraulic cylinder, while the second dynamic pressure sensor is positioned close to the hydraulic source side of the main actuator, with the two separated by a known physical distance along the pipeline axis.

[0043] Controller: As the control core, such as an embedded controller based on a digital signal processor or field programmable gate array, it is electrically connected to the main actuator, high-frequency actuator and sensor group. The controller has an algorithm program for executing the constant tension control method of the present invention embedded or running inside.

[0044] By using first and second dynamic pressure sensors at different locations and combining cross-correlation function analysis, it is possible to determine online whether the source of pressure disturbance is the load side or the hydraulic source side. Based on this determination, the corresponding control components can be selectively activated for targeted suppression. This method makes the control action more targeted, improves the efficiency and accuracy of disturbance suppression, and solves the problem that in a complex disturbance environment, the existing single control strategy cannot simultaneously and effectively suppress the pressure change caused by load nonlinearity and the high-frequency pulsation of the hydraulic source itself, thus leading to insufficient system control accuracy and dynamic stability.

[0045] See attached document Figure 2 The constant tension control method of the hydraulic pushing mechanism of the present invention, in a specific embodiment, includes the following steps: Signal Acquisition: The controller, through its built-in multi-channel high-speed analog-to-digital converter, acquires the analog voltage outputs of each sensor in real time at a preset high sampling frequency, such as 20kHz, and converts the analog voltage outputs into digital signals. Specifically, it acquires the steady-state pressure signal output by the steady-state pressure sensor. Acquire the first dynamic pressure signal output by the first dynamic pressure sensor. and the acquisition of the second dynamic pressure signal output by the second dynamic pressure sensor. .

[0046] To support subsequent calculations, the controller preprocesses the acquired signals, especially the first dynamic pressure signal. Second dynamic pressure signal Digital high-pass filtering is performed to obtain high-frequency dynamic pressure components that do not contain DC and steady-state components. and .

[0047] Digital high-pass filtering can be implemented using a second-order Butterworth high-pass filter, whose transfer function is: ; in, Let be the transfer function of the high-pass filter, which describes the relationship between the filter's output signal and the input signal in the complex frequency domain. For the Laplace operator, This is the cutoff frequency of the filter. is the damping ratio.

[0048] Preprocessing ensures that subsequent disturbance source identification is performed only for high-frequency disturbances.

[0049] Parallel control computation: In each sampling period of the controller, two computational tasks, low-frequency steady-state thrust control and high-frequency dynamic suppression control, are executed in parallel.

[0050] The low-frequency steady-state thrust control steps include: The steady-state error between the target pressure and the steady-state pressure signal is calculated using a proportional-integral-derivative (PID) control law. The main actuator control signal is then calculated based on the steady-state error. The PID control law includes preset proportional gain, integral gain, and derivative gain.

[0051] Low-frequency steady-state thrust control: Performed by a separate logic module or processor core, designed to maintain stable average pressure in the system. This step includes: First, calculate the preset target pressure. (For example, calculated based on the desired tension value) and the steady-state pressure signal after low-pass filtering. steady-state error between

[0052] ; in, For at a certain point in time steady-state error, The target pressure is a pre-set desired pressure value, which is the target that the control system needs to track. For at a certain point in time The actual measured steady-state pressure signal.

[0053] Subsequently, the digital proportional-integral-derivative control law is based on the steady-state error. Calculate the main actuator control signal .

[0054] ; in, To be in discrete time steps The calculated master actuator control signal, The index is the discrete time step, representing the current time. The proportional gain determines the controller's response to the current error. The response intensity To be in discrete time steps steady-state error, The integral gain determines the strength of the controller's response to accumulated historical errors and is mainly used to eliminate the steady-state error of the system.

[0055] This is the integral term of the error, which is given from the initial time. up to the current moment The sum of all historical errors is a discrete approximation of continuous-time integrals. The differential gain determines the strength of the controller's response to the rate of change of error. It has predictive and damping effects, which can speed up the system response and reduce overshoot. The differential term of the error approximates the rate of change of the error by using the difference between the current error and the error at the previous sampling time. The sampling period is the time interval between two consecutive samples.

[0056] See attached document Figure 3 The high-frequency dynamic suppression control task further includes the following steps: The disturbance source identification steps include: performing high-pass filtering on the first dynamic pressure signal and the second dynamic pressure signal to obtain high-frequency dynamic pressure components; calculating the cross-correlation function between the high-frequency dynamic pressure components and determining the peak delay of the cross-correlation function; comparing the sign of the peak delay with a preset time delay dead zone threshold, outputting a disturbance source flag bit, and identifying whether the disturbance originates from the load side or the hydraulic source side.

[0057] The purpose of disturbance source identification is to identify the propagation direction of pressure fluctuations online. First, a length of... Using a sliding window, calculate two high-frequency dynamic pressure components. and Cross-correlation function between : ; in, Let be the cross-correlation function value of signals A and B, which is related to the delay. A function that measures the delay of signal B relative to signal A. The degree of similarity after each sampling point The length of the sliding window is the number of samples used to calculate the correlation. For the discrete time step index of the summation, This is the index of the discrete time step at the current moment. To represent summation within a sliding window containing N most recent samples, For at any time The amplitude of signal A (the high-frequency component of the first dynamic pressure signal). For at any time The amplitude of signal B (the high-frequency component of the second dynamic pressure signal), Delay or time offset, measured in the number of sampling points.

[0058] ; in, Peak delay, i.e., the delay corresponding to the maximum value of the cross-correlation function. value.

[0059] The operator returns the value of the argument that makes the function reach its maximum value, not the maximum value of the function itself. Define search The range, that is, within the preset maximum delay. Search within.

[0060] The upper bound of the search range here It is based on the sensor spacing and the speed of sound in hydraulic oil (Typically around 1200-1400 m / s) Calculated in advance; ; in, For maximum delay, For sensor spacing, The velocity of sound in hydraulic oil.

[0061] Improved robustness of discrimination, finally, based on The value and a preset dead time interval value (For example, 2 sampling points) are compared, and a three-valued disturbance source flag is output. : ; in, The disturbance source flag bit, the value of the disturbance source flag bit ( , , () represent different sources of disturbance. The time delay dead zone threshold, a small positive integer, is used to create an insensitive region to avoid delays caused by minor noise or calculation errors. Misjudgment.

[0062] Calculate the pressure change rate of the steady-state pressure signal. When the pressure change rate is continuously greater than the first positive threshold within the time window and the variance of the pressure change rate is less than the second threshold, the system is determined to be in a viscous state. When the system is in a viscous state and the steady-state pressure signal reaches or exceeds the static friction threshold, the predicted trigger signal is set to the trigger state.

[0063] State prediction is used to anticipate slippage events caused by sudden changes in friction. In a specific embodiment, the specific steps of state prediction include: First, calculate the steady-state pressure signal. Pressure change rate : ; in, For at a certain point in time (corresponding to discrete time) The rate of change of pressure, For the current moment The steady-state pressure value, For the previous sampling time The steady-state pressure value, The sampling period.

[0064] Then, in a length of Within the time window, simultaneously evaluate the following two conditions: Pressure change rate Continuously greater than the preset first positive threshold (This indicates that pressure is accumulating steadily.) Variance of pressure change rate Less than the preset second threshold (This indicates that the accumulation process is smooth and without significant fluctuations.)

[0065] When both conditions are met, the system is determined to have entered a viscous state, and the flag is set to true. Under the condition that the system is in a viscous state, further judgment is made... Does the amplitude reach or exceed the preset static friction threshold? Once the condition is met, it is assumed that the slippage will occur in the next moment or a few moments later, and the prediction trigger signal is immediately generated. Set from 0 to 1.

[0066] In the high-frequency compensation control law synthesis step, the high-frequency actuator control signal is composed of the load disturbance suppression signal, the source disturbance suppression signal, and the prediction feedforward compensation signal superimposed.

[0067] The steps for synthesizing a high-frequency compensation control law include: When the disturbance source flag indicates that the disturbance originates from the load side, the load disturbance suppression signal is activated; when the disturbance source flag indicates that the disturbance originates from the hydraulic source side, the source disturbance suppression signal is activated; when the prediction trigger signal generates a rising edge, the prediction feedforward compensation signal is activated. Based on the aforementioned discrimination and prediction results, the high-frequency compensation control law synthesis adaptively combines to generate the final high-frequency actuator control signal. In one embodiment, the high-frequency actuator control signal is composed of three weighted components superimposed: ; in, For at a certain point in time The synthesized, final output total control signal to the high-frequency actuator; These are the dynamic weights or activation switches for the three control components, with values ​​of 0 or 1, determined based on the disturbance source flag and the state of the predicted trigger signal. These represent the load disturbance suppression signal, the source disturbance suppression signal, and the prediction feedforward compensation signal, respectively.

[0068] The load disturbance suppression signal is calculated based on the reference disturbance signal and the first dynamic compensator. The reference disturbance signal is the high-frequency component of the first dynamic pressure signal. The first dynamic compensator is used to compensate for the dynamic lag of the high-frequency actuator and to counteract it.

[0069] The source disturbance suppression signal is calculated based on the reference disturbance signal and the second dynamic compensator, which is designed to actively absorb pressure pulsations from the hydraulic source.

[0070] Load disturbance suppression signal :when hour, The formula for calculating the load disturbance suppression signal is: ; in, This is a load disturbance suppression signal. As a reference disturbance signal It is the first dynamic compensator used to compensate for the dynamic hysteresis of high-frequency actuators. The parameters of the first dynamic compensator are precisely designed after being identified by the system to ensure that the output compensation pressure wave and the disturbance wave achieve precise anti-cancellation in phase.

[0071] Source-end disturbance suppression signal :when hour, The formula for calculating the source-end disturbance suppression signal is: ; Among them, using one closer to the source end As a reference signal. It is a second dynamic compensator, which can be designed to handle disturbances of specific frequencies (such as the natural frequency of a hydraulic pump). Digital notch filter: ; in, Here is the transfer function of the second dynamic compensator, which is a digital notch filter. This refers to the DC gain or passband gain of the notch filter. For specific disturbance angular frequencies that need to be suppressed, such as the pulsation angular frequency of a hydraulic pump, To convert continuous angular frequencies into corresponding position parameters in the discrete domain, , These represent delays of one and two sampling periods, respectively.

[0072] The radius of the pole is a positive number slightly less than 1 (e.g., 0.99). It determines the bandwidth of the notch filter. The closer it is to 1, the narrower and deeper the notch is, the stronger the suppression effect on a specific frequency, but the more sensitive it is to frequency changes.

[0073] The predictive feedforward compensation signal is a preset pressure relief pulse waveform, which has a pre-calibrated amplitude, shape, and duration.

[0074] Predictive feedforward compensation signal :when When a rising edge is generated, The predictive feedforward compensation signal is a preset pressure relief pulse waveform loaded from a lookup table. The pressure relief pulse waveform has an amplitude, shape (e.g., Gaussian pulse or trapezoidal wave), and duration pre-calibrated based on numerous offline experiments or simulations. This drives a high-frequency actuator to produce a precise, instantaneous pressure relief action, thus offsetting the sudden pressure drop that would be caused by a slippage event.

[0075] The peak delay search is performed within a preset delay range, which is determined based on the distance between the first dynamic pressure sensor and the second dynamic pressure sensor and the speed of sound of the pressure wave propagating in the pipeline.

[0076] In this method, the controller executes two computational tasks in parallel within each sampling period (e.g., every 50 microseconds): low-frequency steady-state thrust control and high-frequency dynamic suppression control. This parallel computation can be implemented by using different cores of a multi-core processor to process the data separately, or by simulating parallelism through precise time-slice round-robin scheduling on a single-core processor. This ensures that the slow adjustment task of the system's basic pressure is not blocked by the instantaneous disturbance suppression task that requires a fast response, and vice versa. This allows the method to maintain long-term steady-state accuracy while having the ability to respond instantly to sudden events.

[0077] The task of low-frequency steady-state thrust control calculation aims to maintain the average thrust of the system at a long-term stable target value. First, the method needs to calculate the steady-state error between the preset target pressure and the filtered steady-state pressure signal.

[0078] The target pressure is the expected pressure value that is preset and stored in the controller according to the process requirements. The steady-state pressure signal acquired in real time contains a small amount of noise, so it is first smoothed by a digital low-pass filter to obtain a more stable average current pressure value.

[0079] Subsequently, the method uses digital proportional, integral, and derivative control laws to calculate the control signal of the main actuator based on the steady-state error.

[0080] The three parts of the digital proportional, integral, and derivative control law each have different responsibilities. The proportional term directly provides a proportional control action based on the current error magnitude, achieving a rapid response. The integral term accumulates the error over a period of time, gradually eliminating long-term steady-state errors caused by continuous system load or model inaccuracy, ensuring that the final pressure accurately reaches the target value. The derivative term calculates based on the rate of error change, predicting future error trends and providing a damping effect to suppress pressure overshoot and accelerate the system stabilization process. The control actions of these three components are superimposed to form the final control signal used to drive the main actuator.

[0081] Composite driver execution: The controller converts the two digital control signals obtained through parallel computation into digital-to-analog conversion channels. and The signal is converted into an analog voltage signal. The main actuator control signal is amplified and then used to drive the main actuator. The high-frequency actuator control signal is driven by a high-voltage, high-speed drive source.

[0082] Loop control: The complete steps S1 to S3 described above are executed precisely once in each sampling cycle of the controller (e.g., every 50 microseconds). At the end of each cycle, the program pointer returns to S1 to start a new round of acquisition, calculation and execution. This high-frequency cyclic control constitutes real-time, composite closed-loop control of the hydraulic cylinder thrust of the hydraulic pushing mechanism, ensuring that the system can respond instantaneously and suppress various disturbances, thereby maintaining ultra-high precision constant tension output.

Claims

1. A constant tension control method for a hydraulic pushing mechanism, characterized in that, Includes the following steps: S1. Signal Acquisition: Real-time acquisition of the steady-state pressure signal output by the steady-state pressure sensor, the first dynamic pressure signal output by the first dynamic pressure sensor, and the second dynamic pressure signal output by the second dynamic pressure sensor in the hydraulic constant tension control system of the hydraulic push mechanism. S2. Parallel control calculation: Parallel execution of low-frequency steady-state thrust control and high-frequency dynamic suppression control; Low-frequency steady-state thrust control: The control signal of the main actuator of the hydraulic constant tension control system is calculated based on the steady-state pressure signal and the preset target pressure. High-frequency dynamic suppression control includes the following steps: Disturbance source identification: Based on the timing relationship between the first dynamic pressure signal and the second dynamic pressure signal, identify the source of pressure fluctuation and output the disturbance source flag bit; State prediction: Based on the steady-state pressure signal and the rate of change of the steady-state pressure signal, predict the slippage event of the hydraulic cylinder of the hydraulic pushing mechanism and output the prediction trigger signal; High-frequency compensation control law synthesis: synthesize high-frequency actuator control signals based on the state of the disturbance source flag and the predicted trigger signal; S3, Composite Drive Execution: The main actuator is driven by the main actuator control signal, and the high-frequency actuator of the hydraulic constant tension control system is driven by the high-frequency actuator control signal. S4. Cyclic Control: Steps S1 to S3 are executed cyclically according to the set control cycle, which is used for the composite closed-loop control of the hydraulic cylinder thrust of the hydraulic pushing mechanism.

2. The constant tension control method for the hydraulic pushing mechanism according to claim 1, characterized in that, The low-frequency steady-state thrust control steps include: The steady-state error between the target pressure and the steady-state pressure signal is calculated, and the main actuator control signal is calculated based on the steady-state error using a proportional-integral-derivative control law. The proportional-integral-derivative control law includes preset proportional gain, integral gain, and derivative gain.

3. The constant tension control method for the hydraulic pushing mechanism according to claim 1, characterized in that, The disturbance source identification step includes: The first dynamic pressure signal and the second dynamic pressure signal are subjected to high-pass filtering to obtain high-frequency dynamic pressure components. Calculate the cross-correlation function between the high-frequency dynamic pressure components and determine the peak time delay of the cross-correlation function; The disturbance source flag is output by comparing the sign of the peak delay with a preset dead zone threshold, and the disturbance is determined to originate from either the load side or the hydraulic source side.

4. The constant tension control method for the hydraulic pushing mechanism according to claim 1, characterized in that, The state prediction step includes: The pressure change rate of the steady-state pressure signal is calculated. When the pressure change rate is continuously greater than a first positive threshold within a time window and the variance of the pressure change rate is less than a second threshold, the system is determined to be in a viscous state. When the system is in a viscous state and the steady-state pressure signal reaches or exceeds the static friction threshold, the predicted trigger signal is set to the trigger state.

5. The constant tension control method for the hydraulic pushing mechanism according to claim 1, characterized in that, In the high-frequency compensation control law synthesis step, the high-frequency actuator control signal is composed of the superposition of the load disturbance suppression signal, the source disturbance suppression signal, and the prediction feedforward compensation signal.

6. The constant tension control method for the hydraulic pushing mechanism according to claim 5, characterized in that, The high-frequency compensation control law synthesis step includes: When the disturbance source flag indicates that the disturbance originates from the load side, the load disturbance suppression signal is activated; When the disturbance source flag indicates that the disturbance originates from the hydraulic source side, the source end disturbance suppression signal is activated; When the prediction trigger signal generates a rising edge, the prediction feedforward compensation signal is activated.

7. The constant tension control method for the hydraulic pushing mechanism according to claim 6, characterized in that, The load disturbance suppression signal is calculated based on the reference disturbance signal and the first dynamic compensator. The reference disturbance signal is the high-frequency component of the first dynamic pressure signal. The first dynamic compensator is used to compensate for the dynamic hysteresis of the high-frequency actuator and to counteract it.

8. The constant tension control method for the hydraulic pushing mechanism according to claim 7, characterized in that, The source disturbance suppression signal is calculated based on the reference disturbance signal and the second dynamic compensator, which is designed to actively absorb pressure pulsations from the hydraulic source.

9. The constant tension control method for the hydraulic pushing mechanism according to claim 6, characterized in that, The predictive feedforward compensation signal is a preset pressure relief pulse waveform, which has a pre-calibrated amplitude, shape, and duration.

10. The constant tension control method for the hydraulic pushing mechanism according to claim 3, characterized in that, The peak delay search is performed within a preset delay range, which is determined based on the distance between the first dynamic pressure sensor and the second dynamic pressure sensor and the speed of sound of the pressure wave propagating in the pipeline.

Images