Hydraulic lifting system synchronous control system and method suitable for complex conditions

Abstract

The invention relates to the technical field of control systems, in particular to a hydraulic lifting system synchronous control system and method suitable for complex situations, and the system comprises a period error monitoring module which is used for collecting the reference position of a hydraulic cylinder and the position fed back by a displacement sensor at multiple time points in a period based on a preset lifting task period; and subtracting the reference position acquired at each time point from the position. According to the method, by collecting and constructing a synchronous error time domain curve, the periodic deviation, caused by mechanical characteristics and load changes, of the hydraulic lifting system in repetitive tasks can be captured, and then a correction feedforward control sequence is generated based on the control input of the last period and the error curve. According to the processing logic, the pure real-time feedback of the control system is converted into a prediction-compensation mode with learning ability, and the processing logic does not depend on a perfect system model, but converts historical error information into pre-compensation for a future control instruction through continuous trial and error learning.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems, and in particular to a synchronous control system and method for a hydraulic lifting system suitable for complex situations. Background Art

[0002] Control systems technology is a core engineering discipline that studies how to manipulate and influence a dynamic system so that its output or state achieves a desired goal. This field builds controllers that apply specific input signals to the controlled object, thereby managing, directing, or regulating its behavior.

[0003] The core weakness of traditional control systems for tasks like hydraulic lift synchronization lies in their inherent reactivity and lack of ability to handle unstructured uncertainty. These systems are regulated based on a fixed mathematical model and real-time error feedback, with their control commands entirely determined by the system deviation at the current moment. This purely reactive approach makes it difficult to foresee and proactively respond to disturbances that recur within each task cycle, such as periodic synchronization errors caused by mechanical friction at a specific location, oil circuit asymmetry, or load eccentricity. As a result, even if the controller parameters are well tuned, the system will still make the same mistakes in each repetitive cycle, resulting in difficult-to-eliminate periodic tracking errors. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a hydraulic lifting system synchronization control system and method suitable for complex situations.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A hydraulic lifting system synchronization control system suitable for complex situations includes: The periodic error monitoring module collects the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the preset lifting task cycle. The reference position collected at each time point is subtracted from the position, and all the differences are arranged in chronological order to obtain the synchronization error time domain curve. The feedforward instruction iteration module, based on the control input of the previous cycle and the synchronization error time domain curve, multiplies each data point in the synchronization error time domain curve by the learning gain coefficient to generate an error correction amount, adds the error correction amount to the control input of the previous cycle at the corresponding time point to generate an updated control signal, and then filters out the signal components in the updated control signal that are higher than the preset frequency threshold to establish a corrected feedforward control sequence.

[0006] Preferably, the system further comprises: A system state observation module, based on the modified feedforward control sequence, the system state space model, and the hydraulic oil flow pulsation characteristics, predicts the current state according to the system state space model and the state estimate at the previous moment, then fuses the measurement values ​​from the displacement sensor and the pressure sensor, calculates the difference between the predicted value and the measured value, and uses the difference to correct the predicted state to obtain the optimal state estimate vector; A real-time feedback compensation module sets a state error weight matrix and a control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, obtains an optimal feedback gain by solving an algebraic Riccati equation, performs matrix multiplication on the optimal feedback gain and the optimal state estimation vector to generate a feedback compensation amount, adds the feedback compensation amount to the modified feedforward control sequence, and generates a compound control instruction for driving the servo valve.

[0007] Preferably, the periodic error monitoring module includes: The position deviation acquisition submodule, based on a preset lifting task cycle, synchronously records the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the cycle. The instantaneous position deviation sequence is obtained by subtracting the corresponding position from the reference position at each time point. The error curve construction submodule pairs each deviation value in the sequence with the acquisition time point based on the instantaneous position deviation sequence, and sorts all paired data according to the chronological order to generate a synchronization error time domain curve.

[0008] Preferably, the feedforward instruction iteration module includes: The error signal weighting submodule generates a time domain error correction signal by multiplying each data point in the synchronization error time domain curve by a preset learning gain coefficient based on the control input and synchronization error time domain curve of the previous cycle; a feedforward instruction updating submodule, which adds the time domain error correction signal and the control input of the previous cycle point by point at the same time point to obtain an unfiltered feedforward instruction based on the time domain error correction signal and the control input of the previous cycle; The high-frequency noise filtering submodule performs a Fourier transform on the instruction sequence based on the unfiltered feedforward instruction to convert it into the frequency domain, sets the amplitude of all frequency components in the frequency domain that are higher than a preset frequency threshold to zero, and then performs an inverse Fourier transform to convert it back to the time domain to establish a corrected feedforward control sequence.

[0009] Preferably, the system status observation module includes: A state prediction submodule, based on the modified feedforward control sequence, the system state space model and the hydraulic oil flow pulsation characteristics, substitutes the state estimate value at the previous moment into the state equation of the system state space model, and uses the modified feedforward control sequence as an input item to calculate the a priori state prediction value at the current moment; The observation gain calculation submodule obtains the state observation gain matrix by solving the prior state prediction value in combination with the process noise covariance matrix and the measurement noise covariance matrix.

[0010] Preferably, the system status observation module further includes: The state correction submodule, based on the prior state prediction value and the state observation gain matrix, subtracts the measurement values ​​from the displacement sensor and the pressure sensor from the corresponding items in the prior state prediction value to obtain a measurement residual, multiplies the measurement residual by the state observation gain matrix, and adds the product result to the prior state prediction value to obtain the optimal state estimation vector.

[0011] Preferably, the real-time feedback compensation module includes: A feedback gain solving submodule sets a state error weight matrix and a control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, and obtains an optimal feedback gain matrix by solving the algebraic Riccati equation associated with the two weight matrices; The compensation signal generating submodule performs a matrix multiplication operation on the negative value of the optimal feedback gain matrix and the optimal state estimation vector based on the optimal state estimation vector and the optimal feedback gain matrix to generate a feedback compensation signal.

[0012] Preferably, the real-time feedback compensation module further includes: The control instruction fusion submodule adds the feedback compensation signal and the modified feedforward control sequence at each time point based on the feedback compensation signal and the modified feedforward control sequence to generate a composite control instruction for driving the servo valve.

[0013] The present invention also provides a synchronous control method for a hydraulic lifting system applicable to complex situations, comprising the following steps: Based on the preset lifting task cycle, the reference position of the hydraulic cylinder and the position fed back by the displacement sensor are collected at multiple time points within the cycle. The reference position collected at each time point is subtracted from the position, and all the differences are arranged in chronological order to obtain the synchronization error time domain curve; Based on the control input of the previous cycle and the synchronization error time domain curve, multiplying each data point in the synchronization error time domain curve by a learning gain coefficient to generate an error correction amount, adding the error correction amount to the control input of the previous cycle at a corresponding time point to generate an updated control signal, and then filtering out signal components above a preset frequency threshold in the updated control signal to establish a modified feedforward control sequence; Based on the modified feedforward control sequence, the system state space model, and the hydraulic oil flow pulsation characteristics, the current state is predicted according to the system state space model and the state estimate at the previous moment, and then the measurement values ​​from the displacement sensor and the pressure sensor are integrated to calculate the difference between the predicted value and the measured value. The predicted state is corrected using the difference to obtain the optimal state estimate vector; Based on the optimal state estimation vector and the corrected feedforward control sequence, a state error weight matrix and a control input weight matrix are set, and the optimal feedback gain is obtained by solving the algebraic Riccati equation. The optimal feedback gain is matrix multiplied by the optimal state estimation vector to generate a feedback compensation amount, and the feedback compensation amount is added to the corrected feedforward control sequence to generate a compound control instruction for driving the servo valve action.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: By collecting and constructing a time-domain synchronization error curve, the present invention captures the periodic deviations caused by mechanical characteristics and load changes in a hydraulic lifting system during repetitive tasks. This allows the generation of a corrected feedforward control sequence based on the control input from the previous cycle and the error curve. This processing logic transforms the control system from a purely real-time feedback-based model to a predictive-compensation model with learning capabilities. Rather than relying on a perfect system model, the system transforms historical error information into pre-compensation for future control commands through continuous trial-and-error learning. This proactively suppresses repeatable disturbances in each subsequent task cycle, progressively reducing synchronization error and achieving close tracking of the target trajectory. Furthermore, by establishing a system state-space model that incorporates the pulsating characteristics of the hydraulic oil flow rate and fusing measurements from displacement and pressure sensors, the optimal state estimate vector is corrected and obtained, overcoming the interference of sensor noise and internal random perturbations on the state observation. The feedback compensation generated based on the optimal state estimate vector is added to the learned and corrected feedforward control sequence to form a composite control command. The system can not only eliminate periodic errors through feedforward learning, but also suppress non-periodic and random disturbances in real time through optimal feedback. While ensuring high synchronization accuracy, it also takes into account the control energy consumption and the stability of system operation, and overall improves the control robustness and comprehensive performance of the hydraulic lifting system under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] See also Figure 1 The present invention provides a technical solution: a hydraulic lifting system synchronization control system suitable for complex situations includes: The periodic error monitoring module collects the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the preset lifting task cycle. The reference position collected at each time point is subtracted from the position, and all the differences are arranged in chronological order to obtain the synchronization error time domain curve. The feedforward instruction iteration module, based on the control input and synchronization error time domain curve of the previous cycle, multiplies each data point in the synchronization error time domain curve by the learning gain coefficient to generate an error correction value, adds the error correction value to the control input of the previous cycle at the corresponding time point to generate an updated control signal, and then filters out the signal components above the preset frequency threshold in the updated control signal to establish a corrected feedforward control sequence; The system state observation module, based on the modified feedforward control sequence, the system state space model, and the hydraulic oil flow pulsation characteristics, predicts the current state according to the system state space model and the state estimate at the previous moment. It then fuses the measurement values ​​from the displacement sensor and the pressure sensor, calculates the difference between the predicted value and the measured value, and uses the difference to correct the predicted state to obtain the optimal state estimate vector; The real-time feedback compensation module sets the state error weight matrix and the control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, obtains the optimal feedback gain by solving the algebraic Riccati equation, performs matrix multiplication on the optimal feedback gain and the optimal state estimation vector to generate the feedback compensation amount, adds the feedback compensation amount to the modified feedforward control sequence, and generates a compound control instruction to drive the servo valve action.

[0018] The periodic error monitoring module includes: The position deviation acquisition submodule, based on a preset lifting task cycle, synchronously records the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the cycle. The instantaneous position deviation sequence is obtained by subtracting the corresponding position from the reference position at each time point. The error curve construction submodule pairs each deviation value in the sequence with the acquisition time point based on the instantaneous position deviation sequence, and sorts all paired data according to chronological order to generate the synchronization error time domain curve.

[0019] Specifically, based on the preset lifting task cycle, for example, a complete lifting and lowering process is defined as 10 seconds. Within this 10-second period, at a fixed sampling frequency, for example, 100 times per second, that is, every 0.01 second, the reference position of the hydraulic cylinder at that moment and the actual position fed back by the displacement sensor are synchronously recorded. The reference position is the ideal trajectory point obtained according to the preset lifting speed and acceleration planning, while the actual position is measured in real time by the displacement sensor (such as magnetostrictive displacement sensor or wire-drawn encoder) installed on the hydraulic cylinder. Synchronous recording is achieved by triggering data acquisition instructions through a unified system clock to ensure that at each sampling time point (in , is the total number of sampling points in the cycle. In this example, ), both obtain the reference position at the same time and actual feedback position , then, for each sampling time point , calculate the instantaneous position deviation , for example, in At 100.5 seconds, if the reference position is 100.5 mm and the sensor feedback position is 100.2 mm, the instantaneous position deviation is , arrange all these calculated instantaneous position deviation values ​​in the order of their occurrence time to form a table containing The time series of deviation values ​​is used to obtain the instantaneous position deviation series.

[0020] Based on the instantaneous position deviation sequence, the sequence is a series of deviation values ​​obtained in the previous step arranged in time order, such as ,in It's at the time In order to construct the synchronization error time domain curve, each deviation value in this sequence is The corresponding collection time point Make explicit pairings to form a series of data point pairs For example, if the first three values ​​of the instantaneous position deviation sequence are 0.1mm, 0.3mm, and 0.2mm, and the corresponding acquisition time points are 0.01s, 0.02s, and 0.03s, then the data point pairs formed are (0.01s, 0.1mm), (0.02s, 0.3mm), and (0.03s, 0.2mm). Since the instantaneous position deviation sequence itself is collected and arranged in time sequence, these data point pairs are logically arranged in time sequence. In actual construction, confirm that this time sequence is strictly maintained and all these paired data points are Collectively, they form a discrete time series data. This ordered sequence consisting of timestamps and their corresponding instantaneous position deviation values ​​can intuitively represent the complete situation of the synchronization error changing with time within a task cycle, and generate a synchronization error time domain curve.

[0021] The feedforward instruction iteration module includes: The error signal weighting submodule generates a time domain error correction signal by multiplying each data point in the synchronization error time domain curve by a preset learning gain coefficient based on the control input and synchronization error time domain curve of the previous cycle; The feedforward instruction update submodule adds the time domain error correction signal and the control input of the previous cycle to obtain the unfiltered feedforward instruction. The high-frequency noise filtering submodule performs Fourier transform on the instruction sequence to the frequency domain based on the unfiltered feedforward instruction, sets the amplitude of all frequency components in the frequency domain that are higher than the preset frequency threshold to zero, and then performs inverse Fourier transform to convert it back to the time domain to establish a corrected feedforward control sequence.

[0022] Specifically, based on the control input of the previous cycle, this is a record of each time point in the previous complete lifting task cycle. Corresponding control command value The time series of the synchronization error calculated in the current cycle is shown as a series of data points. ,in It's at the time The core of the processing is to convert each data point in the synchronization error time domain curve into Multiply by a preset learning gain coefficient , this learning gain coefficient It is a key parameter, and its setting has a direct impact on the learning speed and stability of the system. The value of is usually determined by experimental methods. Initially, a smaller value can be set based on experience, such as 0.2, and then adjusted by observing the convergence of the synchronization error after several cycles. If the error decreases slowly, it can be appropriately increased. , for example, increase by 0.1 each time. If oscillation or divergence occurs, reduce , until an optimal value is found that can quickly reduce the error without causing system instability, for example, it is finally determined to be , for each data point in the synchronization error time domain curve , calculate the corresponding error correction amount , for example, if at time Synchronization error is 0.5 mm (the error unit here needs to be processed later or the gain itself contains a unit conversion factor to match the unit of the control input. For example, here it has been processed or the gain is dimensionless, and the correction amount acts directly on the value of the control signal), and , then the error correction amount of this point is , the error correction amounts calculated at all time points are arranged in chronological order to generate a time domain error correction signal.

[0023] Based on the time domain error correction signal, which is a signal composed of Error correction amount The time series composed of the control input of the previous cycle is a time series composed of the control input of the previous cycle. The control instruction value of the previous cycle In order to update the feedforward instruction, the time domain error correction signal is added to the value of the control input of the previous cycle at the same time point. The specific operation is as follows: for each discrete time point in the cycle (from arrive ,in is the total number of sampling points in the period), take out the time domain error correction signal value corresponding to the time point and the control input value of the previous cycle , then add these two values ​​to get the current time point The unfiltered feedforward command value ,Right now , for example, at time point , if the control input of the previous cycle =5.2 (unit, such as volts or milliamperes, depending on the servo valve drive requirements), the corresponding time domain error correction signal value is 0.325 (consistent with the control input unit or converted), then the unfiltered feedforward command value at this time point is , for all tasks in the entire cycle The same addition operation is performed at each time point, resulting in a indivual Get the unfiltered feedforward instructions for the complete sequence of values.

[0024] Based on the unfiltered feedforward instruction, this is a Updated control command value First, the instruction sequence is Fourier transformed, usually using the Fast Fourier Transform (FFT) algorithm, to convert it from the time domain to the frequency domain, and obtain the amplitude and phase information corresponding to each frequency component, that is, the frequency domain representation Then, all the frequencies in the frequency domain that are higher than the preset frequency threshold are The amplitude of the frequency component is set to zero, this preset frequency threshold The setting of depends on the dynamic response characteristics of the hydraulic system and the expected operating frequency range. For example, if the hydraulic system cannot effectively follow the control signal changes above 30Hz, and the main working command frequency is concentrated in 0-20Hz, then Set to 25Hz. This value can be determined by analyzing the spectrum of the system's open-loop step response to find the system's effective bandwidth, or by setting a value slightly higher than the highest frequency of the expected signal and lower than the starting frequency of the known high-frequency noise based on experience. For example, if the system's expected response frequency does not exceed 20Hz and the sensor noise is mainly above 40Hz, then Can be set to , to retain most of the useful signal and filter out some of the noise, for all frequencies , and its corresponding frequency domain amplitude Set to 0 (or directly represent the complex number Finally, an inverse Fourier transform is performed on the modified frequency domain signal, usually using an inverse fast Fourier transform (IFFT) algorithm, to convert it back to the time domain and establish a modified feedforward control sequence.

[0025] The system status observation module includes: The state prediction submodule, based on the modified feedforward control sequence, the system state space model and the hydraulic oil flow pulsation characteristics, substitutes the state estimate value at the previous moment into the state equation of the system state space model, and combines the modified feedforward control sequence as an input item to calculate the a priori state prediction value at the current moment; The observation gain calculation submodule obtains the state observation gain matrix by solving the prior state prediction value and combining the process noise covariance matrix and the measurement noise covariance matrix; The state correction submodule, based on the prior state prediction value and the state observation gain matrix, subtracts the measurement values ​​from the displacement sensor and pressure sensor from the corresponding items in the prior state prediction value to obtain the measurement residual, multiplies the measurement residual with the state observation gain matrix, and adds the product result to the prior state prediction value to obtain the optimal state estimation vector.

[0026] Specifically, based on the modified feedforward control sequence , system state space model, and hydraulic oil flow pulsation characteristics. The system state space model here is usually obtained by mechanism modeling or system identification of the hydraulic system. For example, the state space model of the hydraulic cylinder system can be expressed in discrete time form: ,in yes The system state vector at the moment includes the position, velocity and cavity pressure of the hydraulic cylinder, is the control input, is the state transition matrix, is the input matrix, is process noise, for example, if the state vector is ,but Maybe one matrix, is a The specific values ​​of matrices are obtained by linearizing and discretizing the hydraulic system dynamics equations (such as the flow continuity equation and Newton's second law), or by parameter estimation through experimental data, such as using the least squares method to identify input and output data. The hydraulic oil flow pulsation characteristics mainly refer to the flow unevenness caused by components such as hydraulic pumps and servo valves. This part of the characteristics can be modeled as the system input A disturbance, or directly reflected in the process noise In the statistical properties of Introduce nonlinear terms or time-varying parameters into the matrix or state equation to reflect the prediction step. The optimal state estimate of Substitute the state transition part of the above discrete state equation and combine it with the current moment The corresponding value in the modified feedforward control sequence As a control input, if the hydraulic oil flow pulsation characteristic is modeled as a disturbance to the input , then the input item is corrected to ,in The characteristics (such as amplitude and frequency) of the pulsation are estimated based on the known pulsation characteristics. For example, if the flow pulsation is mainly caused by the number of plungers in the pump and speed Determined by, its fundamental frequency is ,but This frequency and its harmonic components may be included if the pulsation characteristics are incorporated into and If the exact modeling of is not possible or it is considered as part of the process noise, then the , therefore, the state prediction equation is ,in That is the prior state prediction value at the current moment.

[0027] Based on the prior state prediction value , combined with the process noise covariance matrix and the measurement noise covariance matrix , perform state observation gain matrix Calculation of process noise covariance matrix It reflects the uncertainty of the system model and the influence of unmodeled dynamics. Its diagonal elements represent the variance of the corresponding state variable process noise, and the off-diagonal elements represent the correlation between the process noise of different state variables. The setting is usually based on experience and debugging, for example, for a state containing position and speed For a system with a velocity that is significantly affected by factors such as unmodeled friction, and a position that is primarily obtained by integrating the velocity, then The variance term corresponding to the speed in It will be set higher than the variance term at the corresponding position Larger, for example, if the speed range is The variance of random fluctuations within ,but , while the process noise at the location may be smaller, such as , the measurement noise covariance matrix It reflects the statistical characteristics of the sensor measurement error. Its diagonal elements are the variance of the measurement noise of each sensor. It can usually be obtained from the sensor data sheet or by statistically analyzing a large amount of data collected by the sensor under static conditions. For example, if the accuracy of the displacement sensor is , for example, the error follows a normal distribution and Covering this range, its variance can be set to If there is a pressure sensor, its noise variance is similarly calculated and filled in For example, the standard deviation of the pressure sensor measurement noise is , then the corresponding variance is , state observation gain matrix (i.e., Kalman gain) is obtained by solving the following standard Kalman filter equations: First, calculate the prior error covariance matrix , then calculate the state observation gain matrix ,in is the observation matrix, which maps the state vector to the measurement vector. Its form is determined by the configuration of the sensor and the measured state variables. For example, if the position and pressure are measured directly, then The corresponding rows of the matrix select the position and pressure components in the state vector and calculate them through the above formula to obtain the state observation gain matrix.

[0028] Based on the prior state prediction value And the state observation gain matrix calculated in the previous link First, obtain the current time from the displacement sensor and pressure sensor The actual measured value of ,For example, Then, these actual measured values With the observation matrix Predicting values ​​from prior states The corresponding predicted measurement values ​​extracted from Perform subtraction to obtain the measurement residual (or innovation) For example, if the position prediction in the prior state prediction value is 100.2mm and the pressure prediction is 5.0MPa, and the actual measurement value of the sensor is 100.1mm and 5.1MPa, and the observation matrix is is the identity matrix (for example, the state is directly measurable), then the measurement residual is , then, the calculated measurement residual and the state observation gain matrix Perform matrix multiplication to obtain the correction amount , this correction represents the correction amplitude of the prior state prediction based on the actual measurement. Finally, this correction is combined with the prior state prediction value Add, that is , thus getting the current moment The optimal state estimate vector , and at the same time, update the posterior error covariance matrix ,in is the identity matrix, and the optimal state estimation vector is obtained.

[0029] Specifically, first, the feedforward instruction iteration module achieves adaptive compensation for system repeatability errors through the core concept of iterative learning control. It uses the synchronization error of the previous cycle to correct the feedforward control instructions of the current cycle point by point. This mechanism can effectively deal with disturbance factors in the hydraulic system that are difficult to accurately model but appear periodically, such as nonlinear friction and changes in oil parameters. It enables the system to approach the ideal trajectory through continuous "practice", thereby improving the synchronization control accuracy cycle by cycle. At the same time, by filtering high-frequency noise from the updated instructions in the frequency domain, the stability of the iterative learning process is ensured, avoiding system oscillations caused by noise amplification. The resulting corrected feedforward control sequence is both accurate and smooth, laying the foundation for achieving high dynamic response and smooth operation.

[0030] Second, the system state observation module provides the system with accurate and complete real-time state information by introducing a Kalman filter, which is crucial for coping with complex non-periodic disturbances. This module optimally fuses the predicted values ​​based on the system model with the actual measured values ​​from multiple sensors such as displacement and pressure. This not only effectively filters out the measurement noise of the sensors themselves, but also accurately estimates state variables such as speed that cannot be directly measured but are crucial for control. In particular, by considering characteristics such as hydraulic oil flow pulsation in the model, the accuracy of state prediction is further improved. The final output of the optimal state estimate vector provides a high-quality, low-noise basis for subsequent real-time feedback compensation, enabling the feedback controller to respond more sensitively and accurately to sudden disturbances and model uncertainties, thereby enhancing the robustness and dynamic stability of the entire control system.

[0031] The real-time feedback compensation module includes: The feedback gain solver submodule sets the state error weight matrix and the control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, and obtains the optimal feedback gain matrix by solving the algebraic Riccati equation associated with these two weight matrices. The compensation signal generation submodule performs a matrix multiplication operation on the negative value of the optimal feedback gain matrix and the optimal state estimation vector based on the optimal state estimation vector and the optimal feedback gain matrix to generate a feedback compensation signal; The control instruction fusion submodule adds the feedback compensation signal and the modified feedforward control sequence at each time point based on the feedback compensation signal and the modified feedforward control sequence to generate a composite control instruction for driving the servo valve.

[0032] Specifically, based on the optimal state estimation vector and modified feedforward control sequence , first you need to set the state error weight matrix and the control input weight matrix , these two matrices are the core parameters of the linear quadratic regulator (LQR) design, which determine the relative importance of the control system to the state error and control energy consumption. The state error weight matrix It is usually a semi-positive symmetric matrix with diagonal elements Represents the state variable The penalty for deviation from the expected value (usually zero, or the error from the reference trajectory in tracking problems), The larger the value, the harder the controller will work. Stay close to the desired value, for example, in a hydraulic cylinder position control, if the state vector contains position error and speed error , you may be more concerned about position accuracy, then Can be set as a diagonal matrix ,in Much greater than For example, if the allowable position error standard deviation is 0.1mm and the speed error standard deviation is 1mm / s, we can use Bryson's law to get and , control input weight matrix It is usually a positive definite symmetric matrix with diagonal elements Indicates the control input The size of the punishment, The larger the value, the smaller the amplitude of the control signal generated by the controller, thereby saving control energy or avoiding actuator saturation. For example, if the control input is the servo valve current, its allowable range is ,but Can be set to These weight values ​​usually need to be repeatedly tested and adjusted according to the simulation or actual system operation effect to achieve satisfactory control performance. and Then, by solving the discrete system state space model , matrix and these two weight matrices , The associated discrete algebraic Riccati equation: , to calculate the steady-state solution Once you obtain , the optimal feedback gain matrix It can be calculated as , obtain the optimal feedback gain matrix.

[0033] Based on the optimal state estimation vector And the optimal feedback gain matrix obtained by solving the algebraic Riccati equation in the previous section , perform the specific matrix multiplication operation, first, the optimal feedback gain matrix Taking the negative value, we get , this negative sign indicates that the feedback control law is usually negative feedback, that is, the control action is opposite to the direction of the state deviation. Then, the negative optimal feedback gain matrix and the optimal state estimation vector at the current moment Perform matrix multiplication, that is, calculate ,in is a column vector with the same dimension as the number of state variables in the system (for example, if the states are position and velocity, then vector), the optimal feedback gain matrix The dimension is the number of control inputs multiplied by the number of state variables (for example, if there is a single input and two states, then Matrix), the multiplication operation follows the standard matrix multiplication rules, Each row of The corresponding elements of are multiplied and summed to obtain the feedback compensation signal vector For example, if and , then the feedback compensation signal ,this It is to calculate the control adjustment amount used to correct the system behavior based on the current system state deviation and generate a feedback compensation signal.

[0034] Based on feedback compensation signal and the established modified feedforward control sequence , at each discrete control time point The values ​​of these two signals are algebraically added to modify the feedforward control sequence It is pre-calculated or iteratively learned based on the desired trajectory and system model, and is intended to drive the system to roughly follow the target in an open loop, while the feedback compensation signal It is calculated based on the deviation between the real-time optimal state estimation vector and the expected state (for example, the expected state is zero, that is, LQR is designed for the regulation problem, or the state vector itself represents the error state) through the optimal feedback gain, which is used to correct the deviation and suppress the disturbance in real time. The fusion process is as follows: for the time point of the current control cycle , take out the corresponding corrected feedforward control sequence value and feedback compensation signal value , and then add them together to get the composite control instruction value at that time point , for example, at time point , if the value of the modified feedforward control sequence is 5.525 (unit, such as volts), and the calculated feedback compensation signal is -0.125 (unit as before), then the composite control instruction at that time point is (Units are the same as before). This addition operation is performed in each control sampling period to form a time-varying compound control instruction sequence that combines the rapidity of feedforward control with the robustness and accuracy of feedback control to generate a compound control instruction that drives the servo valve action.

Claims

1. A synchronous control system for a hydraulic lifting system suitable for complex situations, characterized in that: The system comprises: The periodic error monitoring module collects the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the preset lifting task cycle. The reference position collected at each time point is subtracted from the position, and all the differences are arranged in chronological order to obtain the synchronization error time domain curve. The feedforward instruction iteration module, based on the control input of the previous cycle and the synchronization error time domain curve, multiplies each data point in the synchronization error time domain curve by the learning gain coefficient to generate an error correction amount, adds the error correction amount to the control input of the previous cycle at the corresponding time point to generate an updated control signal, and then filters out the signal components in the updated control signal that are higher than the preset frequency threshold to establish a corrected feedforward control sequence.

2. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 1 is characterized in that: The system further comprises: A system state observation module, based on the modified feedforward control sequence, the system state space model, and the hydraulic oil flow pulsation characteristics, predicts the current state according to the system state space model and the state estimate at the previous moment, then fuses the measurement values ​​from the displacement sensor and the pressure sensor, calculates the difference between the predicted value and the measured value, and uses the difference to correct the predicted state to obtain the optimal state estimate vector; A real-time feedback compensation module sets a state error weight matrix and a control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, obtains an optimal feedback gain by solving an algebraic Riccati equation, performs matrix multiplication on the optimal feedback gain and the optimal state estimation vector to generate a feedback compensation amount, adds the feedback compensation amount to the modified feedforward control sequence, and generates a compound control instruction for driving the servo valve.

3. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 1 is characterized in that: The periodic error monitoring module includes: The position deviation acquisition submodule, based on a preset lifting task cycle, synchronously records the reference position of the hydraulic cylinder and the position fed back by the displacement sensor at multiple time points within the cycle. The instantaneous position deviation sequence is obtained by subtracting the corresponding position from the reference position at each time point. The error curve construction submodule pairs each deviation value in the sequence with the acquisition time point based on the instantaneous position deviation sequence, and sorts all paired data according to the chronological order to generate a synchronization error time domain curve.

4. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 1 is characterized in that: The feedforward instruction iteration module includes: The error signal weighting submodule generates a time domain error correction signal by multiplying each data point in the synchronization error time domain curve by a preset learning gain coefficient based on the control input and synchronization error time domain curve of the previous cycle; a feedforward instruction updating submodule, which adds the time domain error correction signal and the control input of the previous cycle point by point at the same time point to obtain an unfiltered feedforward instruction based on the time domain error correction signal and the control input of the previous cycle; The high-frequency noise filtering submodule performs a Fourier transform on the instruction sequence based on the unfiltered feedforward instruction to convert it into the frequency domain, sets the amplitude of all frequency components in the frequency domain that are higher than a preset frequency threshold to zero, and then performs an inverse Fourier transform to convert it back to the time domain to establish a corrected feedforward control sequence.

5. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 2 is characterized in that: The system status observation module includes: A state prediction submodule, based on the modified feedforward control sequence, the system state space model and the hydraulic oil flow pulsation characteristics, substitutes the state estimate value at the previous moment into the state equation of the system state space model, and uses the modified feedforward control sequence as an input item to calculate the a priori state prediction value at the current moment; The observation gain calculation submodule obtains the state observation gain matrix by solving the prior state prediction value in combination with the process noise covariance matrix and the measurement noise covariance matrix.

6. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 5 is characterized in that: The system status observation module also includes: The state correction submodule, based on the prior state prediction value and the state observation gain matrix, subtracts the measurement values ​​from the displacement sensor and the pressure sensor from the corresponding items in the prior state prediction value to obtain a measurement residual, multiplies the measurement residual by the state observation gain matrix, and adds the product result to the prior state prediction value to obtain the optimal state estimation vector.

7. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 2 is characterized in that: The real-time feedback compensation module includes: A feedback gain solving submodule sets a state error weight matrix and a control input weight matrix based on the optimal state estimation vector and the modified feedforward control sequence, and obtains an optimal feedback gain matrix by solving the algebraic Riccati equation associated with the two weight matrices; The compensation signal generating submodule performs a matrix multiplication operation on the negative value of the optimal feedback gain matrix and the optimal state estimation vector based on the optimal state estimation vector and the optimal feedback gain matrix to generate a feedback compensation signal.

8. The synchronous control system of the hydraulic lifting system applicable to complex situations according to claim 7 is characterized in that: The real-time feedback compensation module further includes: The control instruction fusion submodule adds the feedback compensation signal and the modified feedforward control sequence at each time point based on the feedback compensation signal and the modified feedforward control sequence to generate a composite control instruction for driving the servo valve.

9. The method for synchronous control of a hydraulic lifting system applicable to complex situations according to any one of claims 1 to 8, characterized in that: The following steps are involved: Based on the preset lifting task cycle, the reference position of the hydraulic cylinder and the position fed back by the displacement sensor are collected at multiple time points within the cycle. The reference position collected at each time point is subtracted from the position, and all the differences are arranged in chronological order to obtain the synchronization error time domain curve; Based on the control input of the previous cycle and the synchronization error time domain curve, multiplying each data point in the synchronization error time domain curve by a learning gain coefficient to generate an error correction amount, adding the error correction amount to the control input of the previous cycle at a corresponding time point to generate an updated control signal, and then filtering out signal components above a preset frequency threshold in the updated control signal to establish a modified feedforward control sequence; Based on the modified feedforward control sequence, the system state space model, and the hydraulic oil flow pulsation characteristics, the current state is predicted according to the system state space model and the state estimate at the previous moment, and then the measurement values ​​from the displacement sensor and the pressure sensor are integrated to calculate the difference between the predicted value and the measured value. The predicted state is corrected using the difference to obtain the optimal state estimate vector; Based on the optimal state estimation vector and the corrected feedforward control sequence, a state error weight matrix and a control input weight matrix are set, and the optimal feedback gain is obtained by solving the algebraic Riccati equation. The optimal feedback gain is matrix multiplied by the optimal state estimation vector to generate a feedback compensation amount, and the feedback compensation amount is added to the corrected feedforward control sequence to generate a compound control instruction for driving the servo valve action.

Images