Servo control method and system for adjusting flexible resistance of equipment

Abstract

The invention relates to an equipment flexible resistance adjustment servo control method and system, and the method comprises the steps: obtaining resistance torque data and a periodic control instruction of target equipment, carrying out the expansion estimation of disturbance parameters, and obtaining the total disturbance torque data and load equivalent stiffness information of a system; acquiring servo motor data of the target equipment, performing multiple feed-forward compensation on the total disturbance torque data of the system to obtain anti-interference feed-forward compensation data, and performing stiffness reference adjustment in combination with the load equivalent stiffness information to obtain a feed-forward reference coefficient; performing torque parameter coupling on the anti-interference feed-forward compensation data according to the feed-forward reference coefficient to obtain rigidity compensation torque data; and according to the servo motor data, the feedforward reference coefficient and the rigidity compensation torque data, performing instruction construction to obtain a servo control instruction. According to the invention, the dynamic characteristics of the system under complex working conditions can be evaluated more accurately, and the precision of servo control is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment control, and in particular to a servo control method and system for adjusting the flexible resistance of equipment. Background Art

[0002] With the increasing demand for intelligent manufacturing and precision control, achieving precise resistance regulation and dynamic response control for equipment has become a key research topic. Existing servo control methods primarily focus on single-point torque or position control, often overlooking disturbances and changes in load characteristics during actual system operation. This traditional control strategy struggles to adapt to the flexible load requirements of complex operating conditions, resulting in slow system response, insufficient control accuracy, and even impacting the overall performance and service life of the equipment. Summary of the Invention

[0003] The main purpose of the present invention is to provide a servo control method and system for adjusting the flexible resistance of equipment, which can more accurately evaluate the dynamic characteristics of the system under complex working conditions and improve the accuracy of servo control.

[0004] To achieve the above objectives, the present invention provides a servo control method for adjusting the flexible resistance of an equipment, comprising: Obtain the resistance torque data and periodic control instructions of the target equipment, perform disturbance parameter expansion estimation, and obtain the system's total disturbance torque data and load equivalent stiffness information; Acquiring servo motor data of the target device, performing multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and performing stiffness reference adjustment in combination with the load equivalent stiffness information to obtain a feedforward reference coefficient; performing torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data; Instructions are constructed according to the servo motor data, the feedforward reference coefficient and the stiffness compensation torque data to obtain a servo control instruction.

[0005] Furthermore, the acquisition of the resistance torque data and periodic control instructions of the target device, performing disturbance parameter expansion estimation, and obtaining the system total disturbance torque and load equivalent stiffness information includes: Performing motor torque acquisition and periodic signal monitoring on the target device to obtain the resistance torque data and the periodic control instruction; Constructing a disturbance observer model based on the resistance torque data and the periodic control instruction to perform multi-dimensional state variable expansion to obtain an initial value of the total disturbance torque of the system; Performing load-end dynamic identification and parameter convergence according to the initial value of the system total disturbance torque and the periodic control instruction to obtain the load equivalent stiffness information; An iterative mapping correction is performed on the initial value of the total disturbance torque of the system and the load equivalent stiffness information to obtain the total disturbance torque data of the system.

[0006] Furthermore, the servo motor data of the target device is obtained, multiple feedforward compensation is performed on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and stiffness reference adjustment is performed in combination with the load equivalent stiffness information to obtain a feedforward reference coefficient, including: Sampling and incrementally decoding the motor encoding signal of the target device to obtain the servo motor data; Performing characteristic decoupling and separation on the total disturbance torque of the system to obtain a low-frequency disturbance torque component and a high-frequency resonance suppression component; Performing phase advance compensation on the low-frequency interference torque component according to the servo motor data to obtain the anti-disturbance feedforward compensation data; Performing energy coupling operation on the high-frequency resonance suppression component according to the servo motor data to obtain a resonance frequency adjustment amount; A stiffness reference damping mapping is performed according to the resonant frequency adjustment amount and the load equivalent stiffness information to obtain the feedforward reference coefficient.

[0007] Furthermore, performing energy coupling operation on the high-frequency resonance suppression component according to the servo motor data to obtain a resonance frequency adjustment amount includes: Performing real-time load inertia identification processing on the servo motor data to obtain an equivalent load inertia change rate; Performing stiffness-inertia matching calculation on the equivalent load inertia change rate and the load equivalent stiffness information to generate a stiffness attenuation factor; performing dynamic notch frequency band adjustment processing on the high-frequency resonance suppression component according to the stiffness attenuation factor to obtain a resonance suppression amount; Calculate the natural frequency according to the servo motor data to obtain the load resonant frequency characteristic value; The resonance suppression amount and the load resonance frequency characteristic value are subjected to dominant frequency energy extraction to obtain the resonance frequency adjustment amount.

[0008] Furthermore, the step of performing torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data includes: Decomposing the feedforward reference coefficient in terms of flexible stiffness dimension to obtain an axial stiffness gain parameter and a radial stiffness compensation coefficient; Performing deformation direction correction on the anti-disturbance feedforward compensation data according to the radial stiffness compensation coefficient to obtain a deformation suppression compensation torque; Performing spatial trajectory calculation on the servo motor data to obtain flexible load motion data; Performing a spatial compensation operation on the axial stiffness gain parameter and the flexible load motion data to obtain an axial stiffness compensation amount; The deformation suppression compensation torque and the axial stiffness compensation amount are subjected to nonlinear resistance gradient fitting to obtain the stiffness compensation torque data.

[0009] Furthermore, performing spatial trajectory calculation on the servo motor data to obtain flexible load motion data includes: Performing flexible deformation friction analysis on the servo motor data to obtain a rotor friction vector; Performing rigid-flexible transmission coupling on the target device according to the rotor friction vector to obtain the load end trajectory coordinates; Calculating the resistance gradient of the load end trajectory coordinates according to the feedforward reference coefficient to obtain a deformation induced resistance gradient; Performing reverse motion conversion on the deformation-induced resistance gradient to obtain a resistance-compensated deformation displacement; The load end trajectory coordinates and the resistance compensation deformation displacement are fused in servo space to output the flexible load motion data.

[0010] Furthermore, constructing an instruction based on the servo motor data, the feedforward reference coefficient, and the stiffness compensation torque data to obtain a servo control instruction includes: Separating the position and speed parameters of the servo motor data to obtain a rotor angular position signal and a rotor angular velocity component; performing phase synchronization on the stiffness compensation torque data according to the rotor angular position signal to obtain a phase compensation torque vector; Performing velocity coupling on the phase compensation torque vector according to the angular velocity component to obtain a velocity-adaptive compensation component; Linearly superimposing the speed adaptive compensation component and the feedforward reference coefficient to obtain anti-disturbance compensation synthetic torque data; The periodic control instruction is reconstructed according to the anti-disturbance compensation synthetic torque data to obtain the servo control instruction.

[0011] Furthermore, performing velocity coupling on the phase compensation torque vector according to the angular velocity component to obtain a velocity-adaptive compensation component includes: Extracting the internal resistance parameter of the flexible material from the angular velocity component to obtain the viscoelastic hysteresis damping parameter; Performing hysteresis nonlinear compensation on the phase compensation torque vector according to the viscoelastic hysteresis damping parameter to obtain a material internal resistance compensation base value; Performing flexible deformation rate analysis on the angular velocity component to obtain a deformation rate coupling factor; Performing viscoelastic work equivalent conversion on the deformation rate coupling factor and the material internal resistance compensation base value to obtain internal resistance compensation torque data; The internal resistance compensation torque data is corrected with time and temperature equivalent according to preset characteristic data of the equipment flexible connector to obtain the speed-adaptive compensation component.

[0012] The present invention further provides a device flexible resistance adjustment servo control system, which is applied to any of the above-mentioned device flexible resistance adjustment servo control methods, comprising: An acquisition module is used to obtain the resistance torque data and periodic control instructions of the target device, perform disturbance parameter expansion estimation, and obtain the system total disturbance torque data and load equivalent stiffness information; an analysis module configured to obtain servo motor data of the target device, perform multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and perform stiffness reference adjustment based on the load equivalent stiffness information to obtain a feedforward reference coefficient; an association module, configured to perform torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data; A processing module is used to construct an instruction according to the servo motor data, the feedforward reference coefficient and the stiffness compensation torque data to obtain a servo control instruction.

[0013] The present invention provides a device flexible resistance adjustment servo control method and system, which has the following beneficial effects: By obtaining total system disturbance torque data and load equivalent stiffness information through disturbance parameter expansion estimation, the system's dynamic characteristics under complex operating conditions can be more accurately assessed, thereby improving servo control accuracy and providing a more reliable foundation for stable equipment operation. The use of multiple feedforward compensation techniques and stiffness baseline adjustment based on load equivalent stiffness information achieves refined adaptation to varying load characteristics, helping the system maintain stable response under varying operating conditions and avoiding over- or undercontrol. Torque parameter coupling with disturbance-resistant feedforward compensation data based on feedforward baseline coefficients ensures efficient system operation across various operating states and load conditions, reducing control delays and oscillations and improving overall system dynamic performance. By comprehensively analyzing servo motor data and stiffness compensation torque data, more optimal servo control instructions are constructed. Intelligent adjustment strategies are then implemented to optimize the entire system's operation, effectively improving response speed and reducing energy consumption and mechanical wear. By considering the target device's periodic control instructions, the control strategy can be flexibly adjusted based on the characteristics and changing requirements of different application scenarios, making the system more adaptable to diverse industrial environments and task requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a servo control method for adjusting the flexible resistance of a device provided by the present invention; Figure 2 This is a structural diagram of a flexible resistance adjustment servo control system for equipment provided by the present invention.

[0015] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. 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] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0018] Reference Figure 1 As shown, the present invention provides a servo control method for adjusting the flexible resistance of an equipment, comprising: Step S1: Obtain the resistance torque data and periodic control instructions of the target device, perform disturbance parameter expansion estimation, and obtain the system total disturbance torque data and load equivalent stiffness information; Step S2: Acquire the servo motor data of the target device, perform multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and perform stiffness reference adjustment in combination with the load equivalent stiffness information to obtain the feedforward reference coefficient; Step S3: coupling the anti-disturbance feedforward compensation data with torque parameters according to the feedforward reference coefficient to obtain stiffness compensation torque data; Step S4: constructing instructions based on the servo motor data, feedforward reference coefficient and stiffness compensation torque data to obtain servo control instructions.

[0019] Based on the above steps, the detailed process is as follows: Step S1: By collecting real-time strain sensor data and motor current ripple characteristics at the end of the target device's powertrain, the resistance torque waveform under periodic operating instructions is obtained. A nonlinear extended state observer is used to decouple and reconstruct multi-source disturbances, including sudden changes in load inertia and harmonic components of flexible deformation displacement, to construct a dynamic differential equation for the system's total disturbance torque.

[0020] Three core operations are performed within the observer: initializing the variable expansion matrix of the Lyapunov stability constraint and injecting the coupling disturbance factor related to the motor speed; separating the quasi-static and dynamic resistance components of the flexible load through high-dimensional phase space projection; and iteratively updating the load-end stiffness parameters based on an adaptive sliding mode convergence algorithm.

[0021] The final output is two sets of key parameters: the total disturbance torque tensor (including the torque fluctuation spectral characteristics) and the load equivalent stiffness change information (including the direction of the main stiffness axis and the stiffness attenuation gradient), laying the physical model foundation for feedforward compensation. The technical implementation of this step relies on the coordinated work of the harmonic wavelet decomposer and recurrent neural network to ensure that the entire calculation process is completed within a 2ms cycle.

[0022] Step S2: The servo motor data output from the incremental encoder undergoes a four-step processing process: electromagnetic harmonic interference is eliminated using an adaptive Kalman filter; frequency-domain cross-correlation analysis is used to analyze the rotor position pulsation characteristics; and based on this, the dual-band components of the system's total disturbance torque data are separated. For the 0-50Hz low-frequency disturbance torque, a leading phase compensator is used to construct a complex vector transfer function. The compensation amount is adjusted to lead the actual disturbance phase by adjusting the zero-pole distribution. For the 50-2000Hz high-frequency resonant component, the Coriolis effect gain is calculated based on the real-time motor speed, and the viscous damping coefficient obtained from the stiffness information is used to reconstruct the notch filter.

[0023] Dynamic adjustment mechanism of stiffness benchmark: The change rate of the elastic modulus of the main stiffness axis from the load equivalent stiffness information is extracted and mapped into a compensation gain scaling factor. At the same time, the stiffness-damping ratio parameter is used to correct the notch depth, so that the feedforward benchmark coefficient becomes a four-dimensional parameter vector that includes amplitude-frequency compensation and phase calibration.

[0024] Step S3: The feedforward reference coefficients undergo a Cartesian decomposition in three-dimensional stiffness space: the axial component along the motor axis is resolved as a stiffness gain parameter (scalar), and the radial component is projected onto the load deflection plane to form a compensation coefficient matrix. A third-order processing is performed on the anti-disturbance feedforward compensation data: a deformation direction correction algorithm is applied in the radial plane, combining the Poisson's ratio characteristics of the flexible load to generate a moment suppression ellipsoidal distribution. Simultaneously, the servo motor data is input into the kinematic chain solver, and the load end trajectory curvature is resolved using the inverse Jacobian matrix.

[0025] The Hadamard product of the axial stiffness gain and the trajectory curvature tensor is performed to generate a dynamic stiffness compensation that varies with curvature. Finally, a tensor contraction operation is performed using a nonlinear resistance gradient fitter: the deformation suppression torque is projected into tangent space and linked to the axial compensation value by translation on a Riemannian manifold, outputting a pseudotensor structure of the stiffness compensation torque. This process utilizes a conformal geometric algebra computational framework to ensure torque direction invariance under complex surface deformation.

[0026] Step S4: Servo motor data preprocessing involves decoupling position and velocity signals using a fourth-order Butterworth filter bank and constructing a dual-channel angular position and angular velocity observer in the rotor coordinate system. The phase synchronization module performs four key operations: aligning the stiffness compensation torque to the current rotor position angle via a Lie group exponential mapping; establishing a rotating coordinate system transformation that includes flux harmonic compensation; and employing phase-locked loop technology to eliminate phase jitter caused by encoder quantization error. The velocity coupling stage implements rigid-flexible dynamics fusion: a nonlinear transfer function for angular velocity-compensation torque is established based on the viscoelastic constitutive equation; and fractional differentials are used to extract the modulation characteristics of the velocity change rate on the drag gradient.

[0027] The disturbance rejection compensation synthesis process is performed in torque space: the velocity-adaptive component is projected onto a hyperplane defined by the feedforward reference coefficients, and linear superposition is achieved using a convex optimization algorithm with boundary constraints. Final command reconstruction is achieved through dynamic harmonic injection technology: the fundamental component is extracted from the original periodic control command spectrum, and the resultant torque is embedded in the command space according to the power angle characteristics, generating a servo control command with seventh-order harmonic suppression capabilities.

[0028] The present invention provides a servo control method for flexible resistance regulation of equipment. This method uses disturbance parameter expansion estimation to obtain total system disturbance torque data and load equivalent stiffness information. This method can more accurately assess the dynamic characteristics of the system under complex operating conditions, thereby improving the precision of servo control and providing a more reliable foundation for stable operation of the equipment. The method utilizes multiple feedforward compensation techniques and combines load equivalent stiffness information for stiffness baseline adjustment, achieving refined adaptation to varying load characteristics. This helps the system maintain stable response under varying operating conditions and avoids over- or under-control. Torque parameter coupling with anti-disturbance feedforward compensation data based on feedforward baseline coefficients ensures efficient system operation across various operating states and load conditions, reduces control delays and oscillations, and improves the overall dynamic performance of the system. By comprehensively analyzing servo motor data and stiffness compensation torque data, more reasonable servo control instructions are constructed. An intelligent adjustment strategy is then used to optimize the operation of the entire system, effectively improving response speed and reducing energy consumption and mechanical wear. By considering the periodic control instructions of the target equipment, the control strategy can be flexibly adjusted according to the characteristics and changing requirements of different application scenarios, making the system more adaptable to diverse industrial environments and task requirements.

[0029] In one embodiment, the resistance torque data and periodic control instructions of the target device are obtained, and disturbance parameter expansion estimation is performed to obtain the total disturbance torque of the system and load equivalent stiffness information, including: A high-precision strain gauge torque sensor is installed on the equipment's power output shaft, capturing raw torque fluctuation signals in real time at a 10kHz sampling rate. The servo drive command bus is simultaneously monitored, and a hardware-level pulse counter records the duty cycle and frequency characteristics of periodic control commands. The raw torque signal undergoes three-stage preprocessing: a temperature compensation circuit eliminates thermal drift errors; an anti-aliasing filter suppresses high-frequency noise; and an adaptive threshold segmentation method is used to extract payload segments.

[0030] The periodic instruction parsing module performs instruction decoding and timing alignment, identifying the characteristic parameters of the acceleration, constant speed, and braking segments in motion control. It outputs a timestamped drag torque data packet (including peak value, mean value, and harmonic distortion) and a phase-stamped periodic control instruction vector, providing a time-domain synchronization data base for disturbance modeling.

[0031] A nonlinear disturbance observer based on generalized momentum is constructed to initialize a five-dimensional state expansion structure: the motor rotor angular displacement, load-end deflection, electromagnetic harmonic disturbances, viscous friction harmonics, and stiffness coupling error. The state transition equations are established using Lie derivatives, and the drag torque data is mapped into the tangent bundle space to generate the torque gradient tensor. Periodic control commands are injected as external excitation terms, and the rigid and flexible disturbance components are separated through phase trajectory differential geometry analysis.

[0032] A Symge integral operation is performed in the multidimensional phase space, expanding the state variables to a nine-dimensional parameter space that includes the Coriolis and centripetal forces. The initial value matrix of the system's total disturbance torque (including the frequency-domain energy distribution spectrum) is output. This matrix has a time-varying, symmetric, and positive-definite property, directly reflecting the intensity of the disturbance coupling at each node in the transmission chain.

[0033] The initial value of the total system disturbance torque is input into the load dynamics solver. Combined with the acceleration command component of the periodic control instructions, the Newton-Euler recursion equation is constructed. A virtual six-dimensional force sensor is placed at the load suspension point, and a model reference adaptive strategy is used to establish the load mass-stiffness-damping correlation model. A three-step convergence operation is performed: identifying the load resonant main frequency based on the torque harmonic components; extracting the stiffness eigenfrequency using envelope spectrum analysis; and solving for the feasible region of the stiffness parameters using the gradient projection method.

[0034] Functional constraints are applied during the parameter convergence process, triggering convergence lock when the rate of change of the stiffness characteristic frequency is less than 0.1 Hz / ms. The load equivalent stiffness information matrix (3×3 symmetric matrix) is output. Its main diagonal elements contain axial, radial, and tangential stiffness values, and the off-diagonal elements represent the stiffness coupling coefficient.

[0035] A stiffness-disturbance dual-variable correction space is established, and the initial value of the system's total disturbance torque is projected onto the tangent plane of this space. A four-layer iterative loop is executed: the disturbance gradient is calculated in the direction of the principal eigenvector of the stiffness tensor; the disturbance components are separated by orthogonalization; the amplitude and phase errors are corrected according to the minimum work principle; and the inertia coefficients of the matrix are updated. After each iteration, the cross-correlation between the stiffness parameter and the disturbance is tested, and the output is jumped when the correlation coefficient exceeds 0.95. The maximum number of iterations is set to 20, and the step size is adaptively adjusted in the range of 0.05-0.2Nm / deg. Finally, the total disturbance torque data of the system with confidence intervals is generated, which includes 12-dimensional characteristic parameters such as fundamental amplitude, resonance point position, and phase delay angle.

[0036] This embodiment dynamically captures the system's total disturbance torque and load equivalent stiffness through real-time acquisition and disturbance parameter expansion estimation, effectively improving the servo control system's adaptability and anti-disturbance performance in complex flexible load scenarios. A multi-dimensional state variable expansion and iterative mapping correction mechanism accurately compensates for resistance fluctuations caused by changes in load-end stiffness, avoiding control instability caused by parameter identification errors. Optimized feedforward compensation and torque parameter coupling reduce energy losses during control, while simultaneously improving command construction efficiency and overall system response speed.

[0037] In one embodiment, the servo motor data of the target device is obtained, multiple feedforward compensations are performed on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and stiffness reference adjustment is performed in combination with the load equivalent stiffness information to obtain the feedforward reference coefficient, including: High-precision position sensors are deployed to monitor the motor shaft angle and collect pulse signals from the motor encoder. High-speed sampling circuits are used during the sampling process to ensure signal integrity and avoid signal loss due to insufficient sampling frequency. Incremental decoding converts the collected pulse signals into meaningful position, velocity, and acceleration information.

[0038] The decoder processes the two-phase orthogonal encoded signals A and B, determines the direction of rotation by determining the phase difference, and accumulates the pulse count to obtain the angular displacement. The decoding process is implemented using hardware decoding circuits or software algorithms. Hardware decoding is suitable for high-speed applications. The decoded data is filtered to eliminate noise interference and improve signal quality. Through differential calculation, the position information is converted into velocity and acceleration information, completing the acquisition of the basic servo motor parameters. These parameters form the data foundation for subsequent control strategies and provide the necessary input conditions for decoupling the system's total disturbance torque and feedforward compensation. The accuracy of servo motor data directly affects the performance of the control system, so the use of high-precision sampling and decoding methods is crucial. Appropriate data smoothing and preprocessing are also key to ensuring data quality.

[0039] The characteristic decoupling and separation of the system's total disturbance torque is based on an analysis of the overall disturbance characteristics, decomposing the complex disturbance torque into a low-frequency disturbance torque component and a high-frequency resonance suppression component. This decoupling process uses frequency domain analysis to convert the time-domain disturbance signal into the frequency domain. In the frequency domain, low-frequency and high-frequency components are distinguished by setting an appropriate cutoff frequency. The low-frequency disturbance torque component primarily arises from slowly varying factors such as load changes, friction, and gravity, manifesting as steady-state deviations from the system response. The high-frequency resonance suppression component, on the other hand, is primarily caused by mechanical resonance, electrical noise, and elastic deformation of the drive train, manifesting as system oscillations and instability.

[0040] During the decoupling process, bandpass filters are applied to extract signals in different frequency bands to ensure effective separation of the two components. After the low-frequency interference signal is extracted, trend analysis is performed to identify its changing patterns and characteristics; the high-frequency component is subjected to resonant frequency identification to determine the main resonant points. The accuracy of characteristic decoupling depends on the reasonable setting of frequency boundaries and the selection of filtering algorithms, so parameter adjustment is required according to the specific application scenario and device characteristics. The two decoupled components enter different compensation links respectively, with the low-frequency component used for phase advance compensation and the high-frequency component for resonance suppression. This separation processing method can adopt more effective compensation strategies for disturbances of different natures, improving the system's anti-disturbance capability.

[0041] Phase advance compensation of low-frequency interference torque components based on servo motor data is the core link of anti-disturbance feedforward compensation. Phase advance compensation aims to predict and eliminate the impact of low-frequency interference on the system in advance. During the compensation process, the position and speed information extracted from the servo motor data is used in combination with the characteristics of the low-frequency interference torque component. By analyzing the relationship between interference and motor response in historical data, the propagation law and time lag characteristics of the interference are identified. The compensator design adopts an advance filter structure, whose transfer function contains zeros and poles, so that a phase advance is generated within a specific frequency range. The calculation of the compensation amount takes into account the amplitude and frequency characteristics of the interference and the dynamic response time of the system to ensure that the compensation signal can reach the actuator before the interference actually affects the system.

[0042] The compensation signal acts directly on the control output through the feedforward channel, bypassing the feedback loop and reducing the lag of feedback control. The compensation effect is evaluated based on the stability of the system response and the reduction in tracking error, and is optimized by adjusting the compensator parameters. The implementation of phase-lead compensation relies on an accurate understanding of interference characteristics, so the compensator parameters should be dynamically adjusted based on actual operating data. The anti-disturbance feedforward compensation data processed by phase-lead compensation can effectively reduce the impact of low-frequency interference on system performance, improve the system's steady-state accuracy and dynamic response characteristics, and provide a more stable control foundation for subsequent stiffness reference adjustment.

[0043] Performing energy coupling calculations on high-frequency resonance suppression components based on servo motor data is a key step in addressing the system's high-frequency dynamic characteristics. Based on the principles of energy transfer and conversion, energy coupling calculations analyze the energy exchange relationship between high-frequency resonance components and various parts of the system. During the calculation, the servo motor's speed and current data are used to calculate the motor's input power and mechanical output power, determining the direction and distribution of energy. Through resonant characteristic analysis, the energy accumulation points and release mechanisms in the system are identified, and a resonant energy model is established. This model considers the elastic potential energy of the mechanical structure, the kinetic energy of the mass, and the energy consumed by damping, comprehensively describing the energy transfer process of high-frequency vibrations. Energy coupling index calculations are based on the energy distribution ratio at different frequencies, quantifying the impact of resonance on system performance.

[0044] The results of the energy coupling calculation reflect the system's resonant sensitivity at different frequencies, with highly sensitive frequency bands marked as key suppression targets. The resonant frequency adjustment is generated by minimizing the system's resonant energy and altering the system's frequency response characteristics by adjusting control parameters. The energy coupling calculation process utilizes real-time computation to ensure a timely response to changes in system status. The accuracy of the calculation results directly impacts the resonance suppression effect, so during implementation, emphasis is placed on data processing accuracy and algorithm stability. The resonant frequency adjustment obtained through the energy coupling calculation provides a basis for adjusting high-frequency dynamic characteristics in the subsequent stiffness reference damping mapping, enabling the system to more effectively address high-frequency disturbances and resonance issues.

[0045] Performing stiffness-referenced damping mapping based on the resonant frequency adjustment and load equivalent stiffness information is a crucial step in determining the feedforward reference coefficient. Stiffness-referenced damping mapping establishes a correspondence between the system stiffness characteristics and the damping requirements, enabling the control system to adaptively adjust control parameters based on the load characteristics. The mapping process uses the load equivalent stiffness as a reference point and combines it with the resonant frequency adjustment to perform a multidimensional spatial mapping. Load equivalent stiffness information, obtained through force sensor measurement or model estimation, reflects the load's resistance to position changes. The mapping function design utilizes a nonlinear mapping relationship, using different mapping curves for different stiffness ranges to ensure appropriate damping characteristics under various load conditions. The mapping result—the feedforward reference coefficient—directly affects the system's impedance characteristics. Larger coefficient values ​​correspond to higher impedance, suitable for precise positioning tasks; smaller coefficient values ​​correspond to lower impedance, suitable for compliant control scenarios. Accurately adjusting the damping characteristics is crucial to system stability. Excessive damping results in sluggish system response, while insufficient damping can cause oscillation. The implementation of stiffness-referenced damping mapping considers the system's physical constraints and safety margins to ensure that the generated feedforward reference coefficient is within a valid range. The dynamic nature of the mapping process enables the system to respond to load changes in real time and maintain optimal performance under different operating conditions.

[0046] This embodiment achieves effective decoupling and targeted compensation of the system disturbance torque by implementing a servo control method for adjusting the flexible resistance of the equipment. This method separates the total disturbance torque of the system into a low-frequency interference component and a high-frequency resonant component, and performs phase advance compensation and energy coupling operations respectively, which significantly improves the system's ability to suppress disturbances of different frequencies. Based on the precise acquisition and decoding of servo motor data, it provides high-quality basic data for the control system, ensuring the accuracy of disturbance identification and compensation. Through the stiffness reference damping mapping of the resonant frequency adjustment amount and the load equivalent stiffness information, the control parameters can be adaptively adjusted according to the actual load characteristics, and the optimal performance can be maintained under different working conditions.

[0047] In one embodiment, an energy coupling operation is performed on the high-frequency resonance suppression component according to the servo motor data to obtain a resonance frequency adjustment value, including: Real-time load inertia identification based on servo motor data is fundamental to understanding the system's dynamic characteristics. This process dynamically estimates load inertia by analyzing the servo motor's torque command, velocity response, and position feedback. The identification process uses a recursive parameter estimation method, establishing a relationship between motor acceleration and output torque. By collecting and processing data during the motor's acceleration and deceleration processes, key information reflecting the load inertia characteristics can be extracted.

[0048] The collected data is filtered and preprocessed to eliminate the effects of measurement noise and interference. Identification results are continuously updated through an adaptive mechanism to ensure timely tracking of load inertia changes. The equivalent load inertia change rate is calculated based on the difference in inertia estimates between adjacent time periods, reflecting the changing trend of the load's dynamic characteristics.

[0049] This rate of change information directly impacts subsequent stiffness-inertia matching calculations, providing a crucial basis for system parameter adjustments. The accuracy of the identification process significantly impacts the performance of the entire control system, so a multi-verification mechanism is employed during implementation to ensure the reliability of the identification results. The equivalent load-inertia rate of change, obtained through real-time identification, provides accurate data support for subsequent stiffness-inertia matching calculations and is a key element in achieving precise control.

[0050] The stiffness-inertia matching calculation is based on the mechanical and dynamic characteristics of the load. By analyzing the relationship between stiffness and inertia, the system's dynamic response capability is evaluated. The calculation correlates the equivalent load inertia change rate with the load's equivalent stiffness to establish a stiffness-inertia characteristic curve. Matching evaluation criteria include stiffness-inertia ratio, natural frequency, and damping characteristics, which comprehensively reflect the system's dynamic performance. The stiffness attenuation factor is generated based on the matching calculation results and is used to adjust the system's stiffness characteristics. The size of the attenuation factor directly affects the system's dynamic response characteristics. A larger attenuation factor indicates that the system requires stronger stiffness suppression, while a smaller attenuation factor indicates that the system can maintain high response characteristics.

[0051] The matching calculation process utilizes real-time processing to ensure prompt response to system state changes. The accuracy of the calculation results directly impacts the effectiveness of subsequent dynamic notch band adjustments, so during implementation, emphasis is placed on accuracy and stability. The stiffness attenuation factor, derived from the stiffness-inertia matching calculation, provides a crucial control parameter for subsequent dynamic notch band adjustments, enabling the system to better adapt to changing load characteristics.

[0052] Dynamic notch band adjustment is based on the change in the stiffness attenuation factor, precisely suppressing high-frequency resonances by adjusting the notch filter's center frequency and bandwidth. During adjustment, the notch filter's parameters are dynamically updated as the stiffness attenuation factor changes, ensuring optimal notching. The notch band selection considers the system's resonant characteristics and operating frequency range to avoid excessive attenuation of the desired signal. The resonance suppression value is generated based on the notch filter's output characteristics and reflects the system's ability to suppress high-frequency resonances.

[0053] The notch filter design utilizes an adaptive structure that automatically adjusts its characteristics based on changes in the resonant frequency. The accuracy of the frequency band adjustment directly impacts the resonance suppression effect, so a high-precision digital filter is employed in its implementation. During the dynamic notch processing, the response is monitored in real time to ensure the stability of the suppression effect. The calculation of the resonance suppression amount takes into account the system's bandwidth requirements and stability constraints, ensuring the suppression effect while avoiding the introduction of new instabilities. The resonance suppression amount obtained through dynamic notch band adjustment provides the foundational data for subsequent dominant frequency energy extraction and is a key step in achieving precise resonance control.

[0054] The natural frequency calculation is based on the servo motor's operating data. By analyzing the system's free vibration characteristics, its natural vibration frequency is identified. The calculation utilizes the motor's speed, position, and current data, and spectral analysis is used to extract the system's frequency response characteristics. Natural frequency identification utilizes a variety of frequency analysis techniques, including fast Fourier transform and wavelet analysis, to ensure accurate extraction of frequency characteristics. The calculation of the load resonant frequency eigenvalue takes into account the system's mechanical structure and motion state, reflecting the load's vibration characteristics under different operating conditions. The frequency characteristics are tracked in real time during eigenvalue extraction. The accuracy of the natural frequency calculation significantly impacts the system's resonant control effectiveness, so high-precision numerical processing methods are employed during the calculation process. The frequency eigenvalue is updated in real time to ensure timely response to changes in the system's state. The load resonant frequency eigenvalue obtained through natural frequency calculation provides an accurate frequency reference for subsequent extraction of dominant frequency energy, laying the foundation for precise resonant control.

[0055] The dominant frequency energy extraction process compares the resonance suppression value with the load resonant frequency characteristic values, identifying the primary resonant components in the system through energy distribution characteristics. Frequency domain analysis is used to quantitatively assess the energy contribution of different frequency components. The dominant frequency is identified based on energy concentration analysis, selecting the frequency component with the largest energy contribution as the target for regulation. The resonant frequency adjustment value is generated by taking into account system stability requirements and control bandwidth limitations to ensure reliable regulation.

[0056] The calculation of the adjustment factor comprehensively considers the influence of multiple frequency components. The accuracy of the energy extraction process directly affects the system's resonant control effectiveness, so high-precision signal processing technology is employed in the implementation. The dominant frequency identification results are kept up to date through a real-time update mechanism, ensuring timely response to changes in system characteristics. The resonant frequency adjustment factor, derived from the dominant frequency energy extraction, completes the processing of the high-frequency resonance suppression component, providing a precise basis for the system's resonant control.

[0057] This embodiment performs real-time load inertia identification processing on the servo motor data to obtain the equivalent load inertia change rate, thereby achieving an accurate grasp of the dynamic characteristics of the system and providing basic data support for flexible resistance adjustment. The equivalent load inertia change rate is matched with the load equivalent stiffness information to generate a stiffness attenuation factor, so that the system can accurately evaluate the degree of matching of dynamic characteristics and improve control accuracy. According to the stiffness attenuation factor, the high-frequency resonance suppression component is dynamically adjusted to obtain the resonance suppression amount, thereby achieving accurate suppression of high-frequency resonance and effectively reducing system vibration. The natural frequency is calculated through the servo motor data to obtain the load resonance frequency characteristic value, providing an accurate frequency reference for the system. The resonance suppression amount and the load resonance frequency characteristic value are subjected to dominant frequency energy extraction to obtain the resonance frequency adjustment amount, completing the processing of the high-frequency resonance suppression component, enabling the system to maintain a stable dynamic response capability when facing different load characteristics, and significantly improving the flexible control performance of the servo system.

[0058] In one embodiment, torque parameter coupling is performed on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data, including: Feedforward reference coefficients represent the device's predictive compensation parameters for external disturbances. For flexible load devices, these parameters need to be converted into more precise control variables through stiffness dimension decomposition. The stiffness dimension decomposition process starts with a unified feedforward reference coefficient and maps it to the two primary spatial dimensions of axial and radial directions.

[0059] During the decomposition process, the feedforward reference coefficients are decomposed into stiffness characteristics in different directions. The axial stiffness gain parameter reflects the stiffness variation along the main axis, while the radial stiffness compensation coefficient characterizes the stiffness characteristics perpendicular to the main axis. This decomposition method conforms to the physical characteristics of flexible loads exhibited in different directions, making control more precise. Dimensional decomposition uses a matrix projection method to project the feedforward reference coefficients into a predefined stiffness feature space. The construction of this feature space is based on the structural characteristics and real-time operating status of the system, including the inherent properties of the flexible material and the current load conditions.

[0060] The decomposition process considers the differences in the dynamic response of flexible loads in different directions. By collecting historical operating data and establishing a stiffness characteristic model, the decomposition results are more consistent with actual operating conditions. The axial stiffness gain parameter typically reflects high-frequency response characteristics, while the radial stiffness compensation coefficient is more representative of the ability to suppress large, low-frequency deformations.

[0061] Correcting the deformation direction of the anti-disturbance feedforward compensation data based on the radial stiffness compensation coefficient is crucial for accurately suppressing deformation of flexible loads. The anti-disturbance feedforward compensation data includes an estimated response to external disturbances. The deformation direction correction process utilizes the radial stiffness compensation coefficient to spatially transform the anti-disturbance feedforward compensation data, ensuring that the direction of the compensation torque aligns with the actual deformation direction.

[0062] The correction process takes into account the nonlinear deformation characteristics of the flexible load under different working conditions, and adjusts the direction of the feedforward compensation data through the radial stiffness compensation coefficient to generate a more accurate deformation suppression compensation torque. The correction algorithm uses a vector decomposition method to decompose the anti-disturbance feedforward compensation data into components parallel and perpendicular to the deformation direction, and then recombine these components based on the radial stiffness compensation coefficient. By considering the anisotropic characteristics of the flexible load in the radial direction, differentiated compensation is provided for deformations in different directions. In the process of generating the deformation suppression compensation torque, based on the inertia characteristics and damping characteristics of the flexible load, the compensation torque can not only suppress the current deformation, but also prevent future deformation trends.

[0063] Calculating the spatial trajectory of the servo motor data is a key step in obtaining the actual motion state of the flexible load. The servo motor data contains the motion state information of the motor, such as the position, velocity, acceleration, etc. Through spatial trajectory calculation, these basic data are converted into comprehensive features that describe the motion of the flexible load. The spatial trajectory calculation process involves data fusion in multiple dimensions, including motor encoder data, current data, and historical motion trajectory data. The calculation process first filters the original servo motor data to eliminate the influence of noise on the trajectory calculation. The filtered data is mapped to the workspace through a preset kinematic forward solution model to obtain the spatial position and posture information of the end effector. The spatial trajectory calculation not only takes into account the movement of the rigid part, but also combines the emerging flexible deformation to predict the deformation state of the flexible load during the movement.

[0064] By analyzing position changes over a continuous time series, the velocity and acceleration distributions of the flexible load are calculated. This distribution information serves as an important basis for subsequent compensation calculations. Spatial trajectory calculation also includes feature point extraction and trajectory smoothing to ensure that the calculated results accurately reflect the motion characteristics of the flexible load. Feature point extraction identifies target points in the trajectory, such as direction changes and velocity extremes, which are crucial for understanding the load's motion pattern. Trajectory smoothing ensures that the calculated flexible load motion data is continuous and smooth, avoiding jumps in compensation calculations.

[0065] The spatial compensation calculation establishes a mapping relationship between stiffness characteristics and the actual motion state, generating precise axial compensation based on the current motion state. The spatial compensation calculation utilizes a vector integration method, using the axial stiffness gain parameter as a weighting function to integrate the flexible load motion data over the spatial domain. This calculation considers the varying stiffness characteristics of the flexible load under different motion states, particularly the dynamic stiffness characteristics during high-speed motion and rapid changes of direction.

[0066] The spatial compensation operation first spatially decomposes the flexible load motion data to identify the axial component of the motion. The decomposed axial motion data is then multiplied by the axial stiffness gain parameter to generate a preliminary axial compensation value. This preliminary compensation value is then adjusted using a nonlinear mapping function to account for changes in stiffness characteristics under different operating conditions. This nonlinear mapping function, constructed based on historical operating data, can reflect stiffness variations under different load conditions and motion states. The spatial compensation operation also considers the system's response delay, performing a time-domain prediction of the axial compensation value to ensure that the compensation action is synchronized with actual requirements. This prediction method, based on current motion trends and historical response characteristics, effectively compensates for system delay effects. Through this compensation operation, which comprehensively considers both spatial and temporal characteristics, the system-generated axial stiffness compensation value accurately corresponds to the actual requirements of the flexible load, effectively improving the system's axial stiffness performance.

[0067] Nonlinear resistance gradient fitting combines two different compensation dimensions into a unified torque data set, while also accounting for the nonlinear characteristics of the system. The fitting process utilizes a piecewise continuous function model to map the compensation characteristics under different operating conditions into a unified torque space. The piecewise function is designed based on the system's response characteristics under varying load and speed conditions, ensuring smooth and continuous compensation torque across the entire operating range.

[0068] Nonlinear resistance gradient fitting first normalizes the deformation suppression compensation torque and axial stiffness compensation to ensure consistency in numerical scale. The normalized data is initially fused using a weighting function, which dynamically adjusts based on the current operating state to ensure the contributions of the two compensation quantities match actual requirements. This initially fused data is then converted into a gradient characteristic using a nonlinear mapping function, reflecting the variation of the compensation torque with the deformation variable. The gradient characteristic curve is smoothed using spline interpolation to eliminate discontinuities.

[0069] The smoothed gradient characteristics are combined with the current state parameters of the system to generate the final stiffness compensation torque data. The stiffness compensation torque data has good directionality and amplitude characteristics, and can provide accurate compensation for deformations in different directions and degrees. Through this nonlinear gradient fitting method, the compensation torque generated by the system can effectively suppress deformation while maintaining the smoothness of movement, significantly improving the motion accuracy and stability of the flexible load system. This embodiment decomposes the feedforward reference coefficient into flexible stiffness dimensions and maps the unified parameters into axial and radial dimensions, so that the control system can fine-tune the stiffness characteristics in different directions, thereby improving the system's adaptability to complex environmental changes. The anti-disturbance feedforward compensation data is corrected for deformation direction based on the radial stiffness compensation coefficient to ensure that the direction of action of the compensation torque is consistent with the actual deformation direction, effectively suppressing radial deformation under various working conditions. By performing spatial trajectory calculations on the servo motor data, the actual motion state of the flexible load is obtained, providing a reliable basis for subsequent compensation calculations. The axial stiffness gain parameter is spatially compensated with the flexible load motion data to generate an accurate axial stiffness compensation amount, effectively improving the system's axial stiffness performance.

[0070] In one embodiment, spatial trajectory calculation is performed on the servo motor data to obtain flexible load motion data, including: Flexible deformation friction analysis is a fundamental step in the flexible resistance-adjusted servo control method. This process begins with raw data from the servo motor, acquiring key parameters such as current, voltage, speed, and position during operation through a data acquisition system. After preprocessing, this data enters the deformation friction analysis module, which maps the motor's operating status to the friction characteristics of the rotor surface.

[0071] During the analysis process, the difference between the motor's output torque and the actual load is analyzed to identify the friction resistance component caused by the deformation of the flexible material. Deformation friction analysis uses dynamic boundary layer theory to correlate microscopic deformations on the contact surface with macroscopic friction forces. The analysis engine measures tiny torque fluctuations during rotor rotation and, combined with parameters from the material property library, constructs a friction stress distribution map on the rotor surface. This distribution map is represented as a vector field in three-dimensional space, where the direction of each vector represents the direction of friction, and its magnitude represents its intensity. The analysis process also considers the effect of temperature on material properties, correcting the friction coefficient using a temperature compensation function. The final rotor friction vector output encompasses three key dimensions: direction, magnitude, and time-varying characteristics, providing accurate friction boundary conditions for subsequent rigid-flexible transmission coupling. Accurate analysis of the rotor friction vector is crucial for reducing system resistance fluctuations and improving control accuracy.

[0072] By establishing a transfer function between the motor output and the actual motion at the load end, the nonlinear deformation problem in the flexible transmission system is solved. The coupling process uses the rotor friction vector as input and, combined with the mechanical structural parameters of the transmission chain, constructs a complete force transmission path. The coupled model divides the transmission system into two parts: a rigid segment and a flexible segment. The rigid segment uses a rigid body mechanics model to describe the relationship between force and motion, while the flexible segment incorporates a material elastic deformation model. The two-segment model maintains continuity in force and displacement through interface conditions, forming a complete coupled system.

[0073] In the coupling calculation, the rotor friction vector is first converted into a torque at the entrance of the transmission chain, and then transmitted along the transmission chain, taking into account factors such as gear ratio, bearing friction, and elasticity of the connector. When the torque is transmitted to the flexible section, the system calculates the deformation of the flexible component based on the stress-strain relationship of the material, and converts the deformation into a displacement deviation at the load end. Through iterative calculation, the system determines the load end position that satisfies force balance and geometric constraints, that is, the load end trajectory coordinates. These coordinates are represented as a sequence of points in three-dimensional space, which fully describes the position, velocity, and acceleration characteristics of the load during motion. The accurate acquisition of the load end trajectory coordinates lays the foundation for the subsequent resistance gradient calculation.

[0074] Resistance gradient calculation is the process of converting load-end trajectory information into a system resistance distribution. This calculation, based on the load-end trajectory coordinates and combined with preset feedforward reference coefficients, analyzes resistance trends in all directions during motion. The feedforward reference coefficients are a set of parameters predetermined based on historical system operating data and material properties, used to calibrate the accuracy of the resistance calculation model. The resistance gradient calculation extracts information about changes in motion direction from the load-end trajectory coordinate sequence, identifying the curvature and torsion characteristics of the motion trajectory. The calculation engine combines these trajectory characteristics with the elastic deformation properties of the material to derive the degree of deformation of the flexible component in all directions.

[0075] The degree of deformation is multiplied by the material's stiffness properties to yield the internal stress distribution generated by the deformation. The projection of this internal stress onto the contact surface is the deformation-induced resistance. The resistance gradient calculation processes every point on the trajectory to form a complete resistance gradient field. This gradient field describes the changes in resistance magnitude and direction during load motion. Regions with large gradient values ​​indicate drastic resistance changes and increased system control difficulty. The calculation of the deformation-induced resistance gradient directly impacts the subsequent resistance compensation effect and is the core of the entire control method. Resistance gradient information enables the control system to foresee resistance changes that will occur during future motion, providing a basis for proactive control.

[0076] Inverse kinematic transformation (IKT) is the computational process that converts drag gradient information into actual compensation. This step uses the deformation-induced drag gradient as input and, through reverse derivation, calculates the compensation displacement required to eliminate or reduce the drag effect. IKT begins by establishing a mapping function between drag and deformation. This function describes how a given deformation generates a corresponding amount of drag.

[0077] By inverting the function, the system derives the conversion relationship from target resistance to desired deformation. This inverse conversion process accounts for the nonlinear properties of the material and improves computational efficiency through piecewise linearization techniques. The conversion algorithm extracts key points from the resistance gradient field, typically located in areas of significant resistance variation. For each key point, the system calculates the deformation required to neutralize the resistance at that point, forming a local compensation solution. This local compensation solution undergoes spatial smoothing to ensure spatial continuity of the compensated deformation and avoid new vibrations during the compensation process.

[0078] The processed compensation deformation forms a complete deformation field, describing the required displacement adjustments for each part of the system. The resistance compensation displacement is represented as a vector field, with each vector pointing in the compensation direction and its magnitude representing the compensation amount. These compensation values ​​are subsequently combined with the original trajectory coordinates in servo spatial fusion to form the corrected motion command. Accurate calculation of the resistance compensation displacement ensures that the system can proactively counteract resistance fluctuations during flexible load movement, improving motion smoothness.

[0079] Servo space fusion is the last step in the flexible resistance adjustment method. This step integrates the original trajectory data with the compensation data to generate optimized motion control instructions. Spatial fusion takes the load-end trajectory coordinates and the resistance compensation deformation displacement as input, and constructs the corrected motion space through coordinate transformation and data synthesis. The fusion process first converts the two sets of data to the same reference coordinate system to ensure consistency between the data. After the conversion, the system applies the corresponding compensation displacement to each point on the trajectory to obtain the corrected spatial position. The fusion algorithm uses a weighted average method to dynamically adjust the weight ratio of the original trajectory and the compensation amount according to the current state of the system. The weight adjustment takes into account factors such as motion speed, load changes and external interference, and maximizes the compensation effect while ensuring control stability.

[0080] The fusion results form complete flexible load motion data, including multi-dimensional information such as position, velocity, acceleration, and torque. This data is organized in a time series format and directly fed into the servo controller for execution. The main difference between flexible load motion data and traditional rigid body control instructions is that it includes pre-compensation for flexible deformation and can actively suppress vibration and position deviation during motion. The output of servo spatial fusion enables the control system to accurately track the set trajectory while adapting to the special needs of flexible loads, achieving high-precision, low-vibration motion control. The fused motion data is fed back to the servo drive, completing the entire flexible resistance adjustment control loop.

[0081] This embodiment performs flexible deformation friction analysis on the servo motor data to accurately obtain the rotor friction vector, providing the system with accurate friction boundary conditions, effectively reducing resistance fluctuations and improving control accuracy. The rigid-flexible transmission coupling technology based on the rotor friction vector solves the nonlinear deformation problem in the flexible transmission system, making the load end trajectory coordinates more accurate. The calculation of the deformation-induced resistance gradient enables the control system to foresee the resistance changes that will occur in future movements, providing a reliable basis for forward-looking control. The precise calculation of the resistance-compensated deformation displacement ensures that the system can actively counteract resistance fluctuations in the flexible load movement, significantly improving the smoothness of the movement. The flexible load motion data finally generated by servo space fusion includes pre-compensation for flexible deformation, which can actively suppress vibration and position deviation during the movement process, achieve high-precision, low-vibration motion control, and greatly improve the response speed and positioning accuracy of the flexible load system.

[0082] In one embodiment, the servo control instruction is constructed based on the servo motor data, the feedforward reference coefficient, and the stiffness compensation torque data, including: Servo motor data contains mixed information of multiple physical quantities. Obtaining the rotor angular position signal and rotor angular velocity component through parameter separation is the basis for achieving precise control. In this step, the raw data collected by the motor encoder includes multi-dimensional information such as position, velocity, and acceleration, which are stored in the form of digital signals. The parameter separation process uses signal analysis technology to deconstruct the composite signal into independent physical quantities. The rotor angular position signal represents the instantaneous angular position of the motor shaft, usually in radians or degrees, and records the rotation angle of the rotor relative to the reference zero point. The angular velocity component reflects the rotation rate of the rotor and represents the change in angular position per unit time. Digital filtering methods are used in the parameter separation process to eliminate measurement noise and interference to ensure that the separated signal has a high signal-to-noise ratio. The filter design takes into account the dynamic response characteristics of the system to avoid introducing excessive phase delay.

[0083] A high-precision interpolation algorithm is used to improve the resolution of the position signal, enabling the system to detect even small angular changes. The velocity component is derived through position differential calculation, with time window smoothing used to reduce transient fluctuations. The accuracy of parameter separation directly impacts the accuracy of subsequent control, so the separation algorithm has been optimized to reduce computational latency and improve real-time performance. This step outputs high-precision rotor angular position signals and angular velocity components, laying the data foundation for subsequent phase synchronization and velocity coupling.

[0084] A phase relationship exists between the stiffness compensation torque data and the rotor angular position, enabling precise application of the compensation torque through phase synchronization. The system reads preset stiffness compensation torque data, which is typically generated based on a system resistance characteristic model and contains the required compensation torque values ​​for different angular positions. The phase synchronization process extracts the torque value corresponding to the current angular position from the stiffness compensation torque data based on the current rotor angular position signal. This correspondence is achieved through table lookup and interpolation, ensuring that the compensation torque accurately matches the actual angular position. To overcome mechanical backlash and friction nonlinearity issues in the system, the phase synchronization process introduces a phase correction mechanism to compensate for time lag effects in the system.

[0085] The phase compensation torque vector is expressed as a function of angular position, with its elements corresponding to the torque compensation values ​​at each angle. The vector elements are arranged in ascending order of angle, facilitating fast indexing and interpolation calculations. The phase synchronization operation takes into account the continuity of the torque over the angular period, avoiding sudden changes in torque when the angle crosses the period boundary. The synchronization process also dynamically adjusts the amplitude of the compensation torque based on the current operating state of the system to adapt to different load conditions. The phase compensation torque vector output by this step precisely corresponds to the current rotor position, ensuring that the compensation torque acts at the most appropriate time, improving the smoothness and accuracy of the system response.

[0086] Angular velocity modulates the compensation torque, and speed coupling enables adaptive adjustment of the compensation torque as speed changes. In this step, the system uses the rotor angular velocity component as a modulation factor to correct the phase compensation torque vector. Speed ​​coupling uses a nonlinear mapping relationship, applying different coupling functions in different speed ranges to adapt to the dynamic characteristics of the system at various speeds. The compensation effect is enhanced in low-speed areas to overcome static friction and starting resistance; compensation is appropriately weakened in high-speed areas to avoid oscillations caused by over-compensation of the system. The speed coupling process takes into account inertia effects and introduces speed look-ahead compensation to predict the system's motion trend in advance and improve the timeliness of compensation.

[0087] The velocity-adaptive compensation component combines position-dependent stiffness characteristics with velocity-dependent damping characteristics to achieve a comprehensive compensation effect. The coupling algorithm employs an adaptive weight distribution mechanism to dynamically adjust the compensation strength based on the rate of velocity change, providing a smooth transition during rapid speed changes. Velocity coupling also considers direction, employing a symmetrical compensation strategy for forward and reverse motion to ensure consistent system performance in both directions. The velocity-adaptive compensation component output by this step automatically adjusts its compensation characteristics as the system speed changes, adapting to various dynamic operating conditions and improving system stability and tracking accuracy under variable speed conditions.

[0088] The feedforward reference coefficient provides the base compensation torque, which is superimposed with the speed-adaptive compensation component to form a complete anti-disturbance compensation strategy. In this step, the preset feedforward reference coefficients are read. These coefficients reflect the basic compensation requirements under standard operating conditions and are obtained by system identification. Linear superposition uses a weighted summation method to combine the speed-adaptive compensation component with the feedforward reference compensation. The superposition weight is dynamically adjusted according to the current state of the system. During steady-state operation, the proportion of reference compensation is increased to improve system stability; during transient processes, the proportion of speed-adaptive compensation is increased to improve system responsiveness. A smooth transition mechanism is introduced into the superposition process to avoid shocks caused by sudden changes in the compensation torque.

[0089] The combined torque data for anti-disturbance compensation includes position-related compensation, speed-related compensation, and reference compensation, forming a comprehensive compensation strategy. The combined torque is amplitude-limited to prevent it from exceeding the system's physical constraints, protecting the motor and mechanical structure. The superposition algorithm considers the frequency characteristics of each compensation component to ensure appropriate compensation at different frequencies. The combined torque data for anti-disturbance compensation output from this step comprehensively considers the static and dynamic characteristics of the system, effectively resisting the influence of external disturbances and internal nonlinear factors, thereby improving the system's robustness and accuracy.

[0090] The fusion of periodic control instructions and compensation torque is the key to achieving flexible resistance regulation, and the final servo control instructions are generated through instruction reconstruction. In this step, the system makes corrections based on the original periodic control instructions in combination with the anti-disturbance compensation synthetic torque data. Instruction reconstruction adopts a feedforward compensation strategy to convert the compensation torque into an equivalent current instruction increment and superimpose it on the basic current instruction. The reconstruction process takes into account the torque coefficient of the motor to achieve accurate conversion from torque instruction to current instruction. Instruction reconstruction also introduces phase correction to compensate for the electrical and mechanical delays in the system to ensure that compensation acts at the best time. The servo control instruction contains three components: position instruction, speed instruction and torque instruction, forming a complete three-loop control structure.

[0091] During the command reconstruction process, the system monitors motor state parameters in real time and implements command limiting measures when the motor approaches saturation to avoid overcurrent and overvoltage. The reconstruction algorithm uses an adaptive regulation mechanism to dynamically adjust the compensation gain based on the system's response characteristics to maintain system stability under various operating conditions. Servo control commands undergo timing optimization to ensure orderly execution within the servo control cycle, avoiding command accumulation and execution conflicts. The servo control commands output in this step integrate basic motion control requirements and anti-disturbance compensation requirements, enabling precise adjustment of flexible resistance and improving the system's adaptability and controllability in complex environments.

[0092] This embodiment achieves precise control of the resistance characteristics of the servo system through the implementation of the flexible resistance adjustment servo control method, thereby improving the system's adaptability under complex working conditions. The position and speed parameters of the servo motor data are separated to obtain high-precision angular position and angular velocity information, laying a data foundation for subsequent compensation. The phase synchronization technology based on the rotor angular position signal ensures that the compensation torque acts at the optimal time, significantly improving the smoothness of the system response. The introduction of the speed coupling mechanism enables the compensation torque to be adaptively adjusted with the speed. Through the linear superposition of the speed-adaptive compensation component and the feedforward reference coefficient, a comprehensive anti-disturbance compensation strategy is constructed, which enhances the system's ability to resist external interference. The final instruction reconstruction integrates basic motion control and anti-disturbance compensation requirements, allowing the system to exhibit excellent flexible resistance characteristics and improve control accuracy and operating experience.

[0093] In one embodiment, velocity coupling is performed on the phase compensation torque vector according to the angular velocity component to obtain a velocity-adaptive compensation component, including: The angular velocity components contain information about the internal resistance characteristics of flexible materials during dynamic processes. These characteristics are quantified into viscoelastic hysteresis damping parameters through parameter extraction. In this step, the angular velocity components of the motor rotor reflect the response behavior of the flexible connector at different deformation rates. The internal resistance parameter extraction process is based on time-domain analysis of the angular velocity components, capturing the corresponding relationship between velocity variation trends and damping effects. The angular velocity data is filtered to eliminate high-frequency noise interference and retain valid dynamic characteristics.

[0094] The extraction process focuses on the rate of change of angular velocity, which reflects the rate of structural rearrangement of the flexible material's molecular chains and is directly related to the material's viscoelastic behavior. The viscoelastic hysteresis damping parameter consists of two main components: the static damping coefficient and the dynamic hysteresis factor. The static damping coefficient describes the material's resistance at steady speed, while the dynamic hysteresis factor characterizes the material's hysteresis effect during velocity changes.

[0095] The parameter extraction process uses a segmented approach to independently identify material properties for different speed ranges, capturing nonlinear characteristics. The periodic variation pattern of angular velocity is used to identify the material's fatigue properties under cyclic loading, which are expressed using a dynamic memory factor. The extracted viscoelastic hysteresis damping parameters are normalized to ensure consistency across different operating conditions. The viscoelastic hysteresis damping parameters output from this step provide the foundational data for subsequent hysteresis nonlinearity compensation, ensuring that the compensation torque matches the material's actual internal resistance characteristics.

[0096] Flexible materials exhibit significant hysteretic nonlinearity. Hysteretic nonlinearity compensation is used to enable the phase compensation torque vector to more accurately reflect actual material behavior. In this step, viscoelastic hysteresis damping parameters are combined with the phase compensation torque vector to construct a hysteresis compensation model. The compensation process is based on the torque-displacement hysteresis curve, which describes the asymmetric response characteristics of flexible materials during loading and unloading. Hysteretic nonlinearity compensation introduces a memory term, making the current compensation value dependent not only on the current state but also on the historical state, thus simulating the material's memory effect.

[0097] The compensation process distinguishes between the rising and falling torque phases, employing different compensation strategies for each phase to accurately restore the hysteresis loop characteristics. The phase-compensated torque vector maintains its angular correlation during the compensation process while also adding velocity history dependency to form a two-dimensional mapping relationship. The base value for material internal resistance compensation includes both static and dynamic components. The static component corresponds to the material's equilibrium resistance, while the dynamic component reflects the additional resistance during deformation. The compensation process considers the material's elastic limit, adjusting the compensation strategy as the limit is approached to prevent model failure.

[0098] Hysteresis compensation uses a progressive iterative approach, approximating realistic material behavior through multiple iterations to improve model accuracy. The base value for material internal resistance compensation is amplitude-calibrated to ensure that the compensation torque amplitude is commensurate with the actual internal resistance. The base value for material internal resistance compensation output in this step accurately reflects the nonlinear internal resistance characteristics of the flexible material, providing foundational data for subsequent rate coupling.

[0099] The internal resistance characteristics of flexible materials are closely related to the deformation rate, and the modulation effect of the rate on the internal resistance is obtained through deformation rate analysis. In this step, the angular velocity component is regarded as an indirect representation of the deformation rate of the flexible connector, and is converted into the actual deformation rate through a geometric mapping relationship. The deformation rate analysis process takes into account the installation configuration and geometric characteristics of the flexible connector, and converts angular motion into linear deformation. The analysis process introduces the material stiffness-strain relationship and considers the nonlinear change of stiffness under different deformation amounts, so that the analysis results are more consistent with the actual material behavior. The deformation rate analysis distinguishes between pure elastic deformation regions and viscoelastic deformation regions, and adopts corresponding rate analysis models for different regions.

[0100] The analytical process focuses on the directionality of the deformation rate, distinguishes between tensile and compressive states, and captures the heterogeneous response of the material in different deformation directions. The deformation rate coupling factor is expressed as a nonlinear function of the rate, showing high sensitivity in the low-speed region and tending to saturation in the high-speed region, reflecting the friction and entanglement characteristics of the molecular chains within the material at different rates. The coupling factor calculation adopts a piecewise continuous method to ensure a smooth transition during the rate change process. The deformation rate analysis also takes into account the multi-axis coupling effect, comprehensively considering the deformation contribution of the primary and secondary axis directions to form a three-dimensional coupling model. The analytical results are normalized so that the coupling factor has consistent physical meaning under different working conditions.

[0101] The viscoelastic work equivalent conversion establishes a corresponding relationship between the energy dissipation and the compensation torque during the material deformation process, thereby achieving accurate compensation of the energy angle. In this step, the deformation rate coupling factor and the material internal resistance compensation base value are associated through the energy equivalence principle to calculate the equivalent internal resistance compensation torque. The conversion process is based on the power balance equation, which equates the energy dissipation rate during deformation to the product of the compensation torque and the angular velocity. The viscoelastic work equivalent takes into account the energy storage and release characteristics of the material during deformation, and distinguishes between elastic energy and dissipated energy. Elastic energy corresponds to recoverable energy, which is compensated by elastic torque; dissipated energy corresponds to irrecoverable energy loss, which is compensated by damping torque.

[0102] The conversion process introduces an energy transfer function to describe the dynamic transformation relationship between energy storage and dissipation states. The internal resistance compensation torque data contains phase-related components and rate-related components, which are comprehensively balanced by the energy weight coefficient. The conversion algorithm takes into account the time lag effect of energy transfer and introduces a time delay term in the torque calculation to simulate the process delay from macroscopic deformation to microscopic dissipation of energy. During the viscoelastic work equivalent conversion process, the system dynamically evaluates the energy dissipation efficiency and adjusts the conversion parameters according to the current state of the material to adapt to the characteristic drift caused by material fatigue and temperature changes. The conversion result is checked for energy consistency to ensure that the power generated by the compensation torque is equivalent to the actual material dissipation power. The internal resistance compensation torque data output by this step accurately simulates the internal resistance characteristics of the flexible material from an energy perspective, providing an energy basis for subsequent time-temperature equivalence corrections.

[0103] The characteristics of flexible materials are significantly affected by time and temperature, and the compensation torque is adapted to the actual working environment through time-temperature equivalence correction. In this step, the preset characteristic data of the flexible connector of the equipment contains the performance parameters of the material under different time scales and temperature conditions. These data usually come from material testing or historical operation records. The time-temperature equivalence correction is based on the principle of time-temperature equivalence, which unifies the influence of time and temperature into an equivalent parameter for processing. The correction process takes into account the aging effect of the material. As the service time increases, the stiffness and damping characteristics of the flexible material change. These changes are quantified by the time correction factor. The influence of temperature on the flexible material is manifested as softening of stiffness and change of viscosity. This influence is compensated by the temperature correction factor.

[0104] The correction algorithm employs a hierarchical structure, first performing time-scale correction, then temperature correction, and finally combining the two to produce a comprehensive correction effect. During the time-temperature equivalent correction process, the system estimates the temperature state of the flexible connector in real time and calculates the actual operating temperature of the material based on the ambient temperature, operating load, and heat conduction model. The correction process also takes into account uneven temperature distribution, adopting a zoned correction strategy for different temperature zones. The corrected speed-adaptive compensation component is environmentally adaptable, maintaining consistent compensation results across different temperatures and usage stages.

[0105] Time-temperature equivalent correction also incorporates a predictive compensation mechanism. This predicts short-term changes in material properties based on temperature trends, allowing for proactive adjustment of compensation parameters to improve system responsiveness. The velocity-adaptive compensation component output from this step comprehensively considers the dynamic, energy, and environmental adaptability of the flexible material, enabling precise adjustment of flexible resistance and improving control quality and consistency in complex environments.

[0106] This embodiment extracts the internal resistance parameters of the flexible material from the angular velocity component, thereby achieving accurate quantification of the viscoelastic hysteresis damping characteristics of the material, providing a reliable basis for subsequent compensation. Hysteresis nonlinear compensation is performed on the phase compensation torque vector based on the extracted parameters, effectively overcoming the problem of insufficient compensation caused by ignoring the material memory effect in traditional control methods. The flexible deformation rate analysis technology can accurately capture the heterogeneous response of the material at different deformation rates, significantly improving the system's adaptability to dynamic working conditions. The viscoelastic work equivalent conversion method establishes a correspondence between the deformation process and the compensation torque from an energy perspective, making the compensation more accurate and the physical meaning clear. The time-temperature equivalent correction mechanism effectively addresses the impact of ambient temperature changes and material aging on system performance, ensuring long-term stable operation. The overall solution achieves precise adjustment of the flexible resistance, improves the control accuracy and consistency of the system under complex working conditions, and provides a higher quality control experience for the flexible mechanical system.

[0107] Reference Figure 2 As shown, the present invention also provides a device flexibility resistance adjustment servo control system, which is applied to any of the above-mentioned device flexibility resistance adjustment servo control methods, including: The acquisition module is used to obtain the resistance torque data and periodic control instructions of the target device, perform disturbance parameter expansion estimation, and obtain the system total disturbance torque data and load equivalent stiffness information; An analysis module is used to obtain the servo motor data of the target device, perform multiple feedforward compensation on the total disturbance torque data of the system, obtain anti-disturbance feedforward compensation data, and perform stiffness reference adjustment in combination with the load equivalent stiffness information to obtain the feedforward reference coefficient; An association module is used to couple torque parameters with anti-disturbance feedforward compensation data according to a feedforward reference coefficient to obtain stiffness compensation torque data; The processing module is used to construct instructions according to the servo motor data, the feedforward reference coefficient and the stiffness compensation torque data to obtain the servo control instructions.

[0108] The present invention provides a servo control system for flexible resistance regulation of equipment. This system obtains total disturbance torque data and load equivalent stiffness information through disturbance parameter expansion estimation, enabling more accurate assessment of the system's dynamic characteristics under complex operating conditions. This improves servo control precision and provides a more reliable foundation for stable equipment operation. Multiple feedforward compensation techniques are employed, combined with load equivalent stiffness information for stiffness baseline adjustment, enabling refined adaptation to varying load characteristics. This helps the system maintain stable response under varying operating conditions and avoids over- or under-control. Torque parameter coupling with anti-disturbance feedforward compensation data based on feedforward baseline coefficients ensures efficient system operation across various operating states and load conditions, reduces control delays and oscillations, and improves the system's overall dynamic performance. By comprehensively analyzing servo motor data and stiffness compensation torque data, more reasonable servo control instructions are constructed. An intelligent adjustment strategy is then used to optimize the operation of the entire system, effectively improving response speed and reducing energy consumption and mechanical wear. By considering the periodic control instructions of the target equipment, the control strategy can be flexibly adjusted according to the characteristics and changing requirements of different application scenarios, making the system more adaptable to diverse industrial environments and task requirements.

[0109] It should be noted that, those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the above-described system and each module can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0110] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A servo control method for adjusting flexible resistance of equipment, characterized in that: include: Obtain the resistance torque data and periodic control instructions of the target equipment, perform disturbance parameter expansion estimation, and obtain the system's total disturbance torque data and load equivalent stiffness information; Acquiring servo motor data of the target device, performing multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and performing stiffness reference adjustment in combination with the load equivalent stiffness information to obtain a feedforward reference coefficient; performing torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data; Instructions are constructed according to the servo motor data, the feedforward reference coefficient and the stiffness compensation torque data to obtain a servo control instruction.

2. The device flexible resistance adjustment servo control method according to claim 1, characterized in that: The method of obtaining the resistance torque data and periodic control instructions of the target device, performing disturbance parameter expansion estimation, and obtaining the total disturbance torque of the system and load equivalent stiffness information includes: Performing motor torque acquisition and periodic signal monitoring on the target device to obtain the resistance torque data and the periodic control instruction; Constructing a disturbance observer model based on the resistance torque data and the periodic control instruction to perform multi-dimensional state variable expansion to obtain an initial value of the total disturbance torque of the system; Performing load-end dynamic identification and parameter convergence according to the initial value of the system total disturbance torque and the periodic control instruction to obtain the load equivalent stiffness information; An iterative mapping correction is performed on the initial value of the total disturbance torque of the system and the load equivalent stiffness information to obtain the total disturbance torque data of the system.

3. The device flexible resistance adjustment servo control method according to claim 1, characterized in that: The step of acquiring the servo motor data of the target device, performing multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and performing stiffness reference adjustment in combination with the load equivalent stiffness information to obtain a feedforward reference coefficient includes: Sampling and incrementally decoding the motor encoding signal of the target device to obtain the servo motor data; Performing characteristic decoupling and separation on the total disturbance torque of the system to obtain a low-frequency disturbance torque component and a high-frequency resonance suppression component; Performing phase advance compensation on the low-frequency interference torque component according to the servo motor data to obtain the anti-disturbance feedforward compensation data; Performing energy coupling operation on the high-frequency resonance suppression component according to the servo motor data to obtain a resonance frequency adjustment amount; A stiffness reference damping mapping is performed according to the resonant frequency adjustment amount and the load equivalent stiffness information to obtain the feedforward reference coefficient.

4. The device flexible resistance adjustment servo control method according to claim 3, characterized in that: The performing of energy coupling operation on the high-frequency resonance suppression component according to the servo motor data to obtain a resonance frequency adjustment amount includes: Performing real-time load inertia identification processing on the servo motor data to obtain an equivalent load inertia change rate; Performing stiffness-inertia matching calculation on the equivalent load inertia change rate and the load equivalent stiffness information to generate a stiffness attenuation factor; performing dynamic notch frequency band adjustment processing on the high-frequency resonance suppression component according to the stiffness attenuation factor to obtain a resonance suppression amount; Calculate the natural frequency according to the servo motor data to obtain the load resonant frequency characteristic value; The resonance suppression amount and the load resonance frequency characteristic value are subjected to dominant frequency energy extraction to obtain the resonance frequency adjustment amount.

5. The device flexible resistance adjustment servo control method according to claim 1, characterized in that: The step of performing torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data includes: Decomposing the feedforward reference coefficient in terms of flexible stiffness dimension to obtain an axial stiffness gain parameter and a radial stiffness compensation coefficient; Performing deformation direction correction on the anti-disturbance feedforward compensation data according to the radial stiffness compensation coefficient to obtain a deformation suppression compensation torque; Performing spatial trajectory calculation on the servo motor data to obtain flexible load motion data; Performing a spatial compensation operation on the axial stiffness gain parameter and the flexible load motion data to obtain an axial stiffness compensation amount; The deformation suppression compensation torque and the axial stiffness compensation amount are subjected to nonlinear resistance gradient fitting to obtain the stiffness compensation torque data.

6. The device flexible resistance adjustment servo control method according to claim 5, characterized in that: The performing spatial trajectory calculation on the servo motor data to obtain flexible load motion data includes: Performing flexible deformation friction analysis on the servo motor data to obtain a rotor friction vector; Performing rigid-flexible transmission coupling on the target device according to the rotor friction vector to obtain the load end trajectory coordinates; Calculating the resistance gradient of the load end trajectory coordinates according to the feedforward reference coefficient to obtain a deformation induced resistance gradient; Performing reverse motion conversion on the deformation-induced resistance gradient to obtain a resistance-compensated deformation displacement; The load end trajectory coordinates and the resistance compensation deformation displacement are fused in servo space to output the flexible load motion data.

7. The device flexible resistance adjustment servo control method according to claim 1, characterized in that: The step of constructing an instruction according to the servo motor data, the feedforward reference coefficient, and the stiffness compensation torque data to obtain a servo control instruction includes: Separating the position and speed parameters of the servo motor data to obtain a rotor angular position signal and a rotor angular velocity component; performing phase synchronization on the stiffness compensation torque data according to the rotor angular position signal to obtain a phase compensation torque vector; Performing velocity coupling on the phase compensation torque vector according to the angular velocity component to obtain a velocity-adaptive compensation component; Linearly superimposing the speed adaptive compensation component and the feedforward reference coefficient to obtain anti-disturbance compensation synthetic torque data; The periodic control instruction is reconstructed according to the anti-disturbance compensation synthetic torque data to obtain the servo control instruction.

8. The device flexible resistance adjustment servo control method according to claim 7, characterized in that: The performing speed coupling on the phase compensation torque vector according to the angular velocity component to obtain a speed-adaptive compensation component includes: Extracting the internal resistance parameter of the flexible material from the angular velocity component to obtain the viscoelastic hysteresis damping parameter; Performing hysteresis nonlinear compensation on the phase compensation torque vector according to the viscoelastic hysteresis damping parameter to obtain a material internal resistance compensation base value; Performing flexible deformation rate analysis on the angular velocity component to obtain a deformation rate coupling factor; Performing viscoelastic work equivalent conversion on the deformation rate coupling factor and the material internal resistance compensation base value to obtain internal resistance compensation torque data; The internal resistance compensation torque data is corrected with time and temperature equivalent according to preset characteristic data of the equipment flexible connector to obtain the speed-adaptive compensation component.

9. A servo control system for adjusting flexible resistance of equipment, characterized in that: The device flexible resistance adjustment servo control method applied to any one of claims 1 to 8 above comprises: An acquisition module is used to obtain the resistance torque data and periodic control instructions of the target device, perform disturbance parameter expansion estimation, and obtain the system total disturbance torque data and load equivalent stiffness information; an analysis module configured to obtain servo motor data of the target device, perform multiple feedforward compensation on the total disturbance torque data of the system to obtain anti-disturbance feedforward compensation data, and perform stiffness reference adjustment based on the load equivalent stiffness information to obtain a feedforward reference coefficient; an association module, configured to perform torque parameter coupling on the anti-disturbance feedforward compensation data according to the feedforward reference coefficient to obtain stiffness compensation torque data; A processing module is used to construct an instruction according to the servo motor data, the feedforward reference coefficient and the stiffness compensation torque data to obtain a servo control instruction.

Images