Independent rotation and dynamic balance control method for large-size floating hub cover

Abstract

The application discloses a large-size suspension hub cover independent rotation maintaining and dynamic balance control method, and belongs to the technical field of automobile part control. The method comprises the following steps: continuously sampling the current rotation angle of the hub cover to generate a rotation phase sequence; establishing a rotation phase deviation function according to the difference between the rotation phase sequence and a reference rotation phase, and mapping to a radial displacement field to generate a radial displacement field distribution diagram; driving a first group of actuators in a suspension actuator array by using the radial displacement field distribution diagram to generate a radial compensation force; simultaneously monitoring an axial tilt angle in real time to generate an axial tilt angle sequence; performing difference processing with a balance tilt angle to generate an axial tilt deviation amount; adjusting a second group of actuators according to the axial tilt deviation amount to generate an axial compensation torque; and fusing the radial compensation force and the axial compensation torque vector to form a three-dimensional dynamic balance control instruction. The application realizes the coordinated dynamic regulation and control of the rotation and axial posture of the hub cover, and improves the balance precision and stability of the suspension system.

Description

Technical Field

[0001] This invention relates to the field of automotive component control technology, specifically to a method for independent rotation holding and dynamic balance control of large-size floating hubcaps. Background Technology

[0002] Large-sized floating hubcaps need to maintain an independent rotational state relative to the hub center during vehicle operation and maintain spatial stability under dynamic conditions. Existing technologies often employ simple displacement closed-loop adjustment methods for hubcap suspension control. This method counteracts eccentric displacement of the hubcap by individually detecting the suspension air gap and adjusting the combined force of the electromagnetic actuator. This approach treats the hubcap as a rigid body, using a preset fixed gap value as the control target. When a deviation is detected, the electromagnetic force in the corresponding direction is uniformly increased or decreased.

[0003] However, large-sized hubcaps are subjected to centrifugal loads and road surface excitation during rotation, causing non-uniform radial deformation of their structure. This deformation is not a uniform rigid body displacement, but rather a complex distribution along the circumference. Single displacement feedback control cannot detect and compensate for this local morphological change caused by rotational phase differences, resulting in significant radial runout of the hubcap during high-speed rotation and a substantial decrease in suspension accuracy. Furthermore, existing control strategies treat radial displacement compensation and axial tilt compensation as two independent loops, neglecting the spatial coupling relationship between radial force and axial moment. When the vehicle tilts or passes through curves, the radial compensation force and axial compensation moment interfere with each other, causing control oscillations and making it difficult to achieve coordinated and stable three-dimensional attitude of the hubcap.

[0004] The correlation between radial deformation and rotation phase of the hubcap, as well as the coupling interference between radial force and axial torque in dynamic control, are difficult problems that existing technologies have not been able to effectively solve. Summary of the Invention

[0005] This invention provides an independent rotation holding and dynamic balance control method for large-size floating hubcaps. The purpose is to solve the problem in the prior art that it is impossible to compensate for non-uniform radial deformation of hubcaps based on rotation phase differences, and at the same time solve the problem of mutual coupling and interference between radial compensation force and axial compensation torque in dynamic control.

[0006] To achieve the above objectives, the present invention provides the following technical solution: To solve the dynamic instability problem caused by rotational phase deviation and axial tilt during vehicle operation of large-size floating hubcaps, the present invention provides an independent rotational maintenance and dynamic balance control method for large-size floating hubcaps. This method continuously samples the current rotation angle of the hubcap to generate a rotational phase sequence. Based on the difference between this rotational phase sequence and a preset reference rotational phase, a rotational phase deviation function is established. This rotational phase deviation function is mapped to the radial displacement field of the hubcap to obtain a radial displacement field distribution map. The radial displacement field distribution map is used to drive the first set of actuators in the floating actuator array to generate a first radial compensation force. This precisely corrects the radial deformation of the hubcap caused by the phase deviation, allowing the hubcap to maintain an independent rotational state during rotation without being disturbed by the vibration of the hub body. Simultaneously, the axial tilt angle of the hubcap is monitored in real time, and an axial tilt angle sequence is generated. The axial tilt angle sequence is differentially processed with a preset balance tilt angle to obtain the axial tilt deviation. Based on the axial tilt deviation, the second set of actuators in the suspension actuator array is adjusted to generate a second axial compensation torque. By actively compensating for the axial tilt, the swaying of the hubcap is suppressed, improving suspension stability and attitude maintenance accuracy. The first radial compensation force and the second axial compensation torque are vector-fused to form a three-dimensional dynamic balance control command, so as to unify and coordinate the distribution of radial and axial suspension forces, and realize the synchronous control of the hubcap's position and posture in three-dimensional space.

[0007] As a preferred technical solution of the present invention, when establishing the rotating phase deviation function, the rotating phase sequence is expanded by Fourier series to extract the fundamental frequency phase component and several harmonic phase components. The first-order deviation value between the fundamental frequency phase component and the preset reference rotating phase, and the second-order deviation value between each harmonic phase component and the corresponding preset harmonic reference phase are calculated respectively. The first-order deviation value and the second-order deviation value are weighted and summed according to the frequency weighting coefficient, so that the deviation function can fully reflect the influence of different frequency components on the rotation accuracy, thereby improving the accuracy of phase deviation estimation and the pertinence of control response.

[0008] As a further improvement of the present invention, the process of mapping the rotational phase deviation function to the radial displacement field includes: discretizing the rotational phase deviation function along the circumferential direction of the hubcap to obtain the phase deviation scalar at each circumferentially divided point; pre-establishing a three-dimensional topography database of the radial deformation of the hubcap under different phase deviation values; obtaining the corresponding radial deformation set by querying the phase deviation scalar; and then performing spatial interpolation to generate a continuous radial displacement field distribution map. Preferably, the three-dimensional topography database is pre-calculated using finite element simulation to calculate the static and dynamic radial deformation of the hubcap under different phase deviation values ​​and stored in a lookup table, thereby enabling rapid retrieval during real-time control, reducing the online calculation burden, and improving system real-time performance.

[0009] When driving the first set of actuators, the gradient direction of the radial displacement field distribution map is used as the force direction, and its amplitude value is extracted as the force amplitude reference. Hysteresis correction is then applied to this force amplitude reference based on a pre-stored correction table of the hysteresis curves of the first set of actuator materials. This generates a corrected force amplitude, and the force direction and corrected force amplitude are then encoded into a pulse width modulation signal and sent to the drive circuit. This compensates for the hysteresis nonlinearity of the actuator materials, making the actual output radial compensation force more accurate and enhancing the correction effect on the radial displacement of the hubcap.

[0010] Real-time monitoring of the axial tilt angle is preferably achieved by evenly distributing four laser rangefinders along the edge of the hubcap. These rangefinders measure the instantaneous distance from the hubcap edge to the hub end face. The difference between the instantaneous distances measured by each pair of laser rangefinders at relative positions is calculated to obtain a first difference vector and a second difference vector. These vectors are then cross-productd to generate an axial tilt angle sequence containing components around the horizontal and vertical axes. This method improves the resolution of minute tilt angles through non-contact measurement and can quickly capture changes in the hubcap's attitude.

[0011] When adjusting the second set of actuators, the axial tilt deviation is decomposed into a first deviation component around the horizontal axis and a second deviation component around the vertical axis. The expected magnitude and direction of the first and second set of compensation torques are calculated separately. The corresponding thrust parameters of each actuator are obtained by looking up a table using pre-stored actuator thrust-current curves, and then combined to generate the drive signal for the second axial compensation torque. Through torque decomposition and table lookup, the distribution of the axial compensation torque is more precise, effectively suppressing the coupling effects between multiple degrees of freedom.

[0012] When vector-fusing the first radial compensation force and the second axial compensation torque, the first radial compensation force is decomposed into X-direction and Y-direction components along the radial coordinate system, and the second axial compensation torque is scalarized along the Z-direction of the same coordinate system. An initial fused control vector is synthesized according to the three-dimensional vector addition rule, and then subjected to amplitude-limiting filtering to generate a three-dimensional dynamic balance control command. The upper and lower limits of the amplitude limit are set according to the physical maximum output value of the suspended actuator array. This vector fusion control method achieves coordinated output of radial and axial control quantities while ensuring that the control of each degree of freedom does not conflict with each other.

[0013] After generating the three-dimensional dynamic balance control command, this invention also sends the control command to the drive control unit of the suspended actuator array. Simultaneously, it collects the actual output force feedback signal of each actuator, compares it dimension-by-dimensionally with the desired output force to generate a force deviation vector, and inputs this force deviation vector into the integral compensation stage for accumulation to obtain the integral correction amount. This correction amount is then superimposed on the three-dimensional dynamic balance control command to form an updated control command. Furthermore, the integral gain of the integral compensation stage is adaptively adjusted according to the magnitude of the force deviation vector. When the magnitude is greater than a preset threshold, the integral gain is increased to accelerate convergence; otherwise, the integral gain is decreased to prevent overshoot. Through this closed-loop force feedback and adaptive integral compensation, the effects of actuator output errors and external disturbances can be eliminated, ensuring that the hubcap maintains high-precision suspension balance and independent rotation under complex driving conditions.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: By establishing a rotational phase deviation function and mapping it to the radial displacement field of the hubcap, a radial displacement field distribution map is generated, which in turn drives the first set of actuators to produce a first radial compensation force associated with the rotational phase. This method introduces rotational phase information into the radial compensation process, ensuring that the compensation force applied by the actuators corresponds to the actual radial deformation morphology of the hubcap at different rotation angles, rather than treating the hubcap as a single rigid body for overall displacement adjustment. When the hubcap undergoes non-uniform radial deformation due to centrifugal force and road load, the radial displacement field distribution map reflects the local deformation differences in the circumferential direction. Based on this, the first set of actuators outputs differentiated compensation forces at corresponding positions, enabling precise suppression of radial runout synchronized with the rotational phase.

[0015] By decomposing and calculating the axial tilt deviation, the second set of actuators is adjusted to generate a second axial compensation torque. This axial compensation torque is then vector-fused with the components of the first radial compensation force along the radial coordinate system to form a three-dimensional dynamic balance control command. This method synthesizes the compensation quantities of both radial and axial degrees of freedom in the same three-dimensional vector space. When dealing with axial tilt, the generation of the compensation torque and the direction component of the radial compensation force are calculated collaboratively, avoiding spatial contradictions in the control commands generated when the two compensation loops operate independently. During vehicle tilting or turning, the adjustment of radial force and axial torque no longer interferes with each other, and the three-dimensional control command can coordinate the drive actuator array to maintain the stability of the wheel hub cover's spatial attitude. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the method for independent rotation maintenance and dynamic balance control of large-size floating hubcaps; Figure 2 This is a schematic diagram illustrating the principle of real-time monitoring of the axial tilt angle of the hubcap; Figure 3 It is a time-domain curve of the rotating phase component and the rotating phase deviation function; Figure 4 This is a radial displacement distribution diagram corresponding to the rotational phase deviation function of the hubcap; Figure 5 It is a time-domain response diagram of three-dimensional dynamic balance control commands. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figure 1 This invention provides an independent rotation holding and dynamic balance control method for a large-size suspended hubcap. The method includes: continuously sampling the current rotation angle of the hubcap to generate a rotation phase sequence; establishing a rotation phase deviation function based on the difference between the rotation phase sequence and a preset reference rotation phase; mapping the rotation phase deviation function to the radial displacement field of the hubcap to generate a radial displacement field distribution map; using the radial displacement field distribution map to drive a first group of actuators in a suspended actuator array to generate a first radial compensation force; simultaneously, real-time monitoring of the axial tilt angle of the hubcap to generate an axial tilt angle sequence; differential processing of the axial tilt angle sequence with a preset balance tilt angle to generate an axial tilt deviation; adjusting a second group of actuators in the suspended actuator array according to the axial tilt deviation to generate a second axial compensation torque; and vector fusion of the first radial compensation force and the second axial compensation torque to form a three-dimensional dynamic balance control command.

[0020] Example 1 In practical implementation, the process of establishing the rotating phase deviation function includes: performing a Fourier series expansion on the rotating phase sequence to extract the fundamental frequency phase component and several harmonic phase components; calculating the first-order deviation value between the fundamental frequency phase component and a preset reference rotating phase, and the second-order deviation value between each harmonic phase component and its corresponding preset harmonic reference phase; and weighting the first-order and second-order deviation values ​​according to frequency weighting coefficients to obtain the rotating phase deviation function. The frequency weighting coefficients are pre-calibrated based on the sensitivity of the fundamental frequency and each harmonic to the rotational stability of the hub cover and are stored in the coefficient table of the control unit.

[0021] In practical implementation, the rotation phase sequence is a set of discrete angle values ​​obtained by continuously collecting the current rotation angle of the hub cover from the angle sensor at a fixed sampling period, represented as: ,in Indicates the first Each sampling time point. When performing a Fourier series expansion on the rotated phase sequence, a discrete Fourier transform algorithm is used to transform the time-domain rotated phase sequence to the frequency domain, obtaining the amplitude spectrum and phase spectrum. The frequency component with the largest amplitude is identified from the amplitude spectrum as the fundamental frequency, and the corresponding phase value of this fundamental frequency in the phase spectrum is read as the fundamental frequency phase component. Simultaneously, according to integer multiples of the fundamental frequency, the second harmonic frequency, the third harmonic frequency, and so on, are sequentially read from the phase spectrum. The phase values ​​corresponding to the subharmonic frequencies are respectively taken as the phase components of each subharmonic. , For harmonic order, , The preset maximum harmonic order, The value of is determined based on the mechanical structural resonance characteristics of the hubcap suspension system, selecting a range of significant harmonic frequencies that may be excited. In one implementation, Set it to 5.

[0022] In practical implementation, the preset reference rotation phase is a standard phase trajectory determined by the ideal uniform rotation motion model of the hub cover, expressed as: Fundamental frequency phase component The reference rotation phase value at the corresponding sampling time is denoted as Calculate the first-order deviation value This serves as the fundamental frequency phase deviation. Similarly, for each harmonic phase component... Each has a corresponding preset harmonic reference phase. These preset harmonic reference phases are ideally zero phase, and the second-order deviation value is calculated. , which serves as the phase deviation of each harmonic.

[0023] In practice, the frequency weighting coefficient includes the fundamental frequency weighting coefficient. and the weighting coefficients of each harmonic , The calibration process for the frequency weighting coefficient is as follows: On the hubcap test bench, a unit amount of fundamental frequency phase deviation and phase deviations of each harmonic are applied separately. The radial runout of the hubcap edge is measured using a laser vibrometer. The rate of change of the radial runout relative to the phase deviation is used as the sensitivity index. The sensitivity index corresponding to the fundamental frequency phase deviation is marked as follows: , No. The sensitivity index corresponding to the subharmonic phase deviation is: Then the fundamental frequency weighting coefficient Harmonic weighting coefficients This ensures that the sum of all weight coefficients is 1. After calibration, and The coefficients are stored in the control unit's coefficient table for use in real-time control processes.

[0024] In practical implementation, the rotation phase deviation function Calculated using the following formula: in, The numerical output of the rotation phase deviation function is a scalar. The fundamental frequency weighting coefficient is a dimensionless constant with a value range of 0 to 1. It is obtained by the sensitivity calibration method described above and stored in the coefficient table. This represents the first-order deviation of the fundamental frequency phase component from the preset reference rotating phase, expressed in radians. Indicates the first The subharmonic weighting coefficient is a dimensionless constant with a value range of 0 to 1. It is also obtained and stored through calibration. Indicates the first The second-order deviation of the subharmonic phase component from the corresponding preset harmonic reference phase is expressed in radians. This indicates the preset maximum harmonic order, which is an integer greater than or equal to 2.

[0025] In practice, the control unit performs the above calculations in each control cycle, calculating the current rotation phase deviation function based on the latest sampled rotation phase sequence. The rotational phase deviation function is then used to map onto the radial displacement field of the hubcap to drive the first set of actuators to generate a first radial compensation force.

[0026] See Figure 3In the figure, the vertical axis represents phase deviation in radians, and the horizontal axis represents time in seconds, ranging from 0 to 0.5 seconds. The legend shows three curves representing the fundamental frequency phase component (solid blue line), the second harmonic phase component (dashed orange line), and the rotating phase deviation function (dotted green line). This figure corresponds to the process in Example 1 of performing a Fourier series expansion of the rotating phase sequence and calculating the rotating phase deviation function.

[0027] The blue curve represents the fundamental frequency phase component, exhibiting a sinusoidal waveform with the largest amplitude and a period of approximately 0.1 seconds. Its peak amplitude is approximately ±0.02 radians, reflecting the dominant role of the fundamental frequency component in the frequency domain. The orange dashed line represents the second harmonic phase component, with a period approximately half that of the fundamental frequency. Its amplitude is smaller, with a peak amplitude of approximately ±0.008 radians, indicating a minor contribution of the second harmonic to the overall phase deviation. The green dotted line represents the rotating phase deviation function, whose coefficients are obtained by multiplying the fundamental frequency and second harmonic phase deviations by their respective weighting coefficients and then summing them. The curve trend is similar to that of the fundamental frequency phase component, but the amplitude is adjusted to approximately ±0.013 radians, falling between the fundamental frequency and the second harmonic, indicating that the rotating phase deviation function effectively integrates the influence of various frequency components.

[0028] From the overall trend of the curves, all three curves exhibit stable periodic changes, consistent with the Fourier expansion characteristics of the fundamental frequency as the dominant frequency, supplemented by subharmonics, as described in Example 1. This indicates that the phase sampling data and Fourier transform operation correctly reflect the rotational state of the hubcap. The smooth waveform of the rotational phase deviation function shows that the weighted processing effectively suppresses the interference of high-order harmonic noise on the control commands, which is beneficial for the smooth drive subsequently mapped to the radial displacement field.

[0029] Example 2 In a specific implementation, the rotational phase deviation function is mapped to the radial displacement field of the hub cover to generate a radial displacement field distribution map. The radial displacement field distribution map is then used to drive the first group of actuators in the suspended actuator array to generate the first radial compensation force.

[0030] In practical implementation, when discretizing the rotational phase deviation function along the circumferential direction of the hubcap, the circumference of the hubcap is divided into M equal angular intervals, where M is an integer power of 2. In one implementation, M is set to 256 to balance angular resolution and real-time computation. The circumferential angular positions corresponding to the M equally divided points are denoted as... ,in Rotate the phase deviation function Projected along the radial axis to each circumferential angular position The phase deviation scalar at each equally divided point of the circumference is calculated using the radial projection function. The formula for the radial projection function is expressed as: in, Indicates the first The phase deviation scalar at each of the circumference's equal division points, in radians; The value of the rotating phase deviation function is obtained by weighting the fundamental frequency phase component and each harmonic phase component using a frequency weighting coefficient, and the unit is radians. Indicates the first The circumference is divided into three equal parts, each corresponding to a circumference angle, in radians. ; This represents the reference phase offset angle, corresponding to the circumferential angle at which the hubcap's rotational phase deviation function reaches its maximum value under ideal conditions. It is calibrated through a single-point radial force excitation test of the hubcap in a static, suspended state, and is expressed in radians. It should be noted that this is a phase deviation scalar... As an index variable for subsequent queries of the 3D topography database, the index field of the 3D topography database also uses radians as the unit, and... The units are completely consistent, thus allowing direct use of... The corresponding radial deformation amount can be obtained by referring to the table.

[0031] In practice, the process of establishing the three-dimensional topography database is as follows: A geometric model and material property model of the hubcap are constructed using finite element simulation software. Radial load distributions under different phase deviation values ​​are applied to the hubcap in the finite element simulation software, and the static and dynamic radial deformations of the hubcap are calculated. The static radial deformation is calculated under constant load conditions, while the dynamic radial deformation is calculated under conditions where the load alternates with the hubcap's rotation frequency. The set of radial deformations corresponding to each phase deviation value is stored in a lookup table, with the index of the lookup table being the phase deviation scalar. The lookup table output is the radial deformation vector corresponding to the phase deviation scalar. The radial deformation vector includes the radial displacement components of each node on the hubcap surface.

[0032] In practical implementation, when calling the 3D topography database, the phase deviation scalar at each equally divided point on the circumference is used. As the input index of the lookup table, the corresponding radial deformation vector is queried one by one. This yields the set of radial deformation values ​​at each equally divided point of the circumference.

[0033] In practical implementation, when performing spatial interpolation on the radial deformation set, a radial basis function interpolation algorithm is used to generate a continuous radial displacement field distribution map. The radial basis function interpolation algorithm takes the form of spline-type radial basis functions, and the interpolation formula is: in, This represents a vector at any position on the surface of the hubcap. The radial displacement field value at the location, in meters; This represents the position vector of any point on the surface of the hubcap, with the geometric center of the hubcap as the origin of the coordinate system. Indicates the first The position vectors of the points on the surface of the hubcap at the points where the circumference is equally divided; Represents position vector With position vector The Euclidean distance between them; where, Indicates the first The interpolation weight coefficients corresponding to the points where the circumference is equally divided. Because The unit is ,and The unit is Therefore, the weighting coefficient The unit is . The following system of linear equations is obtained by solving: The system of equations is derived from... Interpolation conditions at points where the circumference is equally divided ( (The known radial deformation at this node) and the zero-crossing constraint at the geometric center of the hubcap. Together they constitute, thus uniquely determining all weight coefficients; This represents a spline-type radial basis function, specifically in the form of: , Let Euclidean distance represent the independent variable. Solve for the interpolation weight coefficients. Then, the position vectors of all discrete grid points on the surface of the hub cover are traversed, and the radial displacement field values ​​of each discrete grid point are calculated by substituting them into the interpolation formula to obtain a continuous radial displacement field distribution map.

[0034] In practical implementation, when driving the first group of actuators in the suspended actuator array using the radial displacement field distribution map, the gradient direction of the radial displacement field distribution map is extracted as the force application direction of the first group of actuators. The gradient direction is determined by the radial displacement field distribution map. The gradient vector field is obtained, and its direction at each location on the hubcap surface is the direction of the radial compensation force required at that location. The amplitude value of the radial displacement field distribution map is extracted as the reference for the force amplitude of the first set of actuators, and the amplitude value is the radial displacement field distribution map. The absolute value at each position.

[0035] In practical implementation, when performing hysteresis correction on the applied force amplitude reference, a pre-stored correction table is invoked. This table is established based on the hysteresis curves of the first group of actuator materials. The hysteresis curves are obtained during the factory testing phase of the first group of actuators by applying increasing and decreasing current excitations and simultaneously measuring the actuator output force, forming two force-current relationship curves: a current rising segment and a current falling segment. The correction table takes the applied force amplitude reference and the force direction change indicator as inputs and outputs the corrected applied force amplitude. When the applied force direction is consistent with the applied force direction of the previous control cycle, the correction value corresponding to the current rising segment curve is retrieved; when the applied force direction reverses compared to the previous control cycle, the correction value corresponding to the current falling segment curve is retrieved, thereby compensating for the influence of hysteresis on the force output accuracy and generating the corrected applied force amplitude.

[0036] In practical implementation, when the applied force direction and the corrected applied force amplitude are encoded as pulse width modulation (PWM) signals, the applied force direction corresponds to the polarity selection signal of each actuator in the first group of actuators, and the corrected applied force amplitude corresponds to the duty cycle value of the PWM signal. The PWM signal is generated by the digital output module of the control unit and sent to the drive circuit of the first group of actuators. The drive circuit includes an H-bridge power amplifier unit, which drives the electromagnetic coils in the first group of actuators according to the polarity selection signal and the duty cycle value, so that each actuator generates a specified radial compensation force component. The radial compensation forces of each actuator are vector-synthesized to form the first radial compensation force.

[0037] See Figure 4 The figure shows a schematic curve of the rotational phase deviation function mapped to the radial displacement field of the hubcap in Example 2. The horizontal axis represents the circumferential angular position of the hubcap in radians, ranging from 0 to approximately 6.28 (i.e., 0 to 2π), corresponding to a full circumference of the hubcap. The vertical axis represents the radial displacement in meters, reflecting the amplitude of the radial deformation of the hubcap surface. The figure contains four curves, each corresponding to a different value of the rotational phase deviation function: 0.01 rad (blue solid line), 0.02 rad (orange dashed line), 0.03 rad (green dotted line), and 0.04 rad (red dotted line).

[0038] The curves exhibit a clear cosine function characteristic, with the radial displacement varying periodically with the circumferential angle position. At circumferential angles of 0 and 2π, the radial displacement reaches a positive peak, corresponding to the maximum radial convex deformation; at approximately π, the radial displacement reaches a negative peak, corresponding to the maximum radial concave deformation. The amplitude of each curve increases linearly with the increase of the rotational phase deviation function value, with the maximum radial displacement range increasing from approximately ±0.01 meters to ±0.04 meters, indicating that the amplitude of the radial displacement field is proportional to the rotational phase deviation function.

[0039] This figure vividly illustrates the relationship between the rotating phase deviation function and the reference phase offset angle. The radial projection calculation results, and the spatial distribution law of the continuous radial displacement field generated by combining the three-dimensional topography database and radial basis function interpolation, are shown in the figure. The radial deformation curves corresponding to different rotational phase deviation functions will be used to drive the first group of actuators in the suspension actuator array to apply radial compensation forces with corresponding amplitudes and directions, so as to realize the dynamic adjustment and compensation of the independent rotation state of the hub cover.

[0040] Example 3 In specific implementation, please refer to Figure 2 The steps for real-time monitoring of the axial tilt angle of the hubcap are as follows: Four laser rangefinders are evenly arranged on the edge of the hubcap to measure the instantaneous distance from the edge of the hubcap to the end face of the hub; the difference between the instantaneous distance values ​​measured by the laser rangefinders at each of two relative positions is calculated to obtain a first difference vector and a second difference vector; the first difference vector and the second difference vector are cross-producted to generate an axial tilt angle sequence; the axial tilt angle sequence includes a first tilt angle component around the horizontal axis and a second tilt angle component around the vertical axis.

[0041] In practical implementation, the four laser rangefinders are arranged as follows: With the geometric center of the wheel hub as the origin, four laser rangefinders are fixedly installed radially at four positions corresponding to 0°, 90°, 180°, and 270° along the circumference of the wheel hub. The radius of the mounting circle for the four laser rangefinders is R. The light spots from the four laser rangefinders are perpendicularly illuminated onto a fixed reflective target surface on the edge of the wheel hub. The reflective target surface is a polished stainless steel plane, and it remains parallel to the wheel hub end face. The four laser rangefinders continuously collect the distance from the edge of the wheel hub to the wheel hub end face at a fixed sampling frequency, outputting four instantaneous distance values, denoted as follows: The original measurement units are micrometers. Before participating in subsequent vector operations, each instantaneous distance value is converted from micrometers to millimeters (i.e., divided by 1000). The converted distance value is denoted as... , , , All units are millimeters. The sampling time is indicated, and the unit for all four instantaneous distance values ​​is micrometers. In one embodiment, the outer diameter of the hubcap is 400 mm, and the mounting radius R of the four laser rangefinder sensors is set to 180 mm.

[0042] In practical implementation, when calculating the first and second difference vectors, a spatial rectangular coordinate system is established with the center of the wheel hub end face as the origin, the direction along the 0° laser rangefinder sensor as the positive x-axis, the direction along the 90° laser rangefinder sensor as the positive y-axis, and the direction perpendicular to the wheel hub end face as the positive z-axis. The spatial coordinates of the measurement points corresponding to the four laser rangefinder sensors in the coordinate system are expressed as follows: , , , The first difference vector consists of the spatial coordinate difference between the measurement points of the 0° laser rangefinder sensor and the 180° laser rangefinder sensor, which are arranged relatively oppositely, and is represented as follows: The second difference vector consists of the spatial coordinate difference between the measurement points of the 90° laser rangefinder sensor and the 270° laser rangefinder sensor, which are arranged relatively oppositely, and is expressed as follows: First difference vector Second difference vector All are three-dimensional vectors, and the units of each component are consistent with the coordinates.

[0043] In practice, the first difference vector and the second difference vector are cross-producted to obtain the normal vector. The formula for the cross product operation is: in, This represents the normal vector obtained by the cross product of the first difference vector and the second difference vector. ; Represents the first difference vector; Indicates the second difference vector; multiplication sign This represents the cross product operation of three-dimensional vectors. According to the rules of the cross product operation, the formulas for calculating the components of the normal vector are as follows: , , .

[0044] In practical implementation, the normal vector Generate an axial tilt angle sequence, axial tilt angle sequence Includes the first tilt angle component about the horizontal axis and the second tilt angle component about the vertical axis The generation method is as follows: Will and Substituting the expression, we obtain the first tilt angle component. Second tilt angle component The unit is radians. Represents the normal vector The component along the x-axis, in square millimeters; Represents the normal vector The component along the y-axis, in square millimeters; This indicates the radius of the mounting circle for the four laser rangefinders, in millimeters. , , , These represent the instantaneous distance values ​​measured by the four laser rangefinders at the same sampling time, in micrometers. Axial tilt angle sequence. The changes over time form a time series, and the tilt state information, which includes the first tilt angle component around the horizontal axis and the second tilt angle component around the vertical axis, is output in real time.

[0045] Example 4 In specific implementation, the process of adjusting the second group of actuators in the suspended actuator array according to the axial tilt deviation to generate the second axial compensation torque is as follows: decompose the axial tilt deviation into a first deviation component around the horizontal axis and a second deviation component around the vertical axis; calculate the expected magnitude and direction of the first group of compensation torque using the first deviation component; calculate the expected magnitude and direction of the second group of compensation torque using the second deviation component; convert the expected magnitude and direction of the first group of compensation torque into the corresponding thrust parameters of each actuator in the second group of actuators, and obtain the corresponding thrust parameters by looking up a table of pre-stored actuator thrust-current curves; convert the expected magnitude and direction of the second group of compensation torque into the corresponding thrust parameters of the remaining actuators in the second group of actuators, and combine them to generate the drive signal of the second axial compensation torque.

[0046] In practice, the axial tilt deviation is generated by subtracting the axial tilt angle sequence from a preset equilibrium tilt angle. The preset equilibrium tilt angle is zero degrees, corresponding to the ideal state where the hub cap end face and the hub end face are completely parallel. The axial tilt deviation is represented as a two-dimensional vector, denoted as [vector name missing]. ,in This represents the first deviation component about the horizontal axis, in radians. This represents the second deviation component about the vertical axis, expressed in radians.

[0047] In practical implementation, the first deviation component is utilized. Calculate the expected magnitude and direction of the first set of compensation torques. The expected magnitude of the first set of compensation torques. The operational relationship is determined by the proportional control law. ,in This represents the torque gain coefficient, expressed in Newton-meters per radian. Torque Gain Coefficient Based on the moment of inertia of the hub cap about the horizontal axis With target control bandwidth The operational relationship is set as follows: Moment of inertia of the hubcap about a horizontal axis Calculated from a 3D digital model, the value is [value to be filled in]. kilograms per square meter; target control bandwidth The value is determined based on the response requirements of the suspension system. Radius per second, therefore torque gain coefficient Set as Newton-meters per radian. The direction of the first set of compensating torques is determined by the first deviation component. The sign of the positive or negative value determines the direction of the compensation torque corresponding to a positive deviation, which is in the negative direction around the horizontal axis to counteract the deviation.

[0048] In practical implementation, the second deviation component is utilized. Calculate the expected magnitude and direction of the second set of compensation torques. The expected magnitude of the second set of compensation torques. Also given by the proportional control law The torque gain coefficient The torque gain coefficient is the same value as in the first set of compensation torque calculations because the moment of inertia of the hubcap about the vertical axis is the same. and Equal. The direction of the second set of compensating torques is determined by the second deviation component. The sign of the positive or negative value determines the direction of the compensation torque corresponding to a positive deviation, which is in the negative direction around the vertical axis to counteract the deviation.

[0049] In practical implementation, the expected magnitude and direction of the first set of compensation torques and the expected magnitude and direction of the second set of compensation torques are converted into the corresponding thrust parameters of each actuator in the second set of actuators. The second set of actuators consists of... It consists of an electromagnetic actuator. The value is 8, and each electromagnetic actuator is along a radius of... The circumference is evenly distributed. The value is taken as 0.18 meters, and the force direction of each electromagnetic actuator is perpendicular to the hub end face. The expected thrust of each electromagnetic actuator is calculated by the torque distribution formula. Desired thrust of an electromagnetic actuator The calculation formula is: in, Indicates the first The desired thrust of an electromagnetic actuator, in Newtons; The radius of the circle containing each electromagnetic actuator is 0.18 meters. This indicates the total number of electromagnetic actuators in the second group of actuators, which is 8; This represents the expected magnitude of the first set of compensation torques, expressed in Newton-meters (N·m). This represents the expected magnitude of the second set of compensation torques, expressed in Newton-meters (N·m). Indicates the first The installation azimuth angle of each electromagnetic actuator on the circumference is taken as the zero-degree azimuth reference, with the 0° position of the electromagnetic actuator as the zero-degree reference. The unit is radians.

[0050] In specific implementation, the first Desired thrust of an electromagnetic actuator Subsequently, the corresponding thrust parameters are obtained by looking up the pre-stored actuator thrust-current curves in a table. The actuator thrust-current curves are measured during the factory calibration phase of the electromagnetic actuator. The calibration method is as follows: the electromagnetic actuator is installed above the force sensor, and a DC current from 0 amperes to the maximum drive current is sequentially applied in steps of 0.05 amperes. The steady-state thrust generated by the electromagnetic actuator at each current value is recorded, forming the rising segment thrust-current data set; then, the current is decreased from the maximum drive current to 0 amperes in the same step, and the falling segment thrust-current data set is recorded. Polynomial fitting is performed on the rising and falling segment data to generate thrust-current curves. A lookup table is created for the thrust value corresponding to each current step point on the curve. The lookup table uses the thrust value as the index and the drive current value as the output, and is stored in the control unit's storage module. During actual operation, the desired thrust is used... The corresponding drive current value is obtained by interpolation search in the lookup table. All The drive current values ​​of the individual electromagnetic actuators are combined to generate a drive signal for the second axial compensation torque. This drive signal contains... The current command for each channel is output to the drive circuit of the second set of actuators via the digital-to-analog converter module of the control unit.

[0051] Example 5 In specific implementation, the process of vector fusion of the first radial compensation force and the second axial compensation torque to form a three-dimensional dynamic balance control command is as follows: the first radial compensation force is decomposed into X-direction and Y-direction components along the radial coordinate system of the hub cover; the second axial compensation torque is scalarized along the Z-direction of the radial coordinate system to obtain a Z-direction torque scalar; the X-direction component, Y-direction component, and Z-direction torque scalar are synthesized according to the three-dimensional vector addition rule to obtain the initial fused control vector; the initial fused control vector is subjected to amplitude limiting filtering, and the upper and lower limits of the amplitude limiting filtering are set according to the physical maximum output value of the suspended actuator array to generate a three-dimensional dynamic balance control command.

[0052] In practical implementation, the radial coordinate system of the hubcap is defined as follows: with the geometric center of the hubcap as the origin, the plane containing the hubcap as the XY plane, and the positive Z-axis pointing outward from the hubcap plane. The X-axis and Y-axis are perpendicular to each other within the hubcap plane. The first radial compensation force is obtained by vector synthesis of all radial forces generated by the first set of actuators, and the first radial compensation force always lies within the hubcap plane. The vector of the first radial compensation force is denoted as... Orthogonally decompose along the X and Y axes of the radial coordinate system to obtain the X-axis component. and Y-direction component Both components are in units of cows.

[0053] In practical implementation, the second axial compensation torque is obtained by synthesizing all the axial forces generated by the second set of actuators around the center of the hub cover. The direction of action of the second axial compensation torque is tilting about the X-axis and tilting about the Y-axis. The second axial compensation torque is expressed in torque vector form, denoted as . When scalarizing the second axial compensation torque along the Z-direction of the radial coordinate system, the scalar value of the equivalent radial force effect generated by the second axial compensation torque in the hub cap plane is extracted. The scalarization process is based on the virtual lever arm transformation principle, and the calculation relationship is as follows: ,in This represents a scalar quantity of torque in the Z direction, measured in Newtons (N). Represents the second axial compensation torque vector The modulus, measured in Newton-meters (N·m); The equivalent force arm length is determined by the average distance from the point of action of the second set of actuators to the geometric center of the hub cap, and is set to 0.18 meters.

[0054] In practical implementation, the X-direction component , Y direction component and Z-direction torque scalar The initial fusion control vector is obtained by synthesizing according to the three-dimensional vector addition rule. The formula is expressed as: in, This represents the initial fusion control vector, which is a three-dimensional column vector containing command values ​​for the three control dimensions; This represents the component of the first radial compensating force along the X-axis of the radial coordinate system, expressed in Newtons (N). This represents the component of the first radial compensating force along the Y-axis of the radial coordinate system, expressed in Newtons (N). This represents the Z-direction torque scalar obtained after scalarization of the second axial compensation torque, in Newtons.

[0055] In practical implementation, the initial fused control vector undergoes amplitude limiting filtering. The physical maximum output value of the suspended actuator array includes the maximum output value of the first group of actuators in each direction of the radial plane and the equivalent maximum output value of the second group of actuators in the axial direction. The maximum output value of the first group of actuators in each direction of the radial plane is denoted as... The value is determined by multiplying the maximum thrust of a single actuator in the first group of actuators by the number of actuators capable of simultaneously exerting force in the same direction. The value is 50 Newtons. The equivalent maximum output force of the second actuator in the axial direction is obtained by taking the maximum value of the second axial compensation torque and the equivalent force arm length. The equivalent maximum output value is obtained through conversion. The value is 40 N. This is the upper bound vector for amplitude limiting filtering. and lower limit Set them to: The initial fusion control vector Each component is compared with its corresponding upper and lower limits. If a component value is greater than the upper limit, the upper limit value is used; if a component value is less than the lower limit, the lower limit value is used; otherwise, the component value remains unchanged. The vector obtained after amplitude limiting filtering is the three-dimensional dynamic balance control command. , The three components correspond to the control force command in the X direction of the hubcap plane, the control force command in the Y direction of the hubcap plane, and the equivalent control force command of the tilting moment, respectively.

[0056] In practice, after generating the three-dimensional dynamic balance control command, it is sent to the drive control unit of the suspended actuator array. The drive control unit consists of a digital signal processor and a power amplifier module, and it receives the three-dimensional dynamic balance control command. The three components of the three-dimensional dynamic balance control command are calculated into independent current commands for each actuator in the floating actuator array based on the actuator layout matrix.

[0057] In practical implementation, the actual output force feedback signals of each actuator in the suspended actuator array are simultaneously acquired. Each actuator integrates a strain gauge force sensor. The voltage signal output by the strain gauge force sensor is amplified by a signal conditioning circuit and converted from analog to digital to obtain the actual output force value of each actuator. The actual output force values ​​of all actuators are synthesized into an actual output force feedback vector according to the inverse mapping relationship of the actuator layout matrix, denoted as [vector]. , It is a three-dimensional vector, with three components corresponding to the actual force in the X direction, the actual force in the Y direction, and the equivalent force in the Z direction, respectively.

[0058] In practical implementation, the actual output force feedback vector will be used. With three-dimensional dynamic balance control commands The expected output force is compared dimension by dimension to generate a force deviation vector. The calculation formula is: , It is a three-dimensional vector. The force deviation vector... The input is fed into the integral compensation stage, which accumulates the historical force deviations to generate an integral correction. The recursive form of the cumulative operation is: ,in This represents the integral correction amount for the current sampling period. This represents the integral correction amount for the previous sampling period. This represents the force deviation vector for the current sampling period. This indicates the sampling period, with a value of 0.5 milliseconds. This represents the integral gain coefficient.

[0059] In practical implementation, the integral gain coefficient Based on the force deviation vector The magnitude is adaptively adjusted. The force deviation vector is calculated. modulus When the modulus Greater than the preset threshold At that time, the integral gain coefficient Values , Set to 0.15 to accelerate convergence; when the modulus... Less than or equal to the preset threshold At that time, the integral gain coefficient Values , Set to 0.03 to prevent overshoot. Preset threshold. Based on the force output resolution and noise level settings of the suspension actuator array, a value of 0.5 N is chosen. This value is one-hundredth of the stress sensor's full-scale range of 50 N, and can reliably distinguish between the actual force deviation and measurement noise. The integral correction amount... Superimposed on three-dimensional dynamic balance control commands Generate updated control commands Updated control commands It replaces the current three-dimensional dynamic balance control command and outputs it to the drive control unit to complete the force feedback closed-loop correction.

[0060] See Figure 5 In the figure, the horizontal axis represents time in seconds, ranging from 0 to 0.5 seconds, and the vertical axis represents force / torque value in Newtons, ranging from approximately -6 to 8 Newtons. The legend includes three curves: the "radial compensation force X component" (solid blue line), the "radial compensation force Y component" (dashed orange line), and the "Z-direction torque scalar" (dotted green line), corresponding to the three components of the three-dimensional dynamic balance control command in Embodiment 5.

[0061] As shown in the figure, all three curves exhibit a decaying oscillation trend, indicating that the system has a relatively large output value at the initial moment (0 seconds), and then the force and torque gradually decrease and tend to zero, reflecting the stabilization process of the dynamic response. Specifically, at the initial moment, the radial compensation force X component is at its maximum of about 8 N, the Y component is about 6 N, and the Z-direction torque scalar is about 3.5 N, indicating that the initial control command has a high demand for radial and axial compensation torque of the hubcap. As time progresses, all three components decay rapidly, and the oscillation amplitude decreases significantly after 0.2 seconds, with the force value basically stabilizing near zero, indicating that the hubcap has achieved a stable state of dynamic balance control after force feedback closed-loop adjustment.

[0062] In the figure, the oscillation frequencies of the X and Y components of the radial compensation force are similar, and the amplitude of the X component is slightly larger than that of the Y component in the initial stage, reflecting the dominant role of the first radial compensation force in controlling the rotational stability of the hubcap. The scalar amplitude of the Z-direction torque is relatively small and decays quickly, indicating that the second axial compensation torque has a rapid and stable control response to the tilt adjustment of the hubcap.

[0063] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for independent rotational holding and dynamic balance control of large-size floating hub caps, characterized in that, The method includes: The current rotation angle of the hub cover is continuously sampled to generate a rotation phase sequence; A rotation phase deviation function is established based on the difference between the rotation phase sequence and the preset reference rotation phase. The rotational phase deviation function is mapped to the radial displacement field of the hub cover to generate a radial displacement field distribution map; The radial displacement field distribution map is used to drive the first group of actuators in the suspended actuator array to generate a first radial compensation force. At the same time, the axial tilt angle of the hub cover is monitored in real time, and an axial tilt angle sequence is generated; The axial tilt angle sequence is differentially processed with the preset balance tilt angle to generate the axial tilt deviation. The second group of actuators in the suspended actuator array is adjusted according to the axial tilt deviation to generate a second axial compensation torque. The first radial compensation force and the second axial compensation torque are vector-fused to form a three-dimensional dynamic balance control command. The vector fusion of the first radial compensation force and the second axial compensation torque includes: The first radial compensation force is decomposed into X-direction components and Y-direction components along the radial coordinate system of the hub cover. The second axial compensation torque is scalarized along the Z direction of the radial coordinate system to obtain a Z-direction torque scalar. The scalar value of the equivalent radial force effect generated by the second axial compensation torque in the hub cap plane is then extracted. The scalarization process is based on the virtual lever arm transformation principle, and the calculation relationship is as follows: ,in This represents a scalar quantity of torque in the Z direction, measured in Newtons (N). Represents the second axial compensation torque vector The modulus, measured in Newton-meters (N·m); The equivalent force arm length is determined by the average distance from the point of action of the second set of actuators to the geometric center of the hub cap. The X-direction component, the Y-direction component, and the Z-direction torque scalar are synthesized according to the three-dimensional vector addition rule to obtain the initial fused control vector, expressed by the formula: ； in, This represents the initial fusion control vector, which is a three-dimensional column vector containing command values ​​for the three control dimensions; This represents the component of the first radial compensating force along the X-axis of the radial coordinate system, expressed in Newtons (N). This represents the component of the first radial compensating force along the Y-axis of the radial coordinate system, expressed in Newtons (N). This represents the Z-direction torque scalar obtained after scalarization of the second axial compensation torque, in Newtons. The initial fusion control vector is subjected to amplitude limiting filtering, and the upper and lower limits of the amplitude limiting filtering are set according to the physical maximum output value of the suspended actuator array to generate the three-dimensional dynamic balance control command.

2. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 1, characterized in that, The establishment of the rotation phase deviation function includes: The rotating phase sequence is expanded using Fourier series to extract the fundamental frequency phase component and several harmonic phase components. Calculate the first-order deviation between the fundamental frequency phase component and the preset reference rotating phase, and the second-order deviation between each harmonic phase component and the corresponding preset harmonic reference phase. The first-order deviation value and the second-order deviation value are weighted and summed according to the frequency weighting coefficient to obtain the rotating phase deviation function.

3. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 1, characterized in that, The process of mapping the rotational phase deviation function to the radial displacement field of the hub cover includes: The rotational phase deviation function is discretized along the circumferential direction of the hub cover to generate phase deviation scalars at each circumferentially divided point. A three-dimensional morphological database of radial deformation of the hubcap under different phase deviation values ​​is pre-established; The three-dimensional topography database is invoked, and the corresponding radial deformation set is obtained by querying the phase deviation scalar at each of the equally divided points of the circumference; Spatial interpolation is performed on the radial deformation set to generate a continuous radial displacement field distribution map.

4. The method for independent rotation holding and dynamic balance control of a large-size floating hub cover according to claim 3, characterized in that, The first group of actuators in the suspended actuator array driven by the radial displacement field distribution map includes: The gradient direction of the radial displacement field distribution map is taken as the force application direction of the first group of actuators; The amplitude value of the radial displacement field distribution map is extracted as the reference for the force amplitude of the first group of actuators; Hysteresis correction processing is performed on the applied force amplitude reference. The hysteresis correction processing is implemented based on the correction table pre-stored hysteresis curve of the first group of actuator materials to generate the corrected applied force amplitude. The applied force direction and the corrected applied force amplitude are encoded into pulse width modulation signals and sent to the drive circuit of the first set of actuators.

5. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 1, characterized in that, The real-time monitoring of the axial tilt angle of the hub cover includes: Four laser rangefinders are evenly arranged along the edge of the hubcap to measure the instantaneous distance from the edge of the hubcap to the end face of the hub. Calculate the difference between the instantaneous distance values ​​measured by the laser rangefinder at every two relative positions to obtain the first difference vector and the second difference vector; Perform a cross product operation between the first difference vector and the second difference vector to generate the axial tilt angle sequence; The axial tilt angle sequence includes a first tilt angle component about the horizontal axis and a second tilt angle component about the vertical axis.

6. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 1, characterized in that, The adjustment of the second group of actuators in the suspension actuator array according to the axial tilt deviation includes: The axial tilt deviation is decomposed into a first deviation component about the horizontal axis and a second deviation component about the vertical axis; The expected magnitude and direction of the first set of compensation torques are calculated using the first deviation component; The expected magnitude and direction of the second set of compensation torques are calculated using the second deviation component; The expected magnitude and direction of the first group of compensation torques are converted into the corresponding thrust parameters of each actuator in the second group of actuators. The corresponding thrust parameters are obtained by looking up a table of pre-stored actuator thrust-current curves. The expected magnitude and direction of the second set of compensation torques are converted into the corresponding thrust parameters of the remaining actuators in the second set of actuators, and then combined to generate the drive signal of the second axial compensation torque.

7. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 6, characterized in that, After generating the three-dimensional dynamic balance control command, the method further includes: The three-dimensional dynamic balance control command is sent to the drive control unit of the suspended actuator array; Simultaneously, the actual output force feedback signal of each actuator in the suspended actuator array is collected; The actual output force feedback signal is compared with the expected output force corresponding to the three-dimensional dynamic balance control command in a dimension-by-dimensional comparison to generate a force deviation vector. The force deviation vector is input to the integral compensation stage, which accumulates the historical force deviations to generate an integral correction amount. The integral correction is superimposed on the three-dimensional dynamic balance control command to generate an updated control command.

8. The method for independent rotation holding and dynamic balance control of a large-size floating hub cover according to claim 7, characterized in that, The integral gain of the integral compensation stage is adaptively adjusted according to the magnitude of the force deviation vector. When the magnitude is greater than a preset threshold, the integral gain is increased to accelerate convergence; otherwise, the integral gain is decreased to prevent overshoot.

9. The method for independent rotation holding and dynamic balance control of a large-size suspended hub cover according to claim 3, characterized in that, The three-dimensional topography database is obtained by pre-calculating the static and dynamic radial deformation of the hub cover under different phase deviation values ​​through finite element simulation and storing it in a lookup table.

Images