Structural optimization and compound control method for ball screw vertical feeding system

Abstract

The invention belongs to the technical field of high-precision coordinate grinder feeding system design and control, and particularly discloses a structure optimization and compound control method for a ball screw vertical feeding system, which comprises the following steps of: establishing a multi-body dynamic model of a ball screw-spindle box system; the model comprises the flexibility characteristic of the ball screw, the modal characteristic of the spindle box and the nonlinear contact characteristic of the coupler; performing dynamic simulation based on the model to obtain dynamic response characteristics of the system; a response surface method and a multi-objective optimization algorithm are adopted to carry out optimization design on structural parameters of the ball screw, so that the dynamic rigidity of a system is improved, and resonance is inhibited; and constructing a state space model containing a flexible mode, designing a model prediction controller and an extended state observer, performing dynamic acceleration compensation in combination with disturbance feed-forward compensation control, and outputting a current control signal, a speed command signal or a position command signal. The dynamic response performance and the structural robustness of the feeding system can be fundamentally improved.

Description

Technical Field

[0001] This application belongs to the field of high-precision coordinate grinding machine feed system design and control technology, and more specifically, relates to a structural optimization and composite control method for a ball screw vertical feed system. Background Technology

[0002] Coordinate grinding machines are key equipment used for high-precision machining of complex curved surfaces. Their spindle boxes mostly use ball screw drives to achieve micron-level displacement control in the vertical direction.

[0003] However, in actual machining processes, the vertical direction is significantly affected by gravity, acceleration, and the flexibility of the mechanism, which can easily lead to problems such as lead screw bending, system resonance, and response lag, limiting machining efficiency and accuracy. Most existing studies focus on improving the static stiffness of the structure or control optimization based on simplified models, and there is still a lack of a methodological system that integrates dynamic modeling, structural optimization, and composite control, making it difficult to fundamentally improve the dynamic response performance and structural robustness of the feed system.

[0004] Therefore, how to fundamentally improve the dynamic response performance and structural robustness of the feed system is an urgent problem to be solved. Summary of the Invention In view of the shortcomings of the prior art, the purpose of this application is to provide a structural optimization and composite control method for a ball screw vertical feed system, which can fundamentally improve the dynamic response performance and structural robustness of the feed system.

[0005] To achieve the above objectives, this application provides a structural optimization and composite control method for a ball screw vertical feed system, applied in a coordinate grinding machine, comprising the following steps: S10. Establish a multibody dynamics model of the ball screw-spindle box system. This model includes the flexibility characteristics of the ball screw, the modal characteristics of the spindle box, and the nonlinear contact characteristics of the coupling. S20, based on the multibody dynamics model, performs dynamic simulation to obtain the dynamic response characteristics of the system; Based on simulation results, the response surface methodology and multi-objective optimization algorithm are used to optimize the structural parameters of the ball screw in order to improve the dynamic stiffness of the system and suppress resonance. S40, based on the optimized system structure, constructs a state-space model containing flexible modes, and designs a model predictive controller based on this to achieve trajectory prediction and constraint control. At the same time, it introduces an extended state observer to estimate the impact of disturbances in real time, and combines disturbance feedforward compensation control to perform dynamic acceleration compensation, and outputs current control signal, speed command signal or position command signal.

[0006] The structural optimization and composite control method for the vertical feed system of the ball screw provided in this application has the following effects: First, in the vertical feed system of the ball screw of the coordinate grinding machine, by establishing a multibody dynamics model that includes the flexible characteristics of the ball screw, the modal characteristics of the spindle box, and the nonlinear contact characteristics of the coupling, the dynamic behavior of the system in actual operation can be accurately reflected. Then, dynamic simulation based on this model can obtain the dynamic response characteristics of the system, providing data support for subsequent optimization. Subsequently, the structural parameters of the ball screw are optimized using the response surface methodology and multi-objective optimization algorithm, which can effectively improve the dynamic stiffness of the system and suppress resonance, thereby improving the dynamic response performance. Furthermore, a state-space model including flexible modes is constructed based on the optimized system structure, and a model predictive controller is designed accordingly to realize trajectory prediction and constraint control. At the same time, an extended state observer is introduced to estimate the impact of disturbances in real time, and dynamic acceleration compensation is performed in combination with disturbance feedforward compensation control. This composite control strategy can effectively cope with various disturbances in the operation of the system, ensuring the stability and accuracy of the system. The various steps in this application work together to form a complete system, from dynamic modeling and structural optimization to composite control, which fundamentally improves the dynamic response performance and structural robustness of the feed system.

[0007] As a further preferred embodiment, in step S10, the flexibility characteristics of the ball screw are discretized using Timoshenko beam elements, taking into account the nonlinearity of axial preload and Hertzian contact; the modal characteristics of the spindle box are reduced in degrees of freedom using the Craig-Bampton substructure method, retaining modes with frequencies greater than 500 Hz; and the nonlinear contact characteristics of the coupling are simulated using the Iwan model to simulate the micro-slip characteristics of the bolt connection surface.

[0008] As a further preferred embodiment, in the multibody dynamics model, the ball screw achieves flexible body coupling through a modal neutral file, and its modal cutoff frequency is not less than 1.5 times the highest operating frequency of the system.

[0009] As a further preferred option, in step S20, the dynamic simulation includes typical acceleration motion conditions such as applying step acceleration, trapezoidal acceleration, and actual lifting trajectory, and the simulation obtains the ball screw deflection, shaft offset, stress distribution, and modal frequency.

[0010] As a further optimization, in step S30, the NSGA-II algorithm is used to optimize the ball screw diameter, support spacing, pitch, and bearing stiffness parameters, with the first-order natural frequency greater than 800 Hz and the axial dynamic stiffness greater than 500 N / micrometer as constraints, so that the resonance peak value is reduced by more than 40%.

[0011] As a further preferred option, in step S40, the spindle box displacement, velocity, and lead screw flexible mode are used as system state variables, and a state space model is constructed considering the motor driving force, disturbance acceleration, and structural flexible coupling effect.

[0012] As a further preferred embodiment, in step S40, the design objective of the model predictive controller is to minimize the spindle box trajectory error and control energy while satisfying physical constraints; the constraints include maximum acceleration, velocity, and force limits.

[0013] As a further preferred embodiment, in step S40, the disturbance feedforward compensation control processes the estimated output of the extended state observer through a low-pass filter to form an input channel that compensates for the disturbance acceleration in advance; the cutoff frequency of the low-pass filter is 1.2 times the closed-loop bandwidth of the system.

[0014] As a further preferred option, the following also includes: S50, establish an experimental verification platform, collect actual responses through accelerometers and laser displacement meters, verify the accuracy and dynamic stability of structural optimization and control strategies, require step response overshoot of less than 5%, positioning error within ±2 micrometers, and FFT analysis showing a reduction of more than 30% in resonance peak amplitude.

[0015] As a further preferred embodiment, in the test platform, the resolution of the laser displacement meter is less than or equal to 0.1 μm, the bandwidth of the accelerometer is greater than or equal to 2 kHz, and the simulation and measured errors after model calibration meet the following requirements: time-domain displacement error RMSE is less than or equal to 3 μm, and frequency-domain resonance peak amplitude deviation is less than or equal to 15%. Attached Figure Description

[0016] Figure 1 This is a flowchart of the structural optimization and composite control method of the ball screw vertical feed system provided in this application; Figure 2 This is a diagram illustrating the implementation steps of the structural optimization and composite control method for the vertical feed system of the ball screw provided in this application embodiment; Figure 3 This is an experimental flowchart provided in a specific implementation case of this application; Figure 4 This is an experimental verification diagram provided in a specific implementation case of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] To fundamentally improve the dynamic response performance and structural robustness of the feed system, this application presents a structural optimization and composite control method for the vertical feed system of a ball screw in a coordinate grinding machine. A transient dynamic model of the system is constructed through multibody dynamics coupled finite element analysis, considering the stress deformation and structural response of the ball screw during high-speed acceleration. Based on the structural response results, a multi-objective optimization algorithm is used to systematically optimize the structural parameters of the ball screw (such as diameter, pitch, and support spacing), thereby increasing the system's modal frequency and enhancing its dynamic stiffness.

[0019] Meanwhile, to further suppress dynamic errors during high-speed motion, this application designs a composite control strategy combining disturbance feedforward and model predictive control (MPC). Based on the system state-space model, an extended state observer (ESO) is introduced to compensate for acceleration disturbances during high-speed acceleration and deceleration, thereby improving the displacement accuracy of the spindle box and suppressing vibration. Finally, an experimental platform is constructed to compare simulations and field measurements, verifying the practical effects and engineering applicability of the proposed method in terms of step response, resonance suppression, and positioning accuracy.

[0020] like Figure 1 As shown, the structural optimization and composite control method for the vertical feed system of the ball screw provided in this application is applied to a coordinate grinding machine, including steps S10 to S40, which are detailed below: Step S10: Establish a multibody dynamics model of the ball screw-spindle box system. This model includes the flexibility characteristics of the ball screw, the modal characteristics of the spindle box, and the nonlinear contact characteristics of the coupling.

[0021] In step S10, the flexibility characteristics of the ball screw are discretized using Timoshenko beam elements, considering the axial preload and Hertzian contact nonlinearity. The modal characteristics of the spindle box are reduced in degrees of freedom using the Craig-Bampton substructure method, retaining modes with frequencies greater than 500 Hz. The nonlinear contact characteristics of the coupling are simulated using the Iwan model to model the micro-slip characteristics of the bolt connection surface.

[0022] Step S20: Perform dynamic simulation based on the multibody dynamics model to obtain the dynamic response characteristics of the system.

[0023] Specifically, step S20 can be: applying typical acceleration motion conditions, including step acceleration, trapezoidal acceleration and actual lifting trajectory, and simulating to obtain the screw deflection, shaft offset, stress distribution and modal frequency.

[0024] Step S30: Based on the simulation results, the response surface methodology and multi-objective optimization algorithm are used to optimize the structural parameters of the ball screw (screw diameter, support spacing, pitch, bearing stiffness, etc.) to improve the dynamic stiffness of the system and suppress resonance.

[0025] Specifically, step S30 can use the NSGA-II algorithm to optimize the ball screw diameter (20-40mm), support spacing (500-1200mm), pitch, and bearing stiffness parameters, with the first-order natural frequency greater than 800 Hz and the axial dynamic stiffness greater than 500 N / micrometer as constraints, so that the resonance peak value is reduced by more than 40%.

[0026] Step S40: Based on the optimized system structure, a state-space model containing flexible modes is constructed, and a model predictive controller is designed accordingly to achieve trajectory prediction and constraint control. At the same time, an extended state observer is introduced to estimate the impact of disturbances in real time, and dynamic acceleration compensation is performed by combining disturbance feedforward compensation control, and output current control signal, speed command signal or position command signal.

[0027] In this application, step S40 can be specifically described as follows: Integrating the control strategy, based on the structurally optimized ball screw drive system, a state-space model is constructed using the spindle box displacement, velocity, and screw flexibility mode as system state variables, considering the motor driving force, disturbance acceleration, and structural flexibility coupling effects. This model is used to dynamically predict the spindle box response, reflecting the influence characteristics of system flexibility and inertial disturbances in high-speed motion. Based on the above state-space model, a disturbance-aware MPC controller is constructed. The controller design objective is to minimize the spindle box trajectory error and control energy while satisfying physical constraints (maximum acceleration, velocity, and force limits). Considering the existence of unknown external disturbances in the vertical feed system (such as inertial disturbances caused by rapid spindle acceleration and deceleration, and screw flexibility vibration coupling), this application designs an Extended State Observer (ESO) to estimate and dynamically compensate for system disturbance terms in real time. To further enhance the system's feedforward response capability, a disturbance feedforward compensation module is added to the control input. The disturbance feedforward term is estimated by the ESO output and processed by a low-pass filter (cutoff frequency is 1.2 times the system closed-loop bandwidth) to form an input channel that compensates for disturbance acceleration in advance, resulting in faster system response and smaller overshoot under high-speed acceleration and deceleration conditions. The final control strategy consists of three parts: 1. A Model Predictive Controller (MPC) for overall trajectory prediction and constraint control; 2. An Extended State Observer (ESO) for real-time estimation of disturbance effects; 3. A disturbance feedforward control channel for dynamic acceleration compensation. The direct outputs of the control strategy include: 1. Current control signal (I): suitable for current loop control mode; 2. Speed ​​command signal (v): suitable for speed loop mode; 3. Position command signal (x): suitable for position loop control or servo position control mode.

[0028] Furthermore, this application may also include: step S50, establishing an experimental verification platform, collecting actual responses through accelerometers and laser displacement meters, and verifying the accuracy and dynamic stability of the structural optimization and control strategy. By comparing simulation and measured data, the step response overshoot is required to be <5%, the positioning error to be within ±2μm, and FFT analysis to show that the resonance peak amplitude is reduced by more than 30%.

[0029] In step S50, the resolution of the laser displacement meter is less than or equal to 0.1 μm, the bandwidth of the accelerometer is greater than or equal to 2 kHz, and the simulation and measured errors after model calibration must meet the following requirements: time-domain displacement error RMSE is less than or equal to 3 μm, and frequency-domain resonance peak amplitude deviation is less than or equal to 15%.

[0030] The structural optimization and composite control method for the vertical feed system of the ball screw provided in this application has the following effects: First, in the vertical feed system of the ball screw of the coordinate grinding machine, by establishing a multibody dynamics model that includes the flexible characteristics of the ball screw, the modal characteristics of the spindle box, and the nonlinear contact characteristics of the coupling, the dynamic behavior of the system in actual operation can be accurately reflected. Then, dynamic simulation based on this model can obtain the dynamic response characteristics of the system, providing data support for subsequent optimization. Subsequently, the structural parameters of the ball screw are optimized using the response surface methodology and multi-objective optimization algorithm, which can effectively improve the dynamic stiffness of the system and suppress resonance, thereby improving the dynamic response performance. Furthermore, a state-space model including flexible modes is constructed based on the optimized system structure, and a model predictive controller is designed accordingly to realize trajectory prediction and constraint control. At the same time, an extended state observer is introduced to estimate the impact of disturbances in real time, and dynamic acceleration compensation is performed in combination with disturbance feedforward compensation control. This composite control strategy can effectively cope with various disturbances in the operation of the system, ensuring the stability and accuracy of the system. The various steps in this application work together to form a complete system, from dynamic modeling and structural optimization to composite control, which fundamentally improves the dynamic response performance and structural robustness of the feed system.

[0031] In one embodiment, the technical solution for achieving the above objective can be specifically as follows: Figure 2 As shown, this embodiment provides a method for ball screw dynamics simulation and structural optimization during the vertical acceleration motion of the spindle box of a coordinate grinding machine. The specific steps include: S1: Multibody Dynamics Modeling: Based on Timoshenko beam theory and nonlinear contact mechanics, a multibody dynamics model including a ball screw, coupling, and headstock is established. The ball screw is modeled as a flexible body, and a modal neutral file (.mnf) is used to achieve high-order mode truncation (frequency not lower than 1.5 times the highest operating frequency of the system). The coupling model considers micro-slippage at the bolt connection surface, and its nonlinear frictional hysteresis behavior is described using the Iwan model. The headstock is modeled using the Craig-Bampton substructure method to reduce its order while retaining the main transmission modes.

[0032] S2: Load Condition Definition and Transient Response Analysis: Construct typical acceleration motion conditions, including step acceleration, trapezoidal velocity curves and actual machining lifting trajectories. Obtain transient deflection response, shaft offset, internal stress distribution and modal frequency drift data of the lead screw in the system through coupled simulation, and evaluate its dynamic performance boundaries.

[0033] S3: Structural Parameter Optimization and Resonance Avoidance Design: Based on the response surface methodology and multi-objective optimization algorithms (such as NSGA-II genetic algorithm or particle swarm optimization), key parameters such as screw diameter, support spacing, pitch, and bearing stiffness are synergistically optimized. The objective function is a weighted combination of the system's maximum deflection, maximum stress, and modal coupling degree. Under the constraints of "first-order natural frequency > 800Hz, axial stiffness > 500N / μm", the optimized scheme reduces the system's resonance amplitude by more than 40%.

[0034] S4: An integrated control strategy is implemented based on the structurally optimized ball screw drive system. The spindle box displacement, velocity, and screw flexibility modes are used as system state variables. A state-space model is constructed considering the motor driving force, disturbance acceleration, and the coupling effect of structural flexibility. This model is used to dynamically predict the spindle box response, reflecting the influence characteristics of system flexibility and inertial disturbances during high-speed motion. Based on the above state-space model, a disturbance-aware MPC controller is constructed. The controller's design objective is to minimize the spindle box trajectory error and control energy while satisfying physical constraints (maximum acceleration, velocity, and force limits). Considering the existence of unknown external disturbances in the vertical feed system (such as inertial disturbances caused by rapid spindle acceleration and deceleration, and screw flexibility vibration coupling), this embodiment designs an Extended State Observer (ESO) to estimate and dynamically compensate for system disturbance terms in real time. To further enhance the system's feedforward response capability, a disturbance feedforward compensation module is added to the control input. The disturbance feedforward term is estimated by the ESO output and processed by a low-pass filter (cutoff frequency is 1.2 times the system closed-loop bandwidth) to form an input channel that compensates for disturbance acceleration in advance, resulting in faster system response and smaller overshoot under high-speed acceleration and deceleration conditions. The final control strategy consists of three parts: 1. A Model Predictive Controller (MPC) for overall trajectory prediction and constraint control; 2. An Extended State Observer (ESO) for real-time estimation of disturbance effects; 3. A disturbance feedforward control channel for dynamic acceleration compensation. The direct outputs of the control strategy include: 1. Current control signal (I): suitable for current loop control mode; 2. Speed ​​command signal (v): suitable for speed loop mode; 3. Position command signal (x): suitable for position loop control or servo position control mode.

[0035] S5: Experimental Verification and Simulation Model Calibration: A test platform for the vertical feed system will be built. Laser displacement and acceleration data under step and continuous acceleration conditions will be collected. The model will be calibrated and error corrected. The positioning error must be within ±2μm, the step response overshoot <5%, and the frequency domain FFT analysis showing a resonance peak deviation ≤15%. This will verify the effectiveness and engineering adaptability of the simulation model and control strategy.

[0036] Compared with the prior art, this embodiment has the following significant advantages and innovations: (1) It introduces multibody flexible modeling and nonlinear contact coupling modeling for the first time, which truly reflects the dynamic response characteristics of the ball screw under multi-source load during acceleration and improves the simulation accuracy; (2) It integrates structural parameter optimization and dynamic mode redistribution methods to achieve the collaborative design of system stiffness enhancement and resonance frequency away from the working bandwidth; (3) It proposes a model predictive control algorithm that includes flexible modes, which significantly improves the response accuracy and anti-disturbance stability of the spindle box during high-speed lifting and lowering; (4) It establishes a complete experimental verification closed-loop mechanism, and realizes the accurate implementation of the theoretical model in engineering practice through high-bandwidth sensing system and simulation comparison analysis.

[0037] The following is a specific implementation example of this application: Before describing the structural optimization and composite control method of the ball screw vertical feed system for coordinate grinding machines provided in this specific embodiment, the system structure, sensor layout, dynamic modeling method, and optimized control strategy will be described in detail first. Figure 3 and 4 As shown.

[0038] (1) System modeling and dynamic simulation First, multibody dynamics modeling is performed. The ball screw flexibility modeling utilizes Timoshenko beam theory, considering the effects of shear deformation and rotational inertia. Meshing is performed in finite element software such as ANSYS, generating a modal neutral file (.mnf). The cutoff frequency is set to 1.5 times the operating frequency (e.g., 1500Hz). Axial preload and nonlinear ball-screw contact stiffness are defined using Hertz contact theory. The spindle box reduction modeling uses the Craig-Bampton substructure method, retaining the first 10 modes (frequency greater than 500Hz). The reduced model is generated using finite element software and imported into multibody dynamics software (e.g., Adams). Nonlinear modeling of the coupling uses the Iwan model to simulate the micro-slip characteristics of the bolt connection surface, considering stiffness decay and energy dissipation.

[0039] During dynamic simulation analysis, typical working conditions are defined as follows: step acceleration (0→0.5m / s completed within 0.1s), trapezoidal velocity curve (acceleration→uniform speed→deceleration simulating the actual machining process), and S-shaped acceleration / deceleration (smooth transition, reducing impact). The simulation outputs the lead screw deflection curve, shaft offset trajectory, and stress distribution cloud map. Modal analysis extracts the first 5 natural frequencies and mode shapes, and identifies resonance risk points.

[0040] (2) Structural parameter optimization Structural parameter optimization begins with experimental design, selecting design variables such as lead screw diameter, support spacing, bearing stiffness, and preload, and constructing a response surface model. Sample points are generated using center composite design (CCD) or Latin hypercube sampling, and the maximum deflection, maximum stress, and first-order frequency of each sample point are calculated through finite element simulation.

[0041] For multi-objective optimization, the NSGA-II genetic algorithm or the MOEA / D multi-objective evolutionary algorithm is used. The objective function is defined as minimizing deflection, minimizing stress, and maximizing the first-order frequency. The constraints are that the first-order frequency is greater than 800Hz and the maximum stress is less than the allowable stress of the material. The optimization process includes initializing the population → calculating the objective function → non-dominated sorting → selection, crossover, and mutation → iterative convergence.

[0042] In the optimization result verification stage, the optimal solution is recalculated using finite element method to confirm whether the stiffness, frequency, and stress meet the standards. If they do not meet the standards, the variable range is adjusted or constraints are added for re-optimization.

[0043] (3) Design of composite control strategy The composite control strategy design begins with state-space modeling, selecting the spindle box displacement, velocity, and acceleration, along with the first two modal coordinates of the leadscrew. System identification is achieved through frequency response function testing or step response testing, fitting the transfer function. Discretization is performed using the zero-order hold method, with a sampling frequency greater than or equal to 5kHz.

[0044] Model predictive control (MPC) constructs a predictive model based on state-space equations to predict the system output N steps ahead. In solving the optimization problem, the objective function is to minimize the tracking error plus control energy, with constraints that acceleration is less than 10 m / s² and velocity is less than 1 m / s². During real-time control, a QP problem is solved in each control cycle to output the optimal control force.

[0045] An extended state observer (ESO) is used for perturbation estimation, treating modeling errors and external disturbances as total perturbations and extending them into new state variables. The observer design employs a Luneburger observer or a sliding mode observer, with a bandwidth set to 3 to 5 times the system bandwidth.

[0046] Feedforward compensation uses acceleration feedforward to calculate the ideal control force F based on the acceleration of the reference trajectory. ff =m⋅a ref High-frequency noise is filtered out using a low-pass filter, with the cutoff frequency set to 1.2 times the system bandwidth.

[0047] (4) Experimental verification method During hardware platform setup, the sensor arrangement includes a laser displacement meter to measure the actual position of the spindle box (resolution 0.1μm), an accelerometer to monitor vibration (bandwidth ≥ 2kHz), and a force sensor to measure the axial force of the leadscrew. The data acquisition system uses a high sampling rate DAQ card (NI PXIe-6368) to simultaneously acquire multi-channel signals.

[0048] The test conditions are divided into open-loop test and closed-loop test. In the open-loop test, a step / sinusoidal excitation is applied and the frequency response function (FRF) is measured to verify the accuracy of the model. In the closed-loop test, the control effects of traditional PID, MPC, and MPC + ESO are compared. The test indicators are overshoot, positioning error, and vibration amplitude.

[0049] Data analysis includes time-domain analysis to calculate rise time, steady-state error, and overshoot, and frequency-domain analysis to observe changes in formant amplitude through FFT transformation (comparison before and after optimization).

[0050] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A structural optimization and composite control method for a ball screw vertical feed system, applied in a coordinate grinding machine, characterized in that, Includes the following steps: S10. Establish a multibody dynamics model of the ball screw-spindle box system. This model includes the flexibility characteristics of the ball screw, the modal characteristics of the spindle box, and the nonlinear contact characteristics of the coupling. S20, based on the multibody dynamics model, performs dynamic simulation to obtain the dynamic response characteristics of the system; Based on simulation results, the response surface methodology and multi-objective optimization algorithm are used to optimize the structural parameters of the ball screw in order to improve the dynamic stiffness of the system and suppress resonance. S40, based on the optimized system structure, constructs a state-space model containing flexible modes, and designs a model predictive controller based on this to achieve trajectory prediction and constraint control. At the same time, it introduces an extended state observer to estimate the impact of disturbances in real time, and combines disturbance feedforward compensation control to perform dynamic acceleration compensation, and outputs current control signal, speed command signal or position command signal.

2. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S10, the flexibility characteristics of the ball screw are discretized using Timoshenko beam elements, taking into account the nonlinearity of axial preload and Hertzian contact; the modal characteristics of the spindle box are reduced in degrees of freedom using the Craig-Bampton substructure method, retaining modes with frequencies greater than 500 Hz; the nonlinear contact characteristics of the coupling are simulated using the Iwan model to simulate the micro-slip characteristics of the bolt connection surface.

3. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In the multibody dynamics model, the ball screw achieves flexible body coupling through a modal neutral file, and its modal cutoff frequency is not less than 1.5 times the highest operating frequency of the system.

4. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S20, the dynamic simulation includes typical acceleration motion conditions such as step acceleration, trapezoidal acceleration, and actual lifting trajectory, and the simulation obtains the ball screw deflection, shaft offset, stress distribution and modal frequency.

5. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S30, the NSGA-II algorithm is used to optimize the ball screw diameter, support spacing, pitch, and bearing stiffness parameters. The first-order natural frequency is greater than 800 Hz and the axial dynamic stiffness is greater than 500 N / micrometer as constraints, so that the resonance peak value is reduced by more than 40%.

6. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S40, the spindle box displacement, velocity, and lead screw flexible mode are used as system state variables, and a state space model is constructed considering the motor driving force, disturbance acceleration, and structural flexible coupling effect.

7. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S40, the design objective of the model predictive controller is to minimize the spindle box trajectory error and control energy while satisfying physical constraints; the constraints include maximum acceleration, velocity, and force limits.

8. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, In step S40, the disturbance feedforward compensation control processes the estimated output of the extended state observer through a low-pass filter to form an input channel that compensates for the disturbance acceleration in advance; the cutoff frequency of the low-pass filter is 1.2 times the closed-loop bandwidth of the system.

9. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 1, characterized in that, Also includes: S50, establish an experimental verification platform, collect actual responses through accelerometers and laser displacement meters, verify the accuracy and dynamic stability of structural optimization and control strategies, require step response overshoot of less than 5%, positioning error within ±2 micrometers, and FFT analysis showing a reduction of more than 30% in resonance peak amplitude.

10. The structural optimization and composite control method for the vertical feed system of a ball screw as described in claim 9, characterized in that, In the experimental platform, the resolution of the laser displacement meter is less than or equal to 0.1 μm, the bandwidth of the accelerometer is greater than or equal to 2 kHz, and the simulation and measurement errors after model calibration meet the following requirements: time-domain displacement error RMSE is less than or equal to 3 μm, and frequency-domain resonant peak amplitude deviation is less than or equal to 15%.

Images