Multi-controller fused double-drive synchronous control method

Through the dual-drive synchronization control method of multi-controller fusion, cross-coupling error calculation, H∞ state feedback control and LPV gain scheduling technology, the robustness of the dual-screw feed system in time-varying and nonlinear environments is solved and Dynamic performance issues achieve higher tracking accuracy and vibration suppression effect.

CN118915432APending Publication Date: 2024-11-08WUHAN POLYTECHNIC UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410877432.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-02
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The time-varying and nonlinear characteristics of the dual-screw feed system are difficult to ensure the stability and robustness of the system in the working domain through traditional PID control methods, especially when facing external interference and parameter perturbation, the controller is robust. The performance is reduced, making it difficult to improve the tracking characteristics and dynamic performance of the system.

Method used

The dual-drive synchronization control method with multi-controller fusion includes a cross-coupled error calculation unit, a synchronous tracking controller, a vibration suppression controller and an interference observer. Multiple designs are designed through the H∞ state feedback controller and the LPV gain scheduling controller. The controllers work together to improve the robustness and dynamic performance of the system.

Benefits of technology

Improves the tracking accuracy of the dual-screw feed system, reduces synchronization errors, suppresses vibrations, enhances the dynamic performance and control bandwidth of the system, and effectively deals with uncertain interference, improving the robustness and stability of the system sex.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915432A_ABST
    Figure CN118915432A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-controller fused double-drive synchronous control method, and a synchronous control system comprises a cross coupling error calculation unit, a synchronous tracking controller, a vibration suppression controller, an interference observer, a motor control unit and a motor control unit. The displacement receiving unit is configured to receive the displacement x1 and the displacement x2 output by double shafts X1 and X2 of the double-lead-screw feeding system at the last moment; according to the displacement x1 and the displacement x2 output at the last moment and the reference position instruction r of the double motors, the tracking error of the single shaft is obtained; outputting a cross coupling error e * through a tracking error subunit, a synchronization error subunit and a cross coupling error subunit in sequence; the synchronous tracking controller is configured to receive the cross coupling error output by the cross coupling error calculation unit; and outputting a synchronous tracking motor control signal ue according to the cross coupling error.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of control technology, and in particular to a dual-drive synchronous control method integrating multiple controllers. Background Art

[0002] The time-varying and nonlinear characteristics of the double-screw feed system are considered, and the LPV model of the double-screw feed system is established and the system identification is performed. How to use the LPV model to design a high-precision robust controller and improve the synchronization performance of the double-screw feed system is a challenge currently faced.

[0003] Traditional control methods mainly focus on PID, which regards the feed system as an LTI system, ignoring the influence of external interference, parameter perturbation and unmodeled dynamics on the system. Although PID control can guarantee a certain accuracy, it is difficult to guarantee the stability of the system in the working domain. One solution is to ensure that the designed LTI controller still has robust stability as the dynamic characteristics of the machine tool change. The robust controller can fully consider the uncertainty of the system, use a single controller to meet multiple parameter perturbations, and ensure that the system has a certain robustness. It can effectively improve the tracking characteristics of the system in the low-frequency stage, but the flexibility is insufficient. It is usually assumed that the uncertainty of the system is perturbed near the nominal value. Once the perturbation range of some parameters is too large, the robustness of the controller will be reduced.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a dual-drive synchronous control method integrating multiple controllers, aiming to solve the above-mentioned problems.

[0006] To achieve the above object, the present invention provides a dual-drive synchronous control method with multi-controller fusion, characterized in that the synchronous control system comprises:

[0007] The cross-coupling error calculation unit is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment; and obtain the tracking error of the single axis according to the output displacements x1 and x2 at the last moment and the reference position command r of the dual motors; and then pass through the tracking error subunit, the synchronization error subunit, and the cross-coupling error subunit in sequence to output the cross-coupling error e * ;

[0008] A synchronous tracking controller is configured to receive the cross-coupling error output by the cross-coupling error calculation unit; and output a synchronous tracking motor control signal u according to the cross-coupling error. e ;

[0009] The vibration suppression controller is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the previous moment; and output the vibration suppression motor control signal u according to the output displacements x1 and x2 at the previous moment v ;

[0010] The disturbance observer is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment, and the output motor control signal u at the last moment. m , output motor deviation correction signal u d ;

[0011] The motor control unit synchronously tracks the motor control signal u e , vibration suppression motor control signal u v And the motor deviation correction signal u d , get the motor control signal u m , used to drive the double screw feed system.

[0012] Preferably, in the dual-drive synchronous control method of multi-controller fusion, when designing the synchronous tracking controller, the double-screw feeding system is simplified to an LTI system, and the LTI system is described by the following state space equation:

[0013]

[0014] in, is the system state vector, is the control input, To measure the output, The regulated output is also called performance output. is the external interference input. The interference considered here is uncertain, but the interference signal energy is bounded, that is, w∈L2;

[0015] Design H ∞ The state feedback controller is u=K sy x, so that the following closed-loop system (2) is asymptotically stable, and the closed-loop transfer function T wz (s) satisfies the bounded real theorem;

[0016]

[0017] Take H ∞ Suboptimal design performance γ = 1, the LMI algorithm is used to obtain the positive definite matrices R and S, so that the following LMI is established:

[0018]

[0019] Among them, the matrix N 12 and N 21Respectively and [C2,D 21 ]The basis of the null space is N 21 =ker[C2,D 21 ];

[0020] Design H ∞ When the state feedback controller is used, the weight function W e is a performance index weighting function, which is used to suppress the cross-coupling error e * , weight function W u The weight function is used to control the input weight function, which is used to limit the driving voltage of the dual motors to prevent them from entering the saturation zone; the weight function W e , W u Combined with the system nominal model G n The augmented matrix model of the double screw feed system can be obtained as follows:

[0021]

[0022] x is the state vector matrix of the augmented system, w=[r],u e =[u e1 u e2 ] T , z e =[z e1 z e2 ] T , z u =[z u1 z u2 ] T , Augmented system model and H ∞ State feedback controller K sy To form a closed-loop control system, the system matrix of the augmented system model is:

[0023]

[0024] D 22 =[0]

[0025] Where B = [B l B r ],C=[C u C d ] T , l and r represent the left and right columns of the matrix, and u and d represent the up and down rows of the matrix; the H of the double screw feed system can be obtained by solving LMI through the hinfsyn function in Matlab. ∞ Synchronous tracking controller K sy .

[0026] Preferably, in the dual-drive synchronous control method of multi-controller fusion, when designing the vibration suppression controller, a gain scheduling controller is designed based on the LPV system, and the workbench position x is used to control the vibration suppression controller. w and workpiece mass m w Gain adjustment is achieved to reduce the vibration peaks of different natural frequencies, meet the closed-loop stability of the double-screw feed system, and ensure the designed synchronous controller K sy synchronization performance.

[0027] Preferably, in the dual-drive synchronous control method of multi-controller fusion, the workbench position x is selected w and workpiece mass m w As the scheduling variable, the LPV robust controller is designed, and the double-screw feed system is described by the polyhedral model, and the following LPV model with a polyhedral structure is obtained:

[0028]

[0029] In the formula,

[0030]

[0031] For the LPV model of the polyhedral structure, the state feedback controllers K1, K2, K3, and K4 that meet the requirements are designed for the time-varying parameter polyhedron vertices ω1, ω2, ω3, and ω4 respectively. They are convex polyhedrons, and the workbench is located at any position x in the convex set. w and any mass m w By combining the controller design of each vertex, we can obtain the LPV multi-cell structure controller K of the double screw feed system. vib ,Right now:

[0032]

[0033] Preferably, in the dual-drive synchronous control method of multi-controller fusion, when designing the vibration suppression controller, the ball screw system G(x w ,m w ) and tracking controller K sy (s) are combined to form a new system, in which K sy (s) = C sy (sI-A sy ) -1 B sy +D sy , new combined system model G n (x w ,m w ) is expressed as follows:

[0034]

[0035] in,

[0036] ,

[0037] C n =[C 0];

[0038] Design of vibration suppression controller K for double screw feed system vib When W is introduced F and W v The weight function designed for the controller, W F It is used to suppress the vibration caused by the motor driving force F and limit the vibration amplitude near the natural frequency; W v Used to limit the output signal of the controller to prevent output oversaturation problems; based on the new combined LPV system G n (x w ,m w ) and the weight function W F and W v The new augmented system can be obtained as:

[0039]

[0040] in, is the state vector of the new combined system;

[0041] It is the control output of the new combined system;

[0042] y=[x1 x2] T is the measured output of the system;

[0043] Dynamic weighting function W F (s) = C F (sI-A F ) -1 B F +D F ;

[0044] W v is a constant gain, then the system matrix in the above formula can be derived as follows:

[0045]

[0046] According to LPV variable gain H ∞ The design theory of robust controller, designed vibration suppression controller K vib .

[0047] Preferably, in the dual-drive synchronous control method of multi-controller fusion, the disturbance observer is a filter for filtering out uncertain external disturbances.

[0048] Preferably, in the dual-drive synchronous control method of multi-controller fusion, in the disturbance observer, the filter is a low-pass filter, which is expressed as:

[0049]

[0050] Among them, l ≥ k + b;

[0051] c0=a0;

[0052] τ is the time constant;

[0053] b is the relative order of Q(s).

[0054] Preferably, in the dual-drive synchronous control method of multi-controller fusion, in the disturbance observer, the relative order of the low-pass filter is taken as 1, and its expression is:

[0055] Q(s) = 1 / (τs+1).

[0056] Preferably, in the dual-drive synchronous control method of multi-controller fusion, the synchronous tracking motor control signal u e , vibration suppression motor control signal u v And the motor deviation correction signal u d , get the motor control signal u m , for driving a double-screw feeding system, the motor control unit is configured as:

[0057] Will synchronously track the motor control signal u e , vibration suppression motor control signal u v Add them together to get u m0 ;

[0058] will u m0 and motor deviation correction signal u d Subtract and get the motor control signal u m .

[0059] Preferably, in the dual-drive synchronous control method of multi-controller fusion, in the cross-coupling error calculation unit,

[0060] The tracking error of a single axis is:

[0061]

[0062] Among them, r is the reference position command of the dual motors;

[0063] e1 and e2 are the tracking errors of single axis respectively;

[0064] The synchronization error between the two axes is as follows:

[0065]

[0066] Wherein, ε1 and ε2 are the synchronization errors of the dual axes X1 and X2 of the dual screw feed system respectively;

[0067] Introducing cross-coupling error, equation (11) is written as the following matrix:

[0068] ε=Te;(12)

[0069] Where ε=[ε1ε2] T ;

[0070] e=[e1 e2] T ;

[0071] Represents the synchronous state transfer matrix, then the cross-coupling error can be defined as follows:

[0072] e * =e+βε; (13)

[0073] in,

[0074]

[0075] β=diag(β1,β2) is the cross-coupling coefficient, and β i >0(i=1,2);

[0076] Substituting formula (12) into formula (13) yields:

[0077] e * =(I+βT)e;

[0078] Where I is the identity matrix;

[0079] (I+βT) is a positive definite matrix;

[0080] e * is the cross coupling error.

[0081] The present invention has at least the following beneficial effects:

[0082] The present invention proposes multiple control objectives for the time-varying dynamic characteristics of the double-screw feed system. The first is to improve the tracking accuracy of the feed system and reduce the synchronization error of the dual-drive structure; the second is to suppress various vibrations of the feed system, including high-frequency resonance, nonlinear vibration, etc., to improve the dynamic performance and control bandwidth of the system; and finally to filter and eliminate the uncertain interference of the double-screw feed system. The present invention adopts a cross-coupling parallel synchronous control strategy, designs corresponding controllers for the above-mentioned control objectives, and performs multi-controller collaborative control. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a synchronous control strategy diagram of the cross-coupled parallel structure provided by the present invention;

[0084] Figure 2 A block diagram of the cross-coupling synchronization error model proposed by the present invention;

[0085] Figure 3 The double screw feeding system H provided by the present invention ∞ Synchronous tracking controller block diagram;

[0086] Figure 4 It is a multi-cell LPV system with time-varying parameters of the double-screw feed system;

[0087] Figure 5 A block diagram of a vibration suppression controller for a double-screw feed system provided by the present invention;

[0088] Figure 6 The weight function W provided by the present invention is e Design;

[0089] Figure 7 The weight function W provided by the present invention is F Design;

[0090] Figure 8 The schematic diagram of the LTI system interference observer provided by the present invention;

[0091] Fig. 9 A simplified block diagram of the interference observer provided by the present invention;

[0092] Fig.10 This is a physical picture of a two-dimensional dual-drive grinding wheel dressing machine equipped with a dual-screw feed system;

[0093] Fig.11 To compare the synchronization error curves of the double-screw feed system under different control experiments;

[0094] Fig.12 To compare the workbench positioning error curves of the double-screw feed system under different control experiments;

[0095] Fig.13Synchronous error curve of the comparative test study of two control strategies for the double screw feed system;

[0096] Fig.14 Positioning error curve of the comparative test study of two control strategies for the double-screw feed system.

[0097] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0098] In the embodiments of the present invention, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0099] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0100] In the embodiments of the present invention, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0101] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0102] The present invention provides a dual-drive synchronous control method with multi-controller fusion, please refer to Figure 1 The synchronous control system includes a cross-coupling error calculation unit, a synchronous tracking controller, a vibration suppression controller, a disturbance observer, and a motor control unit.

[0103] The design and manufacture of ball screw transmission systems usually have a natural frequency higher than the servo bandwidth. Therefore, the control design problem can be decomposed into two sub-problems related to different frequency ranges. That is, a synchronous tracking controller is designed in the low-frequency band to reduce the synchronization error and tracking error of the dual-screw feed system. A vibration suppression controller is designed in the high-frequency band to suppress the dynamic vibration of the worktable caused by changes in the worktable position and workpiece mass, thereby improving the dynamic characteristics and robustness of the dual-screw feed system.

[0104] The present invention decomposes a multi-objective control problem into two single-objective sub-problems, thereby proposing the following Figure 1 The synchronous control strategy of the cross-coupled parallel structure shown in the figure is composed of three controllers: synchronous tracking controller K sy , vibration suppression controller K vib and a disturbance observer K DOB . K sy and K vib As the main controller, the two form a parallel structure. Synchronous tracking controller K sy Used to minimize the reference trajectory r and the workpiece table position x w The tracking error e and the cross-coupling error e * , and maximize the closed-loop bandwidth of the system for fast tracking. Vibration suppression controller K vib By feeding back the position signals x1 and x2 of the dual feed axes on both sides of the worktable, the structural vibration caused by nonlinear force, cutting force and external interference is alleviated, and the dynamic characteristics of the system are improved. The outputs of the two controllers are u s and u v Indicates that the sum of the two is used as the input control instruction u of the servo motor m . Disturbance Observer K DOB By receiving the dual-axis positions x1 and x2 on both sides of the workbench and the motor signal u m As input, it estimates and eliminates the interference of the screw friction and low-frequency signal, and sends the deviation correction signal u to the motor d , further reducing the tracking error and synchronization error of the double-screw feed system.

[0105] The cross-coupling error calculation unit is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment; and obtain the tracking error of the single axis according to the output displacements x1 and x2 at the last moment and the reference position command r of the dual motors; and then pass through the tracking error subunit, the synchronization error subunit, and the cross-coupling error subunit in sequence to output the cross-coupling error e * .

[0106] The motion performance of the double-screw feed system needs to comprehensively consider two situations, namely the tracking error of each axis and the synchronization error between the two axes. In the cross-coupling parallel synchronization control strategy of the present invention, the mechanical coupling between the two axes is taken into account, starting from the tracking error of a single axis and then introducing the synchronization error of the two axes, and the error coupling between the two axes is used to ensure the synchronous motion of the two axes.

[0107] In the cross-coupling error calculation unit, the tracking error of a single axis is:

[0108]

[0109] Among them, r is the reference position command of the dual motors;

[0110] e1 and e2 are the tracking errors of single axis respectively;

[0111] x1 and x2 are the output displacements of the X1 and X2 axes of the double-screw feed system at the previous moment, which can be detected in real time by the grating rulers on both sides of the workbench;

[0112] The synchronization error between the two axes is as follows:

[0113]

[0114] Wherein, ε1 and ε2 are the synchronization errors of the dual axes X1 and X2 of the dual screw feed system respectively;

[0115] It can be clearly seen from the above formula that if the synchronization error is equal to 0, that is, e1=e2, then the synchronization performance between the two axes can be satisfied; however, the single-axis tracking error at this time is not necessarily 0, which will lead to poor positioning performance of the dual-drive workbench. In order to ensure the synchronization performance of the two axes while reducing the tracking error of the single axis and improving the positioning performance of the double-screw feed system, this paper introduces the cross-coupling error and writes formula (1-3) into the following matrix:

[0116] ε=Te; (1-4)

[0117] Where ε=[ε1ε2] T ;

[0118] e=[e1 e2] T ;

[0119] Represents the synchronous state transfer matrix, then the cross-coupling error can be defined as follows:

[0120] e * =e+βε; (1-5)

[0121] in,

[0122]

[0123] β=diag(β1,β2) is the cross-coupling coefficient, and β i >0(i=1,2);

[0124] Substituting formula (1-4) into formula (1-5) yields:

[0125] e * =(I+βT)e; (1-6)

[0126] Where I is the identity matrix;

[0127] (I+βT) is a positive definite matrix;

[0128] e * is the cross coupling error.

[0129] According to the definition of formula (1-6), if the cross-coupling error e * =0, then it can be deduced that e = 0, and further, the synchronization error ε = 0 can be deduced by equation (1-4). In other words, the cross-coupling error can be used to link the tracking error of the single axis with the synchronization error between the two axes. As long as the cross-coupling error is 0, the single-axis tracking error and the two-axis synchronization error are both 0. Therefore, the control goal of the present invention is to design a cross-coupling controller so that e * →0, which can ensure both good tracking performance of a single axis and synchronization performance between two axes. Figure 2 This is a block diagram of the cross-coupling synchronization error model proposed by the present invention.

[0130] The synchronous tracking controller is configured to receive the cross-coupling error output by the cross-coupling error calculation unit; and output the synchronous tracking motor control signal ue according to the cross-coupling error. sy Using H ∞ Robust controller.

[0131] Although the transfer function model of the twin-screw feed system depends on the position of the worktable and the mass of the workpiece, for the design of the synchronous tracking controller, it is usually only necessary to consider the nominal rigid body model and ignore its dynamic changes; this simplification is considered reasonable in the ball screw transmission system because the reference signal of the worktable motion is usually given at a frequency lower than the resonant mode. Therefore, the twin-screw feed system can be simplified to an LTI system, with x w =0.1m,m w =0kg means the workbench is in the middle position and the system model is studied as a nominal rigid body model when it is unloaded.

[0132] Design the synchronous tracking controller K sy The goal is to reduce the synchronization error between the two axes of the double-screw feed system, improve the positioning accuracy of the worktable, improve the control bandwidth of the system for rapid processing, and ensure the robustness of the system under closed-loop conditions. The synchronous tracking controller is designed using the LTI model, but in order to ensure that the system can suppress the influence of various uncertainties on the positioning performance, this section designs H ∞ The robust controller acts as a synchronous tracking controller, such as Figure 3 The double screw feed system H provided by the present invention is shown as follows ∞Block diagram of synchronous tracking controller. The control objective of the system is to make the closed-loop transfer function T wz (s) has the minimum H when the external disturbance input is w and the performance output is z ∞ Norm.

[0133] When designing the synchronous tracking controller, the double-screw feeding system is simplified to an LTI system, assuming that the LTI system is described by the following state space equation:

[0134]

[0135] in, is the system state vector, is the control input, To measure the output, The regulated output is also called performance output. is the external disturbance input. The disturbance considered here is uncertain (such as cutting force, noise, unmodeled dynamics, etc.), but the energy of the disturbance signal is bounded, that is, w∈L2;

[0136] Design H ∞ The state feedback controller is u=K sy x; (2-2)

[0137] Make the following closed-loop system (2-3) asymptotically stable, and the closed-loop transfer function T wz (s) satisfies the bounded real theorem (2-4);

[0138]

[0139] ||T wz (s)||=||(C1+D 12 K sy )[sI-(A+B2K sy )] -1 B1+D 11 || ∞ <1; (2-4)

[0140] Take H ∞ Suboptimal design performance γ = 1, using the LMI algorithm to obtain the positive definite matrices R and S, so that the following LMI holds:

[0141]

[0142] Among them, the matrix N 12 and N 21 Respectively and [C2,D 21 ]The basis of the null space is N 21 =ker[C2,D21 ]; According to the above analysis, the control goal is to reduce the cross-coupling error. In the low-frequency stage, the LTI nominal model of the double-screw feed system is used to design H ∞ Robust controller to improve synchronization and single-axis tracking performance of dual-screw feed systems.

[0143] Design H ∞ When the state feedback controller is used, the weight function W e is a performance index weighting function, which is used to suppress the cross-coupling error e * , so e * As the objective function, it is usually in the form of a low-pass filter. The weight function W u To control the input weight function, its function is to limit the driving voltage of the dual motors to prevent it from entering the saturation region. It is usually selected as a constant. e =[u e1 u e2 ] T is the input voltage of the dual-axis servo motor, F=[F1 F2] T It is driven by dual motors.

[0144] The solution provided by the present invention does not affect the synchronous tracking performance of the double-screw feed system in the low-frequency range, and can effectively suppress high-frequency vibration and environmental noise in the high-frequency range due to the high gain. sy is a synchronous tracking controller, G n is the LTI nominal model of the double screw feed system. e (s):=C e (sI-A e ) -1 B e +D e , W u is a constant diagonal constant matrix.

[0145] Weighting function W e , W u Combined with the system nominal model G n The augmented matrix model of the double screw feed system can be obtained as follows:

[0146]

[0147] x is the state vector matrix of the augmented system, w=[r],u e =[u e1 u e2 ] T , z e =[z e1 z e2 ] T , zu =[z u1 z u2 ] T , Figure 3 It can be seen that the augmented system model is consistent with H ∞ State feedback controller K sy To form a closed-loop control system, the system matrix of the augmented system model is:

[0148]

[0149] D 22 =[0]

[0150] Where B = [B l B r ],C=[C u C d ] T , l and r represent the left and right columns of the matrix, and u and d represent the up and down rows of the matrix; the H of the double screw feed system can be obtained by solving LMI through the hinfsyn function in Matlab. ∞ Synchronous tracking controller K sy .

[0151] The vibration suppression controller is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment; and output the vibration suppression motor control signal u according to the output displacements x1 and x2 at the last moment v . Vibration Suppression Controller K vib An LPV gain scheduling controller is adopted.

[0152] The dynamic characteristics of the double-screw feed system are time-varying. In order to suppress the dynamic time-varying vibration mode of the system and make the system have high stability and high bandwidth, the present invention designs a gain scheduling controller based on the LPV system. w and workpiece mass m w Gain adjustment is achieved to reduce the vibration peaks of different natural frequencies, meet the closed-loop stability of the double-screw feed system, and ensure the designed synchronous controller K sy synchronization performance.

[0153] Select workbench position x w and workpiece mass m w As the scheduling variable, the LPV robust controller is designed, and the double-screw feed system is described by the polyhedral model, and the following LPV model with a polyhedral structure is obtained:

[0154]

[0155] In the formula,

[0156]

[0157] For the LPV model (3-1) of the polyhedral structure, the state feedback controllers K1, K2, K3, K4 that meet the requirements are designed for the time-varying parameter polyhedron vertices ω1, ω2, ω3, ω4 respectively. They are convex polyhedrons, and the workbench is located at any position x in the convex set. w and any mass m w By combining the controller design of each vertex, we can obtain the LPV multi-cell structure controller K of the double screw feed system. vib ,Right now:

[0158]

[0159] To meet the control objectives, consider Figure 5 The closed-loop control structure shown in the figure, in order to ensure the tracking characteristics of the system, the ball screw system G(x w ,m w ) and tracking controller K sy (s) are combined to form a new system, in which K sy (s) = C sy (sI-A sy ) -1 B sy +D sy , new combined system model G n (x w ,m w ) is expressed as follows:

[0160]

[0161] in,

[0162]

[0163] C n =[C 0];

[0164] Figure 5 K is the vibration suppression controller for the double screw feed system vib Refer to the design block diagram of Figure 5 , design of vibration suppression controller K for double screw feed system vib When W is introduced F and W v The weight function designed for the controller, W F It is used to suppress the vibration caused by the motor driving force F and limit the vibration amplitude near the natural frequency; W vUsed to limit the output signal of the controller to prevent output oversaturation problems; based on the new combined LPV system G n (x w ,m w ) and the weight function W F and W v The new augmented system can be obtained as:

[0165]

[0166] in, is the state vector of the new combined system;

[0167] It is the control output of the new combined system;

[0168] y=[x1 x2] T is the measured output of the system;

[0169] Dynamic weighting function W F (s) = C F (sI-A F ) -1 B F +D F ;

[0170] W v is a constant gain, then the system matrix in the above formula can be derived as follows:

[0171]

[0172] According to LPV variable gain H ∞ The design theory of robust controller, designed vibration suppression controller K vib .

[0173] In the controller design process, the key is to select an appropriate weight function to solve the controller. The weight function W introduced in the above content of the present invention is e , W u , W F , W v Make appropriate selection to achieve the control target of the double screw feed system. Weighting function W u With W v Generally, it is a constant gain, which can effectively constrain the voltage amplitude at the control input to prevent oversaturation. u =diag(W u1 ,W u2 ), W v =diag(W v1 ,W v2 ). For the double screw feed system, without loss of generality, let W u1 =Wu2 , W v1 =W v2 . Weighting function W e and W F It is usually a dynamic transfer function, which requires multiple and repeated adjustments based on existing experience to determine the corresponding parameter value. e =diag(W e1 ,W e2 ), W F =diag(W F1 ,W F2 ), and similarly let W e1 =W e2 , W F1 =W F2 .

[0174] Adjust W e The goal is to first minimize the closed-loop transfer function T from r to e zw H ∞ norm to improve the stability of the closed-loop system; secondly, the gain of the closed-loop transfer function in the low-frequency stage is made to reduce the cross-coupling error of the system as much as possible; finally, in the closed-loop system, the crossover frequency of the transfer function is increased as much as possible to increase the servo bandwidth of the closed-loop system, and T is used to calculate the crossover frequency of the transfer function in the closed-loop system. re Indicates. F The goal of regulation is to make the closed-loop transfer function from F to x ∞ The norm is minimized to suppress the amplitude of the system under the influence of the driving force, thereby improving the dynamic performance of the system and improving the processing accuracy and stability of the CNC equipment. Fx W e and W F The parameter setting can be done using the following formula:

[0175]

[0176] Table 1. Adjustment values ​​of weight functions of cross-coupled parallel synchronous controller

[0177] Weight function k <![CDATA[M L (dB)]]> <![CDATA[M H (dB)]]> <![CDATA[ω b (Hz)]]> <![CDATA[W e1 IN e2 ]]> 2 -120 4 45 <![CDATA[W F1 IN F2 ]]> 1 -110 -180 6

[0178] Here, M L W -1 Low frequency gain, M H W -1 High frequency gain, ω b is the crossover frequency, and k is the order of the weighting function W(s). e The high frequency gain M HThe value is 2, and the order k is 1. When adjusting the parameters, it can be appropriately increased as needed, and adjusted multiple times. The method of suppressing the resonance mode can reduce the vibration amplitude of the double screw feed system during operation. At the same time, W can be increased according to needs. F The low frequency gain M L And reduce the high frequency gain M H And appropriately increase the k value. Usually W F The crossover frequency ω b Of course, it must be smaller than the first-order natural frequency of the system, W F The gain change rate is achieved by adjusting ω b After repeated adjustments, W u =diag(0.005,0.005),W v =diag(100,100). Figure 6 and Figure 7 is the weight function W e and W F The frequency domain curve after repeated adjustments, Figure 6 W e The frequency response curve of Figure 7 W F Frequency response curve; Table 1 is the weight function W e and W F Adjusted parameter value.

[0179] The disturbance observer is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the previous moment, and the output motor control signal u at the previous moment. m , output motor deviation correction signal u d .

[0180] During the operation of the double-screw feed system, it will be disturbed by various external factors such as motor load fluctuations, measurement noise, cutting force, etc. These factors will also cause positioning errors and synchronization errors in the dual-drive worktable. Therefore, the impact of external disturbances on the system must be fully considered to further improve the control accuracy of the system. During the operation of CNC equipment, the external interference of the system cannot be predicted, is constantly changing, and cannot be described by a physical model. Therefore, an interference observer must be designed to compensate and suppress it. Figure 8 The principle block diagram of the interference observer is shown, which is essentially a filter used to filter out uncertain external interference.

[0181] like Figure 8 As shown, r represents the system reference input, y represents the system output, and K(s) represents the nominal model G n (s) is the designed baseline controller, Q(s) represents the low-pass filter, n represents the feedback noise, represents the disturbance estimation, G(s) represents the controlled system, represents the disturbance estimate, represents the inverse of the controlled system. So the closed-loop transfer function from the input r to the output y is:

[0182]

[0183] exist When G ry =G(s), indicating that the disturbance observer does not affect the transfer characteristics of the system. Therefore, the closed-loop transfer function from system disturbance to output is:

[0184]

[0185] Similarly, when When Q(s) = 1, G dy = 0, which means that the disturbance has been completely suppressed and has no effect on the system output. The closed-loop transfer function from the high-frequency noise of the sensor to the output is:

[0186]

[0187] Similarly, when When Q(s)=0, G dy =0, indicating that the high-frequency noise of the sensor has been completely suppressed at this time, and the high-frequency noise output transfer function is 0. In summary, the filter Q(s) is a low-pass filter. Transforming equation (4-1) with the Mersenne formula yields:

[0188]

[0189] Therefore, simplifying the above design framework Figure 8 , can obtain Fig. 9 The simplified model shown in Figure 1. Q(s) design is a very important part of the interference observer design. In the low frequency band, Q(s)≈1. After adding the interference observer, the gain of the system in the low frequency band can be increased, thereby improving its anti-interference ability.

[0190] In the disturbance observer, the filter is a low-pass filter, which is expressed as:

[0191]

[0192] Among them, l ≥ k + b;

[0193] c0=a0;

[0194] τ is the time constant;

[0195] b is the relative order of Q(s).

[0196] Based on the design criteria of low-pass filter, Q(s) with good stability and strong anti-interference performance is designed. In motion control systems, disturbance observers can be divided into two types, one is the velocity disturbance observer with the actual model being the first-order model, and the other is the position disturbance observer with the actual model being the second-order model. Obviously, the velocity disturbance observer has a lower relative order and is easier to implement.

[0197] In the disturbance observer, the relative order of the low-pass filter is taken as 1, which is expressed as:

[0198] Q(s)=1 / (τs+1); (4-6)

[0199] By simulating the filter in Matlab, it can be determined that the time constant in the above formula is τ = 0.05.

[0200] The motor control unit tracks the motor control signal u according to the synchronization e , vibration suppression motor control signal u v And the motor deviation correction signal u d , get the motor control signal u m , used to drive the double screw feed system.

[0201] Specifically, Figure 1 As shown, the motor control unit is configured to: synchronously track the motor control signal u e , vibration suppression motor control signal u v Add them together to get u m0 ; will u m0 and motor deviation correction signal u d Subtract and get the motor control signal u m .

[0202] In order to verify the effect of the present invention, a motion control experimental platform was built. The platform can be divided into a mechanical body and a control system according to its function. Fig.10 As shown in the figure, the experimental platform has dual servo motors symmetrically arranged in the X and Y directions, which are fixed to the base through motor supports. The motor drives the ball screw through an elastic coupling to indirectly drive the workbench to move, forming a dual-screw feed system. The inductive travel switch is installed at both ends of the workbench for limit protection. In order to perform real-time detection and feedback of the dual-axis displacement of the dual-drive workbench, the experimental platform is equipped with a Fagor SA-270 high-precision grating ruler with a measurement resolution of 0.1μm and an accuracy level of ±3μm. It supports the EL 5001 communication interface and adopts a fixed installation method for the grating ruler motion reading head. In order to reduce vibration, the feed experimental platform is fixed to a heavy-duty cast vibration isolation table with 8 bolts.

[0203] Build a motion control experimental platform, which can be divided into mechanical body and control system according to its function. Fig.10 As shown in the figure, the experimental platform has dual servo motors symmetrically arranged in the X and Y directions, which are fixed to the base through motor supports. The motor drives the ball screw through an elastic coupling to indirectly drive the workbench to move, forming a dual-screw feed system. The inductive travel switch is installed at both ends of the workbench for limit protection. In order to perform real-time detection and feedback of the dual-axis displacement of the dual-drive workbench, the experimental platform is equipped with a Fagor SA-270 high-precision grating ruler with a measurement resolution of 0.1μm and an accuracy level of ±3μm. It supports the EL 5001 communication interface and adopts a fixed installation method for the grating ruler motion reading head. In order to reduce vibration, the feed experimental platform is fixed to a heavy-duty cast vibration isolation table with 8 bolts.

[0204] The dual-drive workbench is composed of X1 axis and X2 axis in the X feed direction, with a stroke of 0.2m. All the moving axes of the feed system use high-precision ball screws and linear motion guides produced by Taiwan HIWIN Company, China, which respectively play the role of driving and supporting the system. The model of the screw nut is: FDW 20-5B2, and the model of the linear guide is: HGH 20CA. All moving axes use precision angular contact ball bearings of FAG Company of Germany, and the bearing model is: FAG: B7202C.T.P4S.UL. The dual-screw feed system uses the AM8531 AC servo motor of Beckhoff Company of Germany. The motor adopts the latest single-wire cable technology and has high dynamic performance and energy efficiency. The basic parameters of the motor are shown in Table 2.

[0205] Table 2 Basic parameters of Beckhoff servo motor (AM8531)

[0206] Rated voltage V Rated power W Rated speed r / min Encoder Type 220 400 3000 Absolute <![CDATA[Moment of inertia kg·mm 2 > Rated current A Rated torque Nm Encoder pulse number 0.27 1 1.37 20bit

[0207] The control system uses a dual-channel servo driver, model AX5203, which can drive and control two servo motors at the same time, and the total current limit of the dual channels reaches 12A. It integrates fast low-latency control technology, which shortens the current control cycle to 31.2μs at the fastest, thereby achieving ultra-high dynamic positioning tasks. This article uses the Beckhoff open motion controller, model C6640-0020, which has a compact ATX motherboard equipped with high-performance components. The controller communicates with the driver, motor and peripheral devices through an EtherCAT bus, and the wiring is simple, convenient and reliable. Various electrical signals related to I / O are transmitted through the EtherCAT bus I / O terminals, including: grating scale displacement signal, proximity switch signal, limit switch signal, indicator light and other various sensor signals.

[0208] In order to verify the superiority of the cross-coupling parallel synchronization control strategy proposed in the present invention, the present invention Fig.10 The synchronous control experiment was carried out on the double-screw feed experimental platform shown in the figure. In order to maintain generality, the cross-coupling coefficient β = diag (1, 1) was selected, and the synchronization error (ε = x1-x2) and the worktable positioning error (e X =r-(x1+x2) / 2) is used to investigate the motion performance of the dual-drive feeding system. The performance indicators are designed as follows:

[0209]

[0210] In the above formula, ε max is the absolute maximum value of synchronization error, ε rms is the RMS value of synchronization error, e Xmax is the maximum absolute value of the positioning error, e Xrms is the root mean square value of the positioning error. Based on the content discussed in the previous chapter and the design of the controller, the present invention conducts the following comparative experimental studies on four synchronization control strategies: (1) Case 1: Only the cross-coupled synchronization tracking controller, i.e., K sy (2) Case 2: Using a cross-coupled parallel structure synchronous controller, namely K sy +K vib ; (3) Case 3: Using P-PI dual-loop synchronous controller based on friction feedforward compensation, namely “P-PI+friction feedforward”; (4) Case 4: Using only P-PI dual-loop synchronous controller, namely P-PI. The four control strategies are all based on the cross-coupling architecture proposed in this paper, where Case 1 and Case 2 both include the disturbance observer K DOB .

[0211] Fig.11 and Fig.12 The error curves of the double screw feed system under different control experiments are compared. Fig.11 is the synchronization error curve, Fig.12 The table positioning error curve is shown in Figure 1. First, by comparing Case 1 and Case 2, we can see that both errors in Case 1 are relatively large, especially at the ends of the lead screw. This is due to the synchronous tracking controller K sy The LTI model is designed only for the midpoint position, so the error curve shows a significant characteristic of large at both ends and small in the middle. The model parameters of the double-screw feed system will change with the position of the worktable. When it is at the two sides, the model parameters have the largest time-varying characteristics, which also verifies the time-varying characteristics of the double-screw feed system.

[0212] In Case 2, the error is small and changes evenly throughout the entire stroke. This is because Case 2 introduces the vibration suppression controller K based on Case 1. vib , and K sy Construct a parallel structure controller, in which the vibration suppression controller K vib It is designed based on the LPV model of the double-screw feed system. The LPV model has high robustness in the entire working range of the workbench, so the error of Case 2 is small in the entire working range.

[0213] Comparing Case 2 and Case 3, it can be seen that the synchronization error and tracking error of Case 3 are both large during the entire operation process. This is because the P-PI dual-loop synchronization control strategy adopted by Case 3 does not take into account the mechanical characteristics of the worktable. Although friction feedforward compensation is used, the robustness is still poor. Table 3 gives the corresponding synchronization control and tracking performance indicators of the double-screw feed system. From the vertical comparison in the table, it can be seen that Case 2 has smaller synchronization error and positioning error than Case 1 and Case 3, where the RMS of the synchronization error and positioning error are 2.25μm and 3.41μm respectively, indicating that the cross-coupling parallel structure synchronization control strategy proposed in this paper has higher accuracy and robustness, and can effectively suppress the uncertainty caused by the change of the worktable position and the change of the system dynamic characteristics, thereby effectively proving the superiority of the dual-drive parallel synchronization control strategy proposed in this paper.

[0214] In order to check the compensation effect of friction feedforward, a comparative experimental study was conducted on Case 3 and Case 4. Fig.13 and Fig.14 Table 3 shows the synchronization error and positioning error curves obtained in the experiment. The corresponding synchronization error and positioning error performance indicators are given in Table 3. Comparing the experimental results of Case 3 and Case 4, it can be seen that the synchronization error and positioning error of Case 3 are smaller than those of Case 4. The synchronization error is reduced from 10.47μm to 7.55μm, a decrease of 38.7%. The positioning error is reduced from 12.62μm to 9.35μm, a decrease of 35%. This is because Case 3 introduces the feedforward compensation link of the nonlinear friction force of the double screw on the basis of Case 4, and accurately compensates based on the nonlinear model of friction, which greatly improves the control accuracy. This also reflects that the nonlinear friction force of the screw has a great influence on the synchronization accuracy of the double screw feed system. Compensating for friction is of great significance to improving the accuracy of the double screw feed system.

[0215] Table 3 Error performance indicators of the double screw feed system under different control strategies

[0216]

[0217] It can be seen from Table 3 that the error of the solution of the present invention is significantly reduced, and the control accuracy is greatly improved.

[0218] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, ordinary technicians in this field can make other different forms of changes or modifications without creative work, which should fall within the scope of protection of the present invention.

Claims

1. A dual-drive synchronous control method with multi-controller fusion, characterized in that: The synchronous control system comprises: The cross-coupling error calculation unit is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment; and obtain the tracking error of the single axis according to the output displacements x1 and x2 at the last moment and the reference position command r of the dual motors; and then pass through the tracking error subunit, the synchronization error subunit, and the cross-coupling error subunit in sequence to output the cross-coupling error e * ; A synchronous tracking controller is configured to receive the cross-coupling error output by the cross-coupling error calculation unit; and output a synchronous tracking motor control signal u according to the cross-coupling error. e ; The vibration suppression controller is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the previous moment; and output the vibration suppression motor control signal u according to the output displacements x1 and x2 at the previous moment v ; The disturbance observer is configured to receive the output displacements x1 and x2 of the dual axes X1 and X2 of the dual screw feed system at the last moment, and the output motor control signal u at the last moment. m , output motor deviation correction signal u d ; The motor control unit synchronously tracks the motor control signal u e , vibration suppression motor control signal u v And the motor deviation correction signal u d , get the motor control signal u m , used to drive the double screw feed system.

2. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: When designing the synchronous tracking controller, the double-screw feeding system is simplified to an LTI system, which is described by the following state space equation: in, is the system state vector, is the control input, To measure the output, The regulated output is also called performance output. is the external interference input. The interference considered here is uncertain, but the interference signal energy is bounded, that is, w∈L2; Design H ∞ The state feedback controller is u=K sy x, so that the following closed-loop system (2) is asymptotically stable, and the closed-loop transfer function T wz (s) satisfies the bounded real theorem; Take H ∞ The suboptimal design performance Y = 1, the LMI algorithm is used to obtain the positive definite matrices R and S, so that the following LMI holds: Among them, the matrix N 12 and N 21 Respectively and [G2, D 21 ]The basis of the null space is N 21 =ker[C2,D 21 ]; Design H ∞ When the state feedback controller is used, the weight function W e is a performance index weighting function, which is used to suppress the cross-coupling error e * , weight function W u The weight function is used to control the input weight function, which is used to limit the driving voltage of the dual motors to prevent them from entering the saturation zone; the weight function W e , W u Combined with the system nominal model G n The augmented matrix model of the double screw feed system can be obtained as follows: x is the state vector matrix of the augmented system, w=[r],u e =[u e1 u e2 ] T , z e =[z e1 z e2 ] T , z u =[z u1 z u2 ] T , Augmented system model and H ∞ State feedback controller K sy To form a closed-loop control system, the system matrix of the augmented system model is: D 22 =[0] Where B = [B l B r ],C=[C u C d ] T , l and r represent the left and right columns of the matrix, and u and d represent the up and down rows of the matrix; the H of the double screw feed system can be obtained by solving LMI through the hinfsyn function in Matlab. ∞ Synchronous tracking controller K sy .

3. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: When designing the vibration suppression controller, a gain scheduling controller is designed based on the LPV system. w and workpiece mass m w Gain adjustment is achieved to reduce the vibration peaks of different natural frequencies, meet the closed-loop stability of the double-screw feed system, and ensure the designed synchronous controller K sy synchronization performance.

4. The dual-drive synchronous control method of multi-controller fusion as claimed in claim 3, characterized in that: Select workbench position x w and workpiece mass m w As the scheduling variable, the LPV robust controller is designed, and the double-screw feed system is described by the polyhedral model, and the following LPV model with a polyhedral structure is obtained: In the formula, For the LPV model of the polyhedral structure, the state feedback controllers K1, K2, K3, and K4 that meet the requirements are designed for the time-varying parameter polyhedron vertices ω1, ω2, ω3, and ω4 respectively. They are convex polyhedrons, and the workbench is located at any position x in the convex set. w and any mass m w By combining the controller design of each vertex, we can obtain the LPV multi-cell structure controller K of the double screw feed system. vib ,Right now:

5. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: When designing the vibration suppression controller, the ball screw system G(x w , m w ) and tracking controller K sy (s) are combined to form a new system, in which K sy (s) = C sy (s I -A sy ) -1 B sy +D sy , new combined system model G n (x w , m w ) is expressed as follows: in, , C n =[C 0]; Design of vibration suppression controller K for double screw feed system vib When W is introduced F and W v The weight function designed for the controller, W F Used to suppress the vibration caused by the motor driving force and limit the vibration amplitude near the natural frequency; W v Used to limit the output signal of the controller to prevent output oversaturation: Based on the new combined LPV system G n (x w , m w ) and the weight function W F and W v The new augmented system can be obtained as: in, is the state vector of the new combined system; It is the control output of the new combined system; y=[x1 x2] T is the measured output of the system; Dynamic weighting function W F (s) = C F (sI-A F ) -1 B F +D F ; W v is a constant gain, then the system matrix in the above formula can be derived as follows: According to the design theory of LPV variable gain H∞ robust controller, the vibration suppression controller K is designed. vib .

6. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: The interference observer is a filter, which is used to filter out uncertain external interference.

7. The dual-drive synchronous control method of multi-controller fusion according to claim 6, characterized in that: In the disturbance observer, the filter is a low-pass filter, which is expressed as: Among them, I ≥ k + b; c0=a0; τ is the time constant; t is the relative order of Q(s).

8. The dual-drive synchronous control method of multi-controller fusion as claimed in claim 7, characterized in that: In the disturbance observer, the relative order of the low-pass filter is taken as 1, which is expressed as: Q(s) = 1 / (τs+1).

9. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: The synchronous tracking motor control signal u e , vibration suppression motor control signal u v And the motor deviation correction signal u d , get the motor control signal u m , for driving a double-screw feeding system, the motor control unit is configured as: Will synchronously track the motor control signal u e , vibration suppression motor control signal u v Add them together to get u m0 ; will u m0 and motor deviation correction signal u d Subtract and get the motor control signal u m .

10. The dual-drive synchronous control method of multi-controller fusion according to claim 1, characterized in that: In the cross-coupling error calculation unit, The tracking error of a single axis is: Among them, r is the reference position command of the dual motors: e1 and e2 are the tracking errors of single axis respectively; The synchronization error between the two axes is as follows: Wherein, ε1 and ε2 are the synchronization errors of the dual axes X1 and X2 of the dual screw feed system respectively; Introducing cross-coupling error, equation (11) is written as the following matrix: ε=Te; (12) Where, ε=[ε1 ε2] T ; and=[and 1e2 ] T ; Represents the synchronous state transfer matrix, then the cross-coupling error can be defined as follows: e * =e+βε; (13) in, β = diag (β1, β2) is the cross-coupling coefficient, and β i >0(i=1,2); Substituting formula (12) into formula (13) yields: And * =(I+βT)e; Where I is the identity matrix; (I+βT) is a positive definite matrix; e * is the cross coupling error.