High-precision synchronous control method and system for double direct-drive platforms

Abstract

The invention is suitable for the technical field of motor control, and provides a high-precision synchronous control method and system for double direct drive platforms, and the method comprises the steps: loading a control strategy library which comprises a variable weight cross coupling strategy and an overall current compensation strategy; wherein a time-varying weight coefficient optimization cross coupling strategy is introduced in advance, and a variable weight cross coupling strategy is obtained; based on the variable weight cross coupling strategy, dynamically adjusting the weights of the synchronization error and the single-axis tracking error of the double direct drive platform; a set self-adaptive super-spiral sliding mode controller is adopted to suppress disturbance, the single-axis tracking precision is improved, and current instructions of all axes are generated at the same time; estimating overall disturbance through the overall current compensation strategy, and compensating the overall disturbance to current instructions of all axes to realize high-precision synchronous control of the double direct-drive platforms; according to the method, the weights of the synchronization error and the single-axis tracking error can be dynamically balanced, and the single-axis precision loss caused by the traditional fixed weight is avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor control, and provides a high-precision synchronous control method and system for a dual direct-drive platform. Background Art

[0002] In recent years, in industrial scenarios such as high-end CNC machine tools, precision manufacturing, laser processing, semiconductor manufacturing, and wafer inspection, dual direct-drive platforms have been widely used in the industrial field due to their high positioning accuracy and large thrust characteristics. However, due to factors such as unknown or perturbed mechanical coupling parameters (such as stiffness and damping), inconsistent friction damping, and load disturbances, the synchronous error of the two-axis motors is significant, resulting in internal force loss between axes and even mechanical damage.

[0003] In the prior art, the operation of a dual direct-drive platform is driven by cross-coupled control (CCC). Although cross-coupled control can improve synchronous performance, its fixed weight coefficient leads to a decrease in single-axis tracking accuracy; traditional disturbance observers cannot effectively compensate for disturbances between axes because they ignore mechanical coupling terms. Therefore, there is an urgent need for a high-precision synchronous control method that can still achieve high-precision synchronous control and reduce internal force loss when mechanical coupling parameters are unknown. Summary of the Invention

[0004] The purpose of the present invention is to provide a high-precision synchronous control method and system for a dual direct-drive platform, aiming to solve the problem that in the prior art, high-precision synchronous control cannot be achieved and there is a large internal force loss when controlling a dual direct-drive platform with unknown mechanical coupling parameters; dynamically balance the weights of synchronous error and single-axis tracking error, and avoid the loss of single-axis accuracy caused by the fixed weight of traditional disturbance observers.

[0005] The first aspect of the present invention is implemented as follows. A high-precision synchronous control method for a dual direct-drive platform includes the following steps:

[0006] Load a control strategy library, where the control strategy library includes a variable-weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient has been pre-introduced to optimize the cross-coupling strategy to obtain a variable-weight cross-coupling strategy;

[0007] Based on the variable-weight cross-coupling strategy, dynamically adjust the weights of the synchronous error and single-axis tracking error of the dual direct-drive platform;

[0008] Use a set adaptive super-twisting sliding mode controller to suppress disturbances and improve single-axis tracking accuracy, and at the same time generate current commands for each axis;

[0009] Estimate the overall disturbance through the overall current compensation strategy and compensate it to the current commands of each axis to achieve high-precision synchronous control of the dual direct-drive platform.

[0010] Further, the method further includes:

[0011] Defining a uniaxial tracking error;

[0012] Setting a time-varying weight coefficient, and constructing a comprehensive error according to the uniaxial tracking error to balance the synchronization performance and the uniaxial tracking;

[0013] Among them, the time-varying weight coefficient is denoted as ε(t),

[0014]

[0015] Among them, ε * ∈[0,1], is the initial weight coefficient, e1 = y r -y1, e2 = y r -y2, e1 and e2 are the uniaxial tracking errors of two motors in the double direct-drive platform, y r is the reference position signal; y1 and y2 are the actual positions of the two motors respectively. In this way, the synchronization error weight is dynamically adjusted through the time-varying weight coefficient ε(t) to balance the synchronization performance and the uniaxial tracking accuracy.

[0016] Further, the setting of the adaptive super-twisting sliding mode controller includes the setting of the sliding mode surface, the control law and the adaptive gain;

[0017] Among them, the setting of the sliding mode surface is:

[0018] Based on the comprehensive error, setting a second-order sliding mode surface:

[0019]

[0020] Among them, c1 and c2 are positive design parameters used to adjust the error convergence speed; is the derivative of the comprehensive error, representing the dynamic characteristics of the error change;

[0021] The setting of the control law and the adaptive gain includes:

[0022] Defining the control law:

[0023]

[0024] β(t) = k3α(t),

[0025] Among them, w i is the intermediate control variable used to calculate the final control input, u i is the auxiliary control variable that plays an auxiliary adjustment role in the control law; the sliding mode surface function is s i (i = 1, 2), α(t) and β(t) represent the adaptive gain, is the derivative of α(t);

[0026] α sw is the gain switching threshold, which is used to balance the rapid response and stability of the control gain;

[0027] u is the thickness of the sliding mode boundary layer, which is used to suppress chattering;

[0028] k1, k2, and k3 are adaptive gain adjustment parameters.

[0029] Furthermore, the high-precision synchronous control method for the dual direct drive platform further includes: setting an overall current compensation strategy and storing it in the control strategy library, specifically including:

[0030] Establishing a dynamic model of the dual direct drive platform;

[0031] Based on the dynamic model of the dual direct drive platform, setting an overall disturbance observer;

[0032] Among them, the dynamic model of the dual direct drive platform satisfies:

[0033]

[0034] Among them,

[0035] i q1 and i q2 represent the q-axis currents of the two motors, k f represents the thrust coefficient of the motor, represents the derivative of the average speed of the two axes, represents the derivatives of the speeds of the two motors, M represents the mass of the mover in the dual direct drive platform, and m represents the mass of the crossbeam; d1 and d2 represent the uniaxial disturbances received by the two motors;

[0036] The overall disturbance observer satisfies:

[0037]

[0038] Among them, is the sliding mode error function, where a > 0, b > 1, p1, p2 < 0, q ∈ (0, 1), represents the observed value of the total disturbance and its differential, represents the observed error of the average acceleration, r is the observer gain, which is used to adjust the disturbance estimation speed.

[0039] In the second aspect of the present invention, a high-precision synchronous control system for a dual direct drive platform is provided. The high-precision synchronous control system for the dual direct drive platform includes:

[0040] A data management module, configured to load a control strategy library, where the control strategy library includes a variable-weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient has been pre-introduced to optimize the cross-coupling strategy to obtain the variable-weight cross-coupling strategy;

[0041] A weight adjustment module, based on the variable-weight cross-coupling strategy, dynamically adjusts the weights of the synchronization error and the single-axis tracking error of the dual direct-drive platform;

[0042] An adaptive control module, configured to use a set adaptive super-twisting sliding mode controller to suppress disturbances, improve the single-axis tracking accuracy, and generate current commands for each axis;

[0043] A current compensation module, which can estimate the overall disturbance through the overall current compensation strategy and compensate it to the current commands of each axis to achieve high-precision synchronization control of the dual direct-drive platform.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] In the present invention, based on the variable-weight cross-coupling strategy, by dynamically adjusting the time-varying weight coefficient ε(t), the synchronization error is preferentially suppressed when the synchronization error is large, and the single-axis accuracy is preferentially improved when the single-axis tracking error is large; the weights of the synchronization error and the single-axis tracking error are dynamically balanced to avoid the loss of single-axis accuracy caused by traditional fixed weights. The set adaptive super-twisting sliding mode controller adjusts the gain according to the operating state data to suppress chattering and enhance the anti-disturbance ability, making the control law smoother and reducing the generation of chattering. The overall current compensation strategy can eliminate the influence of unknown mechanical coupling parameters on synchronization control and reduce the internal force loss between axes. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a schematic diagram of the structure and mechanical decomposition of the dual direct-drive platform in the embodiment of the present invention;

[0047] Figure 2 is a schematic diagram of the synchronous motion of the dual direct-drive platform in the embodiment of the present invention;

[0048] Figure 3 is a waveform of the control gain change of a high-precision synchronization control method for a dual direct-drive platform in the embodiment of the present invention;

[0049] Figure 4 、 Figure 5 are respectively the weight coefficient and position waveforms when the motor is no-load and the centroid of the crossbeam is offset;

[0050] Figures 6 to 12 is an experimental comparison waveform of various disturbance controls in this embodiment;

[0051] Figure 13Flow chart of a high-precision synchronization control method for a dual direct drive platform provided by an embodiment of the present invention;

[0052] Figure 14 Block diagram of the structure of a high-precision synchronization control system for a dual direct drive platform provided by an embodiment of the present invention;

[0053] In the drawings: 101 - First ball guide rail; 102 - Limit spring; 103 - Permanent magnet; 104 - Cross beam; 105 - Second ball guide rail; 106 - Heavy object. Detailed implementation manners

[0054] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] Glossary:

[0056] Cross-Coupled Control with Variable weight, CCCV;

[0057] Compensate of Overall Current, COC;

[0058] PMLSM: Permanent Magnet Linear Synchronous Motor.

[0059] As Figure 13 shown, in one embodiment, a high-precision synchronization control method for a dual direct drive platform is proposed. The method may specifically include the following steps S101 to S104;

[0060] Step S101: Load a control strategy library, where the control strategy library includes a variable weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient has been pre-introduced to optimize the cross-coupling strategy to obtain the variable weight cross-coupling strategy;

[0061] Step S102: Based on the variable weight cross-coupling strategy, dynamically adjust the weights of the synchronization error and the single-axis tracking error of the dual direct drive platform;

[0062] Step S103: Use a set adaptive super-twisting sliding mode controller to suppress disturbances, improve the single-axis tracking accuracy, and generate current commands for each axis;

[0063] Step S104: Estimate the overall disturbance through the overall current compensation strategy and compensate it to the current commands of each axis to achieve high-precision synchronization control of the dual direct drive platform.

[0064] Exemplarily, asFigure 1 As shown, the dual direct drive platform applied in this embodiment includes two motors. For the specific structure, please refer to Figure 1 (a) in, which includes: ball guide rail 101 and ball guide rail 105 that are parallel to each other and fixed through a base. Permanent magnets 103 are linearly arranged and fixed on ball guide rail 101 and ball guide rail 105; two sliders that are slidably connected to ball guide rail 101 and ball guide rail 105, and a cross beam 104 connected between the two sliders. Limiting springs 102 are connected to both sides of the connection between cross beam 104 and the slider;

[0065] Please refer to Figure 1 (b) in, which shows the uniform motion when the friction damping is unequal: revealing the synchronous error caused by the difference in friction force. (c) shows the accelerated motion when the centroid deviates: explaining the internal axial force caused by the inertial force.

[0066] In Figure 2 , (a), initial asynchronous state: the single-axis tracking error and the synchronous error coexist. (b), high-weight synchronization priority process: ε(t) increases, and the synchronous error is preferentially eliminated. (c), low-weight tracking priority process: ε(t) decreases, focusing on single-axis positioning. (d), synchronous completion state: the errors all converge to zero.

[0067] In this embodiment, the method further includes:

[0068] Defining the single-axis tracking error; the specific equation can be expressed as follows:

[0069] e1 = y r - y1, e2 = y r - y2 (1);

[0070] e1 and e2 are the single-axis tracking errors of the two motors in the dual direct drive platform, and y r is the reference position signal; y1 and y2 are the actual positions of the two motors respectively. Please refer to Figure 1 (a) in.

[0071] Setting a time-varying weight coefficient, constructing a comprehensive error based on the single-axis tracking error to balance the synchronization performance and single-axis tracking; the comprehensive error is expressed as follows:

[0072] e c1 = e1 + ε(t) * (e1 - e2), e c2 = e2 + ε(t) * (e2 - e1) (2);

[0073] Among them, the time-varying weight coefficient is denoted as ε(t),

[0074]

[0075] Among them, ε* ∈ [0,1] is the initial weight coefficient, which ensures that the synchronization error weight changes dynamically with the error ratio. In this way, through the time-varying weight coefficient ε(t), the synchronization error is preferentially suppressed when the synchronization error is large, and the single-axis tracking accuracy is preferentially improved when the single-axis tracking error is large; realizing dynamic adjustment of the synchronization error weight and balancing the synchronization performance and the single-axis tracking accuracy, as Figure 2 shown.

[0076] Exemplarily, the setting of the adaptive super-twisting sliding mode controller includes the setting of the sliding mode surface, the control law, and the adaptive gain;

[0077] The adaptive super-twisting sliding mode controller uses the Adaptive Super-Twisting Algorithm (ASTA) to converge the comprehensive error to 0, suppress the matching disturbances (such as inconsistent friction damping and load changes), and improve the tracking accuracy.

[0078] Among them, the setting of the sliding mode surface is:

[0079] Based on the comprehensive error, a second-order sliding mode surface is set:

[0080]

[0081] where c1 and c2 are positive design parameters used to adjust the error convergence speed; is the derivative term of the comprehensive error, which characterizes the dynamic characteristics of the error change;

[0082] The setting of the control law and the adaptive gain includes:

[0083] Define the control law:

[0084]

[0085] β(t) = k3α(t) (6),

[0086] where w i is the intermediate control variable used to calculate the final control input, and u i is the auxiliary control variable that plays an auxiliary adjustment role in the control law;

[0087] Refer to Figure 3 ; The waveforms of the adaptive gains α1 and α2 of the two motors changing in real time with time, as can be seen from Figure 3 when the load is opposite to the motor, α rises from the given α sw = 80, and then drops back to α after the disturbance is suppressed swThis effectively suppresses the influence of matching disturbance on single-axis tracking, making the synchronization control accuracy and single-axis control accuracy of CCCV (variable weight cross-coupling control)-ASTA (adaptive super-helical sliding mode control algorithm, or adaptive super-helical algorithm) better than CCCV-STA (super-helical algorithm).

[0088] In the control law, w i Combined sliding surface s i Information and auxiliary control variables u i The impact of u i The derivative of According to the sliding surface s i The sign of is adjusted so that the controller of the dual direct drive platform can quickly respond to changes in the state of the system (i.e., the dual direct drive platform servo system);

[0089] The sliding surface function is s i (i=1,2), i=1,2 correspond to the sliding mode surfaces of the two motors respectively; ɑ(t), β(t) represent adaptive gains, is the derivative of α(t); thus, the obtained control law is an adaptive super-helical sliding mode control law, which suppresses chattering and enhances anti-disturbance capability through dynamic gain adjustment;

[0090] α(t) and β(t) are functions of time t and are dynamically adjusted based on the servo system's operating state. They suppress chattering, a common phenomenon in sliding-mode control, and enhance the servo system's anti-interference capabilities. Dynamically adjusting α(t) and β(t) smoothes the control law and reduces chattering.

[0091] α sw is the gain switching threshold, which is used to balance the fast response and stability of the control gain;

[0092] u is the thickness of the sliding mode boundary layer, which is used to suppress buffeting;

[0093] k1, k2, and k3 are adaptive gain adjustment parameters. In this way, the adaptive gain of the control law is dynamically adjusted according to the state of the sliding surface. When the sliding mode variable exceeds the boundary layer, the gain is increased to quickly suppress the disturbance, and vice versa, the gain is reduced to suppress chattering.

[0094] exist Figure 4 、 Figure 5 middle, Figure 4 The weight change when no-load is shown: the weight coefficient ε(t) increases significantly at the maximum speed (position zero point) when no-load, corresponding to the dominant error caused by inconsistent friction damping; Figure 5 The figure shows the weight change when the beam center of mass shifts: the weight coefficient increases significantly at the point of maximum acceleration when the beam center of mass shifts, corresponding to the dominant error of the inertial coupling term.

[0095] In this embodiment, in order to eliminate the influence of unknown mechanical coupling parameters on synchronous control and reduce the internal force loss between axes, the high-precision synchronous control method for the dual direct-drive platform further includes: setting an overall current compensation strategy and storing it in the control strategy library, specifically including:

[0096] Establishing a dynamic model of the dual direct-drive platform; the dynamic model can refer to (c) in Figure 1 ;

[0097] Based on the dynamic model of the dual direct-drive platform, setting an overall disturbance observer;

[0098] Among them, the dynamic model of the dual direct-drive platform satisfies:

[0099]

[0100] Among them,

[0101] i q1 、i q2 represent the q-axis currents of two motors, k f represents the thrust coefficient of the motor, represents the derivative of the average speed of the two axes, represents the derivatives of the speeds of two motors, that is, accelerations; M represents the mass of the mover in the dual direct-drive platform, m represents the mass of the crossbeam; d1, d2 represent the uniaxial disturbances received by the two motors, specifically including disturbances caused by factors such as friction, load change, coupling terms, etc., which are lumped disturbances;

[0102] The q-axis current is mainly used to generate electromagnetic thrust to drive the motor to move. In the overall current compensation strategy, it is necessary to adjust i q1 、i q2 according to the dynamic model and the disturbance observation results to compensate for the disturbance of the servo system and improve the synchronization performance of the system.

[0103] The overall disturbance observer in this embodiment satisfies:

[0104]

[0105] Among them, is the sliding mode error function, which is used to quickly converge the disturbance estimation value; where a>0, b>1, p1, p2<0, q∈(0,1), represents the observed value of the total disturbance and its differential, represents the observed error of the average acceleration, r is the observer gain, which is used to adjust the disturbance estimation speed.

[0106] In the overall disturbance observer, a and b are the design parameters of the sliding mode surface S. Specifically, a is the gain coefficient used to adjust the speed of the reaching law. Increasing a can accelerate the error convergence. b is the exponential term coefficient that controls the dynamic characteristics of the reaching law. p1 and p2 are the linear and nonlinear term coefficients in the reaching law equation. p1 adjusts the weight of the linear term and controls the stability of the reaching speed. p2 adjusts the weight of the nonlinear term and enhances the robustness to errors. q is the exponent of the nonlinear term in the reaching law, which controls the nonlinear degree of the reaching law.

[0107] In step S104 of this embodiment, the overall disturbance is estimated through the overall current compensation strategy and compensated to the current commands of each axis. The specific operations can be as follows:

[0108] After obtaining the overall disturbance observation value it is possible to use Equation (7) to calculate the sum of the ideal q-axis currents:

[0109]

[0110] where, (i q1 + i q2 ) * represents the sum of the ideal q-axis currents, which can be calculated through the dynamic model and disturbance estimation, namely Equations (7) and (8);

[0111] After that, the difference is taken between the sum of the ideal q-axis currents and the sum of the q-axis current commands calculated by the adaptive super-twisting sliding mode controller; half of this difference, that is, is used as the current compensation amount and distributed to each axis to suppress the internal force loss.

[0112] As Figures 6 to 12 shown, this is the result of the comparative experiment on a high-precision synchronous control method for a dual direct-drive platform in this embodiment.

[0113] Figure 6 This is a comparison between CCCV (variable-weight cross-coupling control) and traditional CCC (cross-coupling control). Specifically, it is a comparison between CCCV and CCC under disturbance observation and an eccentric load of 80 N. It includes the synchronous error, single-axis tracking error #1, single-axis tracking error #2, and q-axis current difference.

[0114] It can be seen from Figure 6 that when using CCCV, the synchronous error of CCCV (26.217 μm) is lower than that of using CCC (35.392 μm). The single-axis tracking errors are 236.836 μm and 218.662 μm respectively, which are significantly lower than 268.529 μm and 247.928 μm of CCC. In addition, the q-axis current difference of 0.771 A is lower than 0.917 A. This indicates that the CCCV proposed in this embodiment has higher synchronous control performance compared to the traditional CCC.

[0115] Figure 7 It is a comparison between ASTA (Adaptive Supertwisting Algorithm) and the traditional STA (Supertwisting Algorithm). Specifically: the comparison between CCCV-ASTA and CCCV-STA with disturbance observation and an eccentric load of 80N. It includes synchronization error, uniaxial tracking error #1, uniaxial tracking error #2, and q-axis current difference.

[0116] From Figure 7 It can be seen that the synchronization control error of CCCV-ASTA (17.066μm) is slightly lower than that of CCCV-STA (18.238μm). The uniaxial tracking errors are 228.029μm and 240.612μm respectively, which are significantly lower than 273.675μm and 287.299μm of CCCV-STA. In addition, the q-axis current difference of 0.216A is much smaller than 0.874A of the latter. This shows that although the improvement of ASTA in synchronization error compared with STA is not obvious, it has an obvious inhibitory effect on the internal force loss between the two axes.

[0117] Figures 8 to 12 It is a comparison between the COC (Overall Current Compensation) strategy and the traditional DSMO (Dual-axis Disturbance Observer). Specifically,

[0118] Figure 8 : The comparison between CCCV-ASTA-COC and CCCV-ASTA-DSMO when the crossbeam is unloaded and the centroid of the crossbeam is not artificially changed. It includes synchronization error, uniaxial tracking error #1, uniaxial tracking error #2, and q-axis current difference.

[0119] Figure 9 : The comparison when a 140N heavy object ( Figure 1 of 106) is applied at the centroid of the crossbeam, and the comparison dimensions are the same as above. Figure 10 : The comparison when a 140N heavy object is applied at the eccentric position of the crossbeam, and the comparison dimensions are the same as above. Figure 11 : The comparison when an 80N heavy object is loaded at the centroid of the crossbeam, and the comparison dimensions are the same as above. Figure 12 : The comparison when an 80N heavy object is loaded at the eccentric position of the crossbeam, and the comparison dimensions are the same as above.

[0120] The application of the heavy object here means: directly placing the heavy object 106 on the crossbeam 104, and the force on the crossbeam 104 is in the vertical direction. Refer to Figure 1 ; Loading means: connecting the heavy object to the crossbeam through a rope, and the heavy object is suspended in the air about 10 - 80cm above the ground (changing with the movement position of the motor). The force on the crossbeam is in the horizontal direction and opposite to the movement direction of the motor.

[0121] From Figure 8It can be seen that the synchronization error and the single-axis tracking error under the COC algorithm are significantly lower than those of the DSMO algorithm when the speed and acceleration are at their maximum. The following experimental data tables (Table 1 and Table 2) show that the synchronization error of COC is 8.205μm, lower than 9.725μm of DSMO. The single-axis tracking errors are 127.851μm and 133.477μm respectively, lower than 132.721μm and 138.818μm of DSMO. The q-axis current difference is 0.154A, lower than 0.167A of DSMO.

[0122] It can be seen that Figure 9 due to the inconsistency between the artificially applied heavy mass center and the ideal position, when a 140N heavy object is applied to the crossbeam mass center, there is a gap in the synchronization error of COC and DSMO in the neighborhood of each maximum acceleration moment. In addition, due to the change in the crossbeam mass, the friction damping term of the single-axis motor increases, and there are significant differences between COC and DSMO at each moment of maximum speed. The following table data show that the synchronization error of COC is 12.609μm, lower than 17.310μm of DSMO. The single-axis tracking errors are 135.026μm and 132.065μm respectively, much lower than 177.814μm and 189.448μm of DSMO. The q-axis current difference of 0.214A is significantly lower than 0.253A of DSMO.

[0123] It can be seen that Figure 10 when a 140N heavy object is applied to the eccentric position of the crossbeam, the center-of-mass position of the system is changed, and obvious differences in the synchronization error of the two algorithms appear in the neighborhood of the maximum speed and acceleration moments. In addition, the difference in the q-axis current is more obvious than Figure 9 that. The following experimental data tables (Table 1 and Table 2) show that the synchronization error of COC is 18.162μm, significantly lower than 29.805μm of DSMO. The single-axis tracking errors are 156.148μm and 182.454μm respectively, both significantly lower than 191.846μm and 207.163μm of DSMO. The q-axis current difference is 0.218A, lower than 0.383A of DSMO.

[0124] It can be seen that Figure 11 and Figure 12 when 80N is loaded at the center of mass, when the load is added, a spike appears in the single-axis tracking error of DSMO that does not exist in COC. More importantly, it can be found that at the starting moment (i.e., after 1s), the synchronization error of DSMO is greater than that of COC. This is because the acceleration at the starting moment is greater than any subsequent moment, and the center-of-mass load exacerbates the asynchronousness of the two motors, and the eccentric load is even more so. Intuitively, the synchronization error at the starting moment with the center-of-mass load is about 200μm, and about 350μm with the eccentric load. The remaining comparison data can be seen in the following experimental data tables, namely Table 1 and Table 2.

[0125] Table 1 is the summary table of the core data of the comparative experiment

[0126]

[0127] Table 2 is the summary table of the core data of the comparative experiment

[0128]

[0129] The experimental results of the above embodiments show that through this method, the synchronization accuracy can be improved, the single-axis tracking can be optimized, and the internal force loss between axes can be reduced; at the same time, it can adapt to unknown or dynamically changing mechanical coupling parameters and improve the robustness during error suppression.

[0130] As Figure 14 shown, in another embodiment, a high-precision synchronization control system for a dual direct-drive platform, the high-precision synchronization control system 100 for the dual direct-drive platform includes:

[0131] A data management module 110, configured to load a control strategy library, where the control strategy library includes a variable-weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient is pre-introduced to optimize the cross-coupling strategy to obtain the variable-weight cross-coupling strategy;

[0132] A weight adjustment module 120, based on the variable-weight cross-coupling strategy, dynamically adjusts the weights of the synchronization error and the single-axis tracking error of the dual direct-drive platform;

[0133] An adaptive control module 130, configured to use a set adaptive super-twisting sliding mode controller to suppress disturbances, improve the single-axis tracking accuracy, and generate current commands for each axis;

[0134] A current compensation module 140, capable of estimating the overall disturbance through the overall current compensation strategy and compensating it to the current commands of each axis to achieve high-precision synchronization control of the dual direct-drive platform.

[0135] In this embodiment, the high-precision synchronization control system for the dual direct-drive platform, as the software part, can be actually verified through a specified dual direct-drive platform servo system; the dual direct-drive platform servo system among them includes: a drive module, an actuator, a controller, and a sensor; the specific hardware configuration is as follows:

[0136]

[0137] The specific parameter configuration is as follows:

[0138] Inner current loop parameters <![CDATA[k p = 10, k i = 5000]]> Position loop CCCV - ASTA parameters <![CDATA[α sw = 80, k1 = k2 = 1000, k3 = 125, c1 = 300, c2 = 1, ε * = 0.05]]> Disturbance observer parameters <![CDATA[p1 = 500, p2 = 1000, g = 0.9, a = 300, b = 1.2, r = -7]]>

[0139] for reference Figures 6 to 12 , and the specific experimental procedure is as follows:

[0140] First, comparative experiment design

[0141] First step, CCCV vs CCC: (corresponding to Figure 6 )

[0142] Condition: Eccentric load of 80 N, and the trajectory of the input reference signal is a sine signal y r = 0.15sin(5t) (maximum speed 0.75 m / s, maximum acceleration 3.75 m / s 2 ).

[0143] Result: The synchronization error of CCCV is 26.217 μm, and that of CCC is 35.392 μm (a decrease of 26%). The q-axis current difference decreases from 0.917 A to 0.771 A, a reduction of 16%, indicating a decrease in the internal force loss between the shafts. The synchronization error of CCCV is significantly lower than that of CCC, and both the single-axis tracking error and the q-axis current difference are smaller, verifying the advantages of CCCV.

[0144] Second step, ASTA vs STA: (corresponding to Figure 7 )

[0145] Condition: The same input reference signal trajectory and load, with a disturbance observer. The input reference signal is the same as y in the previous experiment r = 0.15sin(5t), the load is an eccentricity of 80 N, and the direction is opposite to the movement direction.

[0146] Result: The current difference of ASTA is 0.216 A, and that of STA is 0.874 A (a decrease of 75%, significantly reduced). The synchronization error of ASTA (17.066 μm) is slightly better than that of STA (18.238 μm); both the synchronization error and the current difference are reduced, proving that ASTA has stronger anti-disturbance ability.

[0147] Third step, COC vs DSMO: (corresponding to Figures 8 to 12 )

[0148] Condition: The two motors are connected by a crossbeam (mechanical coupling exists), and are tested without load, with a 140 N weight applied to the centroid / eccentricity of the crossbeam, and with an 80 N weight applied to the centroid / eccentricity of the crossbeam. The input reference signal is still a sine trajectory, the same as before.

[0149] Result: When a 140 N weight is applied to the centroid, the synchronization error of COC is 8.205 μm, and that of DSMO is 9.725 μm (a decrease of 15%);

[0150] When a 140 N weight is applied eccentrically, the synchronization error of COC is 18.162 μm, and that of DSMO is 29.805 μm (a decrease of 39%). The remaining experimental data are shown in Table 1 and Table 2.

[0151] Second, experimental conclusion:

[0152] CCCV: The dynamic weight effectively balances synchronization and single-axis performance and is applicable to load change scenarios.

[0153] ASTA: The adaptive gain significantly improves the anti-interference ability and reduces internal force loss.

[0154] COC: Overall compensation does not require coupling parameters and has better performance when coupling exists or the centroid is offset.

[0155] A high-precision synchronization control method for a dual direct-drive platform provided in this embodiment, and a high-precision synchronization control system for a dual direct-drive platform is provided based on this high-precision synchronization control method for a dual direct-drive platform. This high-precision synchronization control method for a dual direct-drive platform is based on a variable weight cross-coupling strategy. By dynamically adjusting the time-varying weight coefficient ε(t), when the synchronization error is large, it preferentially suppresses the synchronization error, and when the single-axis tracking error is large, it preferentially improves the single-axis accuracy; dynamically balances the weights of the synchronization error and the single-axis tracking error, avoiding the loss of single-axis accuracy caused by traditional fixed weights. And the set adaptive super-twisting sliding mode controller, according to the operating state data, suppresses chattering and enhances the anti-interference ability through dynamic gain adjustment, making the control law smoother and reducing the generation of chattering. And the overall current compensation strategy can eliminate the influence of unknown mechanical coupling parameters on synchronization control and reduce the internal force loss between axes.

[0156] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the appended claims.

[0157] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high-precision synchronous control method for a double direct-drive platform, characterized in that, The high-precision synchronous control method for the dual direct-drive platform includes the following steps: Load the control strategy library, which includes a variable-weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient has been pre-introduced to optimize the cross-coupling strategy to obtain the variable-weight cross-coupling strategy; Based on the variable-weight cross-coupling strategy, dynamically adjust the weights of the synchronous error and the single-axis tracking error of the dual direct-drive platform; Use a set adaptive super-twisting sliding mode controller to suppress disturbances, improve the single-axis tracking accuracy, and generate current commands for each axis at the same time; Estimate the overall disturbance through the overall current compensation strategy and compensate it to the current commands of each axis to achieve high-precision synchronous control of the dual direct-drive platform.

2. The high-precision synchronous control method for a dual direct-drive platform according to claim 1, wherein The high-precision synchronous control method for the dual direct-drive platform further includes the following steps: Define the single-axis tracking error; Set the time-varying weight coefficient, and construct a comprehensive error based on the single-axis tracking error to balance the synchronous performance and the single-axis tracking; Among them, the time-varying weight coefficient is denoted as ε(t), Among them, ε * ∈ [0, 1], is the initial weight coefficient, e1 = y r - y1, e2 = y r - y2, e1 and e2 are the single-axis tracking errors of the two motors in the dual direct drive platform, y r is the reference position signal; y1 and y2 are the actual positions of the two motors respectively.

3. The high-precision synchronous control method for a dual direct-drive platform according to claim 1, characterized in that, The setting of the adaptive super-twisting sliding mode controller includes the setting of the sliding mode surface, the control law and the adaptive gain; Among them, the setting of the sliding mode surface is: Based on the comprehensive error, set the second-order sliding mode surface: Among them, c1 and c2 are positive design parameters used to adjust the error convergence speed; is the differential term of the comprehensive error, representing the dynamic characteristics of the error change; The setting of the control law and the adaptive gain includes: Define the control law: β(t) = k3α(t), where, w i is an intermediate control variable for calculating the final control input, and u i is an auxiliary control variable that plays an auxiliary adjustment role in the control law; the sliding mode surface function is s i (i = 1, 2), where α(t) and β(t) represent the adaptive gains, is the derivative of α(t); α sw is the gain switching threshold for balancing the fast response and stability of the control gain; u is the thickness of the sliding mode boundary layer, which is used to suppress chattering; k1, k2, k3 are adaptive gain adjustment parameters.

4. The high-precision synchronous control method for the double direct drive platform according to claim 1, characterized in that The high-precision synchronous control method for the dual direct-drive platform further includes: setting the overall current compensation strategy and storing it in the control strategy library, specifically including: Establish the dynamic model of the dual direct-drive platform; Based on the dynamic model of the dual direct-drive platform, set the overall disturbance observer.

5. The high-precision synchronous control method for a dual direct-drive platform according to claim 4, characterized in that The dynamic model of the dual direct-drive platform satisfies: Among them, i q1 and i q2 represent the q-axis currents of two motors, k f represents the thrust coefficient of the motor, represents the derivative of the average speed of the two axes, represents the derivative of the speeds of two motors, M represents the mass of the mover in the double direct drive platform, m represents the mass of the crossbeam; d1, d2 represent the uniaxial disturbances received by the two motors.

6. The high-precision synchronous control method for the dual direct drive platform according to claim 5, characterized in that The overall disturbance observer satisfies: Among them, is the sliding mode error function, where a > 0, b > 1, p1, p2 < 0, q ∈ (0, 1), represents the observed value of the total disturbance and its differential, represents the observed error of the average acceleration, r is the observer gain, used to adjust the disturbance estimation speed.

7. A high-precision synchronization control system for a dual direct-drive platform, which is used for the high-precision synchronization control method of the dual direct-drive platform as described in any one of claims 1-6, characterized in that, The high-precision synchronous control system for the dual direct-drive platform includes: A data management module for loading the control strategy library, which includes a variable-weight cross-coupling strategy and an overall current compensation strategy; among them, a time-varying weight coefficient has been pre-introduced to optimize the cross-coupling strategy to obtain the variable-weight cross-coupling strategy; A weight adjustment module that dynamically adjusts the weights of the synchronous error and the single-axis tracking error of the dual direct-drive platform based on the variable-weight cross-coupling strategy; An adaptive control module for using a set adaptive super-twisting sliding mode controller to suppress disturbances, improve the single-axis tracking accuracy, and generate current commands for each axis at the same time; A current compensation module that can estimate the overall disturbance through the overall current compensation strategy and compensate it to the current commands of each axis to achieve high-precision synchronous control of the dual direct-drive platform.

Images