A variable-speed servo motor cogging torque suppression method and device based on double-channel equivalent input disturbance

Abstract

This invention discloses a method for suppressing cogging torque in a variable-speed servo motor based on dual-channel equivalent input disturbances, comprising: S1, establishing a state-space model of a permanent magnet synchronous servo motor to form equivalent input disturbances; S2, outputting the total equivalent disturbance signal using an adaptive super-helical sliding mode observer; S3, calculating the time-varying delay of the rotor and constructing a time-varying delay operator; S4, generating a periodic disturbance compensation amount using the time-varying delay operator and a low-pass filter; S5, adding the remaining disturbance compensation amount to obtain the total disturbance compensation amount; S6, superimposing the total disturbance compensation amount with the basic control amount to generate the control input amount. This invention also discloses a device for suppressing cogging torque in a variable-speed servo motor based on dual-channel equivalent input disturbances. This invention can solve the problem of noise amplification and chattering caused by the coupling between high-frequency periodic components and low-frequency residual components in cogging torque and the observation gain, and can be widely applied in the field of permanent magnet synchronous servo motor control.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet synchronous servo motor control technology, and in particular to a method and device for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are characterized by high power density, compact structure, and good dynamic performance, and are widely used in high-precision industrial servo systems. Due to the interaction between the permanent magnets and the stator slots, the motor generates cogging torque that varies periodically with the rotor position during rotation. This disturbance, combined with load variations, nonlinear friction, and model parameter perturbations, can cause speed ripple and reduce closed-loop tracking accuracy.

[0003] Fixed-delay repetitive controllers can match the fixed time period of cogging torque under constant speed conditions. However, when the motor performs acceleration, deceleration, or continuous speed change tasks, the instantaneous frequency of the cogging torque changes with the speed, and the frequency matching capability of the fixed-delay internal mode decreases. Although spatial domain repetitive control and variable sampling repetitive control can handle spatial periodic disturbances, they respectively have the problems of difficulty in integration with conventional time-domain control frameworks or additional requirements for hardware sampling.

[0004] On the other hand, repetitive control alone is insufficient to compensate for aperiodic loads and unmodeled dynamics. Existing equivalent input disturbance control typically treats the observed combined disturbance as a single signal for direct compensation, causing the same observation and compensation path to simultaneously handle high-frequency periodic cogging torque and low-frequency load disturbances. Furthermore, the high observation gain set to track rapid oscillation components easily amplifies measurement noise and introduces steady-state chattering. Therefore, a servo motor disturbance suppression method is needed that can maintain periodic matching under variable speed conditions and allocate compensation tasks according to the disturbance frequency characteristics. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the above-mentioned background technology and provide a method and device for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance. This invention aims to solve the problems of the period of cogging torque changing with the speed under variable speed conditions, the mutual coupling of high-frequency periodic components and low-frequency residual components in the comprehensive disturbance, and the noise amplification and chattering caused by a fixed high observation gain.

[0006] This invention provides a method for suppressing cogging torque in a variable-speed servo motor based on dual-channel equivalent input disturbance, comprising the following steps: S1. Establish a state-space model of the permanent magnet synchronous servo motor, and project the combined effects of cogging torque, load torque, friction and model parameter perturbation on the motor output onto the control input channel to form an equivalent input disturbance. S2. Construct an observation error based on the output measurement value and state observation value of the state space model, input the observation error into an adaptive superhelical sliding mode observer, and output the total equivalent disturbance signal by adaptively adjusting the superhelical gain; wherein, the superhelical gain increases with the increase of the observation error, and has a preset minimum robust gain lower limit, and decays towards the minimum robust gain lower limit after the observation error converges; S3. Determine the basic spatial period of cogging torque based on the number of stator slots and rotor poles of the permanent magnet synchronous motor, and calculate the time-varying delay required for the rotor to cross one basic spatial period based on the basic spatial period and the rotor mechanical angular velocity, and then construct a time-varying delay operator. S4. Input the total equivalent disturbance signal into the time-varying repetitive controller, and the time-varying repetitive controller uses the time-varying delay operator and low-pass filter to generate a periodic disturbance compensation amount that matches the spatial period of the cogging torque. S5. Based on the remaining disturbance components in the total equivalent input disturbance that were not compensated by the periodic disturbance compensation amount, generate the remaining disturbance compensation amount through the remaining disturbance integral compensation channel, and add the periodic disturbance compensation amount to the remaining disturbance compensation amount to obtain the total disturbance compensation amount. S6. The total disturbance compensation quantity is superimposed with the basic control quantity to generate the system control input quantity. The basic control quantity is jointly generated by the state feedback loop and the speed error integral loop. The state feedback loop generates the state feedback control quantity based on the deviation between the state reference vector and the state observation value through the feedback gain matrix. The speed error integral loop generates the speed error integral quantity based on the deviation between the rotor mechanical angular velocity reference value and the actual measured value through the integral gain. The final control input quantity is obtained by adding the state feedback control quantity, the speed error integral quantity and the total disturbance compensation quantity.

[0007] Furthermore, in step S1, the dynamic equations of the state-space model of the permanent magnet synchronous servo motor are specifically as follows: (1) in, and These represent the stator current and stator voltage in the coordinate system, respectively. and These are the stator resistance and stator inductance, respectively. It is the extreme logarithm; and These are the moment of inertia and the coefficient of viscous friction, respectively. For permanent magnet flux linkage; This is the load torque; Represents spatial periodic cogging torque; Define state vector With scalar input The controlled object is rewritten as: (2) in, , , , , , , For system output; Since these disturbances, each with different physical properties, collectively affect the system, a unified framework is needed for subsequent dual-channel design. To this end, an equivalent input disturbance method is introduced, projecting all external disturbances and internal dynamics onto the control input channel, thus affecting... The combined effect is modeled as a lumped input signal Therefore, the above system can be equivalently rewritten as: (3) in, Indicates equivalent perturbation, Indicates cogging torque. Includes external loads and all unmodeled dynamics.

[0008] Furthermore, in step S2, the adaptive superspiral sliding mode observer uses the observation error... (4) As input, the output of the adaptive superspiral sliding mode observer is obtained as follows: (5) in, positive number The power is a fractional power; this parameter is specifically designed to ensure exponential convergence even with large initial errors. For symbolic functions, In the integration state, For adaptive superspiral gain; In traditional sliding mode observers, a fixed high gain is used to suppress severe high-frequency pulses of gear torque, which inevitably amplifies steady-state chatter. Conversely, a fixed low gain severely degrades dynamic tracking performance to cope with sudden non-periodic load steps. To fundamentally solve this dilemma and explicitly coordinate with the dual-channel cooperative mechanism, the adaptive superspiral gain is designed to be dynamically adaptive. (6) in, For error response coefficients, The attenuation coefficient is... For gain attenuation sensitivity coefficient, This represents the minimum robustness gain.

[0009] Furthermore, in step S3, the cogging torque is expressed using a spatial Fourier series as follows: (7) in, and The first Amplitude and phase of subspace harmonics, The rotor's mechanical angular position. It is the least common multiple of the number of stator slots and the number of rotor poles; The basic spatial period is: (8) The time-varying delay Determined by the following formula: (9) in, The rotor's mechanical angular velocity, The basic spatial period is the cogging torque; the historical moment is determined by tracing back one of these basic spatial periods from the current rotor mechanical angle position to the historical rotor mechanical angle position. .

[0010] Furthermore, in step S3, the rate of change of the time-varying delay satisfies: (10) ensure The validity of this condition is an absolute prerequisite for conducting causal and stability analyses. The above formula can be written as: To verify the causal boundary, the kinematic properties of the system are analyzed: During any acceleration phase, the integral is non-negative, and we can obtain... The causal condition is unconditionally satisfied during the deceleration phase; conversely, during the deceleration phase, constrained by the maximum torque and the system's moment of inertia, the physical deceleration process strictly satisfies the boundary conditions. ; Furthermore, since the time-varying repetitive controller operates only in the non-zero speed range... Applying the mean value theorem for definite integrals to the above equation, we can find that there exists a certain constant. Make Established; Substituting the physical boundary into the absolute value of the derivative expression, we can derive its supremum. for:

[0011] Therefore, as long as the motor meets the requirements This ensures the strict causality of the internal model of the variable structure.

[0012] Furthermore, in step S4, the time-varying repetitive controller includes a first-order low-pass filter and a time-varying delay internal model, designed as follows: (11) in, Represents a first-order low-pass filter ; The cutoff frequency of this filter must be selected to satisfy the following conditions: ; The highest frequency of the time-varying periodic reference input signal.

[0013] Furthermore, in step S4, the time-varying repetitive controller repeatedly controls the state. Achieve and satisfy: (12) in, This is the output of the time-varying periodic repetitive controller; The cogging torque compensation is: (13) in, It is a repetitive control gain.

[0014] Furthermore, in step S5, the residual disturbance integral compensation channel integrates the comprehensive disturbance estimate and multiplies it by a gain to generate the residual disturbance compensation amount corresponding to the load disturbance and the uncertainty of the slow-varying model. The comprehensive disturbance estimate and the residual disturbance compensation in the technical steps satisfy the following relationship: (14) in, This is the channel gain for compensating for remaining disturbances. It is the remaining disturbance compensation amount. It is a comprehensive disturbance estimate; The total disturbance compensation amount satisfies: (15) in, It is the total disturbance compensation amount. It is the cogging torque compensation amount.

[0015] Furthermore, in step S6, the control input is generated from the basic control input and the total disturbance compensation input, and the system control input satisfies: (16) in, , It is integral gain; It is the feedback gain matrix.

[0016] The present invention also provides a cogging torque suppression device for a variable speed servo motor based on dual-channel equivalent input disturbance, which has a computer program that can execute a method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance.

[0017] This invention relates to a method and apparatus for suppressing cogging torque in a variable-speed servo motor based on dual-channel equivalent input disturbance, which has the following beneficial effects: 1. The basic spatial period of the cogging torque is converted into an instantaneous time period by integrating the mechanical angular velocity, and the variable time delay repetitive controller is updated accordingly, so that the periodic disturbance compensation channel can maintain frequency matching with the cogging torque under acceleration, deceleration and continuous speed change conditions.

[0018] 2. By processing the comprehensive disturbance estimate in parallel through the periodic disturbance compensation channel and the residual disturbance compensation channel, the high-frequency periodic cogging torque is compensated by the variable time delay repetitive controller, and the load and slow-varying uncertainty are compensated by the integral channel, thus structurally reducing the burden of a single compensation path handling disturbances of different frequency bands at the same time.

[0019] 3. By using an adaptive sliding mode gain associated with the output observation error, the observer increases its gain during large error transients and falls back to near the minimum robust gain during small error steady states, thus taking into account comprehensive disturbance tracking, noise suppression, and reduction of steady-state chattering. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating Embodiment 1 of the method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance of the present invention. Figure 2 This is a schematic diagram of the control architecture of Embodiment 2 of the variable speed servo motor cogging torque suppression method based on dual-channel equivalent input disturbance of the present invention, showing the connection relationship between the adaptive super-torsion sliding mode observer, the variable time delay repetitive control channel, the residual disturbance compensation channel and the servo motor closed loop; Figure 3 This is a schematic diagram of the architecture of the variable speed servo motor cogging torque suppression device based on dual-channel equivalent input disturbance in Embodiment 3 of the present invention; Figure 4 The above is a simulation diagram of two reference inputs (trapezoidal reference input and sinusoidal reference input) in Embodiment 4 of the method and device for suppressing cogging torque of variable speed servo motor based on dual-channel equivalent input disturbance of the present invention. Figure 5 This is a simulation diagram of the comprehensive disturbance observation results under input 1 in Embodiment 4 of the present invention, which is based on the method and device for suppressing cogging torque of variable speed servo motors with dual-channel equivalent input disturbance. Figure 6 This is the dual-channel compensation output of Input 1 in Embodiment 4 of the present invention, which describes the method and device for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance. Figure 7 This is a simulation diagram of the speed tracking error at output 1 in Embodiment 4 of the present invention, which is based on the method and device for suppressing cogging torque of a variable speed servo motor using dual-channel equivalent input disturbance. Figure 8 This is a simulation diagram of the comprehensive disturbance observation results under input 2 in Embodiment 4 of the present invention, which is based on the method and device for suppressing cogging torque of variable speed servo motor based on dual-channel equivalent input disturbance. Figure 9 This is the dual-channel compensation output of input 2 in Embodiment 4 of the present invention, which is based on the dual-channel equivalent input disturbance suppression method and device for variable speed servo motor cogging torque. Figure 10 This is a simulation diagram of the speed tracking error under input 2 in Embodiment 4 of the present invention, which is a method and device for suppressing cogging torque of variable speed servo motor based on dual-channel equivalent input disturbance. Figure 11 This is a simulation diagram of the combined disturbance observation and speed tracking error under measurement noise conditions in Embodiment 4 of the present invention, which is based on the method and device for suppressing cogging torque of variable speed servo motors with dual-channel equivalent input disturbance. Figure 12 This is a simulation diagram of the speed tracking error under motor parameter perturbation conditions in Embodiment 4 of the present invention, which is based on the method and device for suppressing cogging torque of variable speed servo motors with dual-channel equivalent input perturbation. Detailed Implementation

[0021] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments, but these embodiments should not be construed as limiting the present invention.

[0022] This invention provides a cogging torque suppression technique for variable-speed servo motors based on dual-channel equivalent input disturbance. The technique first establishes a state model of the permanent magnet synchronous servo motor and projects the cogging torque, load torque, friction, and parameter perturbations onto the motor control input channel. Using an adaptive super-torsion sliding mode observer, the sliding mode gain is adjusted online based on the output observation error to obtain a comprehensive disturbance estimate. When the observation error is large, the sliding mode gain is increased; when the observation error decreases, the sliding mode gain is allowed to fall back to near the minimum robust gain. A variable time delay function is calculated based on the instantaneous duration required for the motor rotor to cross one fundamental spatial period of the cogging torque. The comprehensive disturbance estimate is fed into a periodic disturbance compensation channel containing this variable time delay function and a residual disturbance compensation channel containing an integral element. The outputs of the two channels are summed and negative feedback compensation is performed at the motor input.

[0023] Example 1 See Figure 1 This embodiment proposes a method for suppressing cogging torque in a variable-speed servo motor based on dual-channel equivalent input disturbance. The technical solution is as follows: S1. Establish a state-space model of the permanent magnet synchronous servo motor, and project the combined effects of cogging torque, load torque, friction and model parameter perturbation on the motor output onto the control input channel to form an equivalent input disturbance. S2. Construct an observation error based on the output measurement value and state observation value of the state space model, input the observation error into an adaptive superhelical sliding mode observer, and output the total equivalent disturbance signal by adaptively adjusting the superhelical gain; wherein, the superhelical gain increases with the increase of the observation error, and has a preset minimum robust gain lower limit, and decays towards the minimum robust gain lower limit after the observation error converges; S3. Determine the basic spatial period of cogging torque based on the number of stator slots and rotor poles of the permanent magnet synchronous motor, and calculate the time-varying delay required for the rotor to cross one basic spatial period based on the basic spatial period and the rotor mechanical angular velocity, and then construct a time-varying delay operator. S4. Input the total equivalent disturbance signal into the time-varying repetitive controller, and the time-varying repetitive controller uses the time-varying delay operator and low-pass filter to generate a periodic disturbance compensation amount that matches the spatial period of the cogging torque. S5. Based on the remaining disturbance components in the total equivalent input disturbance that were not compensated by the periodic disturbance compensation amount, generate the remaining disturbance compensation amount through the remaining disturbance integral compensation channel, and add the periodic disturbance compensation amount to the remaining disturbance compensation amount to obtain the total disturbance compensation amount. S6. The total disturbance compensation quantity is superimposed with the basic control quantity to generate the system control input quantity. The basic control quantity is jointly generated by the state feedback loop and the speed error integral loop. The state feedback loop generates the state feedback control quantity based on the deviation between the state reference vector and the state observation value through the feedback gain matrix. The speed error integral loop generates the speed error integral quantity based on the deviation between the rotor mechanical angular velocity reference value and the actual measured value through the integral gain. The final control input quantity is obtained by adding the state feedback control quantity, the speed error integral quantity and the total disturbance compensation quantity.

[0024] Example 2 This embodiment is basically the same as Embodiment 1, except that: This embodiment is a further detailed explanation of the method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance in Embodiment 1, such as... Figure 2 As shown: In step S1, the dynamic equations of the state-space model of the permanent magnet synchronous servo motor are specifically as follows: (1) in, and These represent the stator current and stator voltage in the coordinate system, respectively. and These are the stator resistance and stator inductance, respectively. It is the extreme logarithm; and These are the moment of inertia and the coefficient of viscous friction, respectively. For permanent magnet flux linkage; This is the load torque; This represents the spatial periodic cogging torque.

[0025] Define state vector With scalar input The controlled object is rewritten as: (2) in, , , , , , , This is the system output.

[0026] Since these disturbances, each with different physical properties, collectively affect the system, a unified framework is needed for subsequent dual-channel design. To this end, an equivalent input disturbance method is introduced. All external disturbances and internal dynamics are projected onto the control input channel, and their effects are considered... The combined effect is modeled as a lumped input signal Therefore, the above system can be equivalently rewritten as: (3) in, This indicates an equivalent perturbation. Indicates cogging torque. Includes external loads and all unmodeled dynamics.

[0027] In step S2, the adaptive superspiral sliding mode observer uses the observation error (4) As input, the output of the adaptive superspiral sliding mode observer is: (5) in, positive number The power is a fractional power; this parameter is specifically designed to ensure exponential convergence even with large initial errors. For symbolic functions, In the integration state, For adaptive superspiral gain.

[0028] In traditional sliding-mode observers, a fixed high gain is used to suppress severe high-frequency pulses of gear torque, which inevitably amplifies steady-state chatter. Conversely, a fixed low gain severely degrades dynamic tracking performance to handle sudden, aperiodic load steps. To fundamentally address this dilemma and explicitly coordinate with the dual-channel cooperative mechanism, the superspiral gain is designed to be dynamically adaptive: (6) in, For error response coefficients, The attenuation coefficient is... For gain attenuation sensitivity coefficient, This represents the minimum robustness gain.

[0029] In step S3, the cogging torque is expressed as a spatial Fourier series as follows: (7) in, and The first Amplitude and phase of subspace harmonics, The rotor's mechanical angular position. The least common multiple of the number of stator slots and the number of rotor poles; the basic spatial period is: (8) The time-varying delay Determined by the following formula: (9) in, The rotor's mechanical angular velocity, The basic spatial period is the cogging torque; the historical moment is determined by tracing back one of these basic spatial periods from the current rotor mechanical angle position to the historical rotor mechanical angle position. .

[0030] The rate of change of the time-varying delay satisfies: (10) ensure The validity of this condition is an absolute prerequisite for conducting causal and stability analyses. The above formula can be written as: To verify the causal boundary, the kinematic properties of the system are analyzed. During any acceleration phase, the integral is non-negative, and we can obtain... The causal condition is unconditionally satisfied. Conversely, during the deceleration phase, constrained by the maximum torque and the system's moment of inertia, the physical deceleration process strictly satisfies the boundary conditions. .

[0031] Furthermore, since the time-varying repetitive controller operates only in the non-zero speed range... Applying the mean value theorem for definite integrals to the above equation, we can find that there exists a certain constant. Make Established.

[0032] Substituting the physical boundary into the absolute value of the derivative expression, we can derive its supremum. for:

[0033] Therefore, as long as the motor meets the requirements This ensures the strict causality of the internal model of the variable structure.

[0034] In step S4, the time-varying repetitive controller includes a first-order low-pass filter and a time-varying delay internal model, which is designed as follows: (11) in, Represents a first-order low-pass filter ; The cutoff frequency of this filter must be selected to satisfy the following conditions: ; The highest frequency of the time-varying periodic reference input signal.

[0035] The time-varying repeating controller controls the repeating state. Achieve and satisfy: (12) in, This is the output of the time-varying periodic repetitive controller.

[0036] The cogging torque compensation is: (13) in, It is a repetitive control gain.

[0037] In step S5, the residual disturbance integral compensation channel integrates the comprehensive disturbance estimate and multiplies it by a gain to generate the residual disturbance compensation amount corresponding to the load disturbance and the uncertainty of the slow-varying model.

[0038] The comprehensive disturbance estimate and the residual disturbance compensation in the technical steps satisfy the following relationship: (14) in, This is the channel gain for compensating for remaining disturbances. It is the remaining disturbance compensation amount. It is a comprehensive disturbance estimate.

[0039] The total disturbance compensation amount satisfies: (15) in, It is the total disturbance compensation amount. It is the cogging torque compensation amount.

[0040] In step S6, the control input is generated from the basic control input and the total disturbance compensation input, and the system control input satisfies: (16) in, , It is integral gain; It is the feedback gain matrix.

[0041] Example 3 See Figure 3 This embodiment of the variable speed servo motor cogging torque suppression device based on dual-channel equivalent input disturbance includes the following: Input disturbance module: Establish a state-space model of the permanent magnet synchronous servo motor, and project the combined effects of cogging torque, load torque, friction and model parameter perturbation on the motor output onto the control input channel to form an equivalent input disturbance; Adaptive superhelical sliding mode observation module: Constructs observation error based on the output measurement value and state observation value of the state space model, inputs the observation error into the adaptive superhelical sliding mode observer, and outputs the total equivalent disturbance signal by adaptively adjusting the superhelical gain; wherein, the superhelical gain increases with the increase of the observation error, has a preset minimum robust gain lower limit, and decays towards the minimum robust gain lower limit after the observation error converges; Time-varying delay operator construction module: Determine the basic spatial period of cogging torque based on the number of stator slots and rotor poles of the permanent magnet synchronous motor, and calculate the time-varying delay required for the rotor to cross one basic spatial period based on the basic spatial period and the rotor mechanical angular velocity, and then construct the time-varying delay operator. Time-varying repetitive control module: The total equivalent disturbance signal is input to the time-varying repetitive controller, which uses the time-varying delay operator and low-pass filter to generate a periodic disturbance compensation amount that matches the spatial period of the cogging torque; Total disturbance compensation module: Based on the remaining disturbance components in the total equivalent input disturbance that are not compensated by the periodic disturbance compensation amount, the module generates a remaining disturbance compensation amount through the remaining disturbance integral compensation channel, and adds the periodic disturbance compensation amount to the remaining disturbance compensation amount to obtain the total disturbance compensation amount; Control input generation module: The total disturbance compensation quantity is superimposed with the basic control quantity to generate the control input quantity. The basic control quantity is jointly generated by the state feedback loop and the speed error integration loop. The state feedback loop generates the state feedback control quantity based on the deviation between the state reference vector and the state observation value through the feedback gain matrix. The speed error integration loop generates the speed error integral quantity based on the deviation between the rotor mechanical angular velocity reference value and the actual measured value through the integration gain. The final control input quantity is obtained by adding the state feedback control quantity, the speed error integral quantity and the total disturbance compensation quantity.

[0042] Example 4 This embodiment provides a detailed description of the technical effects of embodiments 1-3 using specific examples: A simulation platform was built to verify the effectiveness of the invention, and the relevant electrical parameters were selected as follows: motor moment of inertia. Viscous damping coefficient .

[0043] definition , , , get (17) Two representative reference inputs were considered: a trapezoidal velocity reference input with acceleration, constant velocity, and deceleration phases, and a sinusoidal reference input. Figure 4 Two reference inputs are displayed.

[0044] Cogging torque for Load torque for .

[0045] Figure 5 and Figure 6 as well as Figure 8 and Figure 9These are simulation diagrams of various embodiments of the present invention under different input conditions. It can be seen that the cogging torque can be accurately and stably identified under different working conditions. This invention is also compared with the disturbance suppression method for repetitive control systems based on phase-compensated equivalent input disturbances and the disturbance suppression method for nonlinear repetitive control systems based on high-order sliding mode equivalent input disturbances (M. Zhang, C. Lu, S. Tian, ​​Y. Wang, M. Wu and M. Iwasaki, "Improved disturbance rejection in repetitive-control systems: phase-compensated equivalent-input-disturbance approach," IEEE Transactions on Industrial Electronics, 71 (10), 12942-12951, 2024; S. Tian, ​​K.-Z. Liu, M. Zhang C. Lu, M. Wu and J. She, "Improved equivalent-input-disturbance method for nonlinear repetitive-control systems: integration with high-ordersliding-mode control," IEEE / ASME Transactions on Mechatronics, 30 (1), 692-702, 2025.). Figure 7 and Figure 10 As shown, it is easy to see that the present invention exhibits a more significant advantage in terms of disturbance suppression, and the addition of an adaptive high-order sliding mode observer gives the proposed structure a great advantage in dynamic response. Furthermore, compared with existing active suppression methods, the present invention greatly improves the system's stability and anti-interference capability.

[0046] Figure 11 , Figure 12 The robustness of the invention was verified. When band-limited white noise with a standard deviation of 0.1 was added to the output, and the moment of inertia was adjusted to 1.2 times the nominal value and the viscous damping was adjusted to 0.8 times the nominal value, the invention still maintained a small speed tracking error.

[0047] The key technical point of this invention is that it achieves dual-channel compensation through a time-varying repetitive controller and an adaptive high-order sliding mode observer, and realizes online updating of the observer gain, which has the advantages of strong anti-disturbance capability and fast speed response.

[0048] This invention discloses a cogging torque suppression scheme for a variable-speed servo motor based on dual-channel equivalent input disturbances. The scheme includes: establishing a state model of a permanent magnet synchronous motor and projecting the cogging torque, load, and model uncertainty as equivalent input disturbances; obtaining a comprehensive disturbance estimate using an adaptive super-torsional sliding mode observer that adjusts the sliding mode gain online based on the output observation error; updating the variable time delay based on the instantaneous duration of the rotor crossing the basic spatial period of the cogging torque; inputting the comprehensive disturbance estimate in parallel into the variable time delay repetitive control channel and the integral compensation channel to generate cogging torque compensation and residual disturbance compensation, respectively; superimposing the two compensations and combining them with the state feedback control quantity to obtain the control command. This scheme enables the motor to maintain periodic disturbance frequency matching under variable-speed conditions while also achieving excellent disturbance suppression capabilities.

[0049] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0050] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A method for suppressing cogging torque of a variable-speed servo motor based on dual-channel equivalent input disturbance, applied to a permanent magnet synchronous servo motor control system, characterized in that: Includes the following steps: S1. Establish a state-space model of the permanent magnet synchronous servo motor, and project the combined effects of cogging torque, load torque, friction and model parameter perturbation on the motor output onto the control input channel to form an equivalent input disturbance. S2. Construct an observation error based on the output measurement value and state observation value of the state space model, input the observation error into an adaptive superhelical sliding mode observer, and output the total equivalent disturbance signal by adaptively adjusting the superhelical gain; wherein, the superhelical gain increases with the increase of the observation error, and has a preset minimum robust gain lower limit, and decays towards the minimum robust gain lower limit after the observation error converges; S3. Determine the basic spatial period of cogging torque based on the number of stator slots and rotor poles of the permanent magnet synchronous motor, and calculate the time-varying delay required for the rotor to cross one basic spatial period based on the basic spatial period and the rotor mechanical angular velocity, and then construct a time-varying delay operator. S4. Input the total equivalent disturbance signal into the time-varying repetitive controller, and the time-varying repetitive controller uses the time-varying delay operator and low-pass filter to generate a periodic disturbance compensation amount that matches the spatial period of the cogging torque. S5. Based on the remaining disturbance components in the total equivalent input disturbance that are not compensated by the periodic disturbance compensation amount, generate the remaining disturbance compensation amount through the remaining disturbance integral compensation channel, and add the periodic disturbance compensation amount to the remaining disturbance compensation amount to obtain the total disturbance compensation amount. S6. The total disturbance compensation quantity is superimposed with the basic control quantity to generate the control input quantity. The basic control quantity is jointly generated by the state feedback loop and the speed error integral loop. The state feedback loop generates the state feedback control quantity based on the deviation between the state reference vector and the state observation value through the feedback gain matrix. The speed error integral loop generates the speed error integral quantity based on the deviation between the rotor mechanical angular velocity reference value and the actual measured value through the integral gain. The final control input quantity is obtained by adding the state feedback control quantity, the speed error integral quantity and the total disturbance compensation quantity.

2. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 1, characterized in that: In step S1, the dynamic equations of the state-space model of the permanent magnet synchronous servo motor are specifically as follows: （1） in, and These represent the stator current and stator voltage in the coordinate system, respectively. and These are the stator resistance and stator inductance, respectively. It is the extreme logarithm; and These are the moment of inertia and the coefficient of viscous friction, respectively. For permanent magnet flux linkage; This is the load torque; Represents spatial periodic cogging torque; Define state vector With scalar input The controlled object is rewritten as: （2） in, , , , , , , For system output; The equivalent input disturbance method is introduced, which projects all external disturbances and internal dynamics onto the control input channel, and then applies the equivalent input disturbance to the system output. The combined effect is modeled as a lumped input signal Therefore, the controlled object described herein is equivalently rewritten as: （3） in, Indicates equivalent perturbation, Indicates cogging torque. Includes external loads and all unmodeled dynamics.

3. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 2, characterized in that: In step S2, the adaptive superspiral sliding mode observer uses the observation error (4) As input, the output of the adaptive superspiral sliding mode observer is obtained as follows: （5） in, positive number The power is a fractional power; this parameter is specifically designed to ensure exponential convergence even with large initial errors. For symbolic functions, In the integration state, For adaptive superspiral gain; The dynamic adaptation of the adaptive superspiral gain is as follows: （6） in, For error response coefficients, The attenuation coefficient is... For gain attenuation sensitivity coefficient, This represents the minimum robustness gain.

4. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 3, characterized in that: In step S3, the cogging torque is represented using a spatial Fourier series method: （7） in, and The first Amplitude and phase of subspace harmonics, The rotor's mechanical angular position. It is the least common multiple of the number of stator slots and the number of rotor poles; The basic spatial period is: （8） The time-varying delay Determined by the following formula: （9） in, The rotor's mechanical angular velocity, The basic spatial period is the cogging torque; the historical moment is determined by tracing back one of these basic spatial periods from the current rotor mechanical angle position to the historical rotor mechanical angle position. .

5. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance according to claim 4, characterized in that: In step S3, the rate of change of the time-varying delay satisfies: （10） because The establishment of this condition is an absolute prerequisite for conducting causal and stability analyses; the above formula can be written as follows: ; Analyzing the kinematic properties of the rate of change of the time-varying delay to verify its causal boundary: In any acceleration phase, the integral is non-negative, thus yielding... The causal condition is unconditionally satisfied; Conversely, during the deceleration phase, constrained by the maximum torque and the system's moment of inertia, the physical deceleration process strictly satisfies the boundary conditions. ; Because the time-varying repetitive controller only operates in the non-zero speed range By applying the mean value theorem of definite integrals to the formula for the rate of change of time-varying delay, a certain constant can be obtained. Make Established; Substituting the physical boundary into the absolute value of the derivative expression, we can derive its supremum. for: It can be seen that as long as the motor meets the requirements... This ensures the strict causality of the internal model of the variable structure.

6. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 5, characterized in that: In step S4, the time-varying repetitive controller includes a first-order low-pass filter and a time-varying delay internal model, which is designed as follows: （11） in, Represents a first-order low-pass filter ; The cutoff frequency of this filter must be selected to satisfy the following conditions: ; The highest frequency of the time-varying periodic reference input signal.

7. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 6, characterized in that: In step S4, the time-varying repetitive controller controls the repetitive state. Achieve and satisfy: （12） in, This is the output of the time-varying periodic repetitive controller; The cogging torque compensation is: （13） in, It is a repetitive control gain.

8. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 7, characterized in that: In step S5, the residual disturbance integral compensation channel integrates the comprehensive disturbance estimate and multiplies it by a gain to generate the residual disturbance compensation amount corresponding to the load disturbance and the uncertainty of the slow-varying model. The integrated disturbance estimate and the residual disturbance compensation satisfy the following relationship: （14） in, The remaining disturbance integral is used to compensate for the channel gain. It is the remaining disturbance compensation amount. It is a comprehensive disturbance estimate; The total disturbance compensation amount satisfies: （15）， in, It is the total disturbance compensation amount. It is the cogging torque compensation amount.

9. The method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in claim 8, characterized in that: In step S6, the control input is generated from the basic control quantity and the total disturbance compensation quantity, and the control input satisfies: （16） in, , It is integral gain; It is the feedback gain matrix.

10. A cogging torque suppression device for a variable speed servo motor based on dual-channel equivalent input disturbance, comprising a computer program, characterized in that: The computer program is capable of executing the method for suppressing cogging torque of a variable speed servo motor based on dual-channel equivalent input disturbance as described in any one of claims 1 to 9.

Images