Digital servo system mechanical resonance suppression method based on frequency offset automatic correction

By improving the mechanical resonance model in the digital servo system and using a notch instead of a low-pass filter, and using a single-parameter frequency self-tracking algorithm to self-tune the notch, the mechanical resonance frequency offset problem in a high-bandwidth digital servo system is solved, and the mechanical resonance frequency offset problem is realized. The oscillation frequency correction and dynamic suppression effects are achieved.

CN119093794AActive Publication Date: 2024-12-06WUHAN UNIV OF SCI & TECH +2

Patent Information

Application Number
CN202411186659.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-12-06
Estimated Expiration
2044-08-28

AI Technical Summary

Technical Problem

In high-bandwidth digital servo systems, the mechanical resonance frequency does not match the true value and causes the notch filter to fail, and the problem of oscillation frequency deviation correction needs to be solved urgently.

Method used

Based on automatic correction of frequency deviation, the mechanical resonance suppression method of digital servo system is improved. By improving the mechanical resonance model of digital servo system, a notch is used instead of a low-pass filter, a single-parameter frequency self-tracking algorithm is used to detect the frequency deviation signal, and a self-tuning notch is used. Dynamically suppress mechanical resonance.

Benefits of technology

The oscillation frequency deviation correction and dynamic mechanical resonance suppression are achieved, reducing the number of oscillation frequency detection times and suppression time, and improving the suppression efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093794A_ABST
    Figure CN119093794A_ABST
Patent Text Reader

Abstract

The invention provides a digital servo system mechanical resonance suppression method based on automatic frequency offset correction, which is improved from two aspects of a correction strategy and a frequency detection method under the condition of vibration frequency offset so as to better suppress mechanical resonance. Firstly, a perfect digital system mechanical resonance model is constructed, according to the highly similar phase characteristics of a notch filter and a second-order low-pass filter, a low-pass filter added in a traditional scheme is replaced by a notch filter, and the problems that a traditional suppression strategy is complex and deviation correction is slow are solved; secondly, designing a single-parameter frequency self-tracking algorithm for frequency offset signal detection, ensuring that the oscillation frequency is rapidly detected without considering initial conditions, damping ratio and amplitude, reducing the number of times of required oscillation frequency detection and shortening the time consumption of resonance suppression, and improving the suppression speed from the aspect of the detection method so as to adjust a notch filter on line; the functions of oscillation frequency correction and dynamic mechanical resonance suppression are realized, and the requirements of a modern high-bandwidth digital servo system are better met.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of AC servo control systems, and in particular relates to a mechanical resonance suppression method for a digital servo system based on automatic frequency deviation correction. Background Art

[0002] Today, permanent magnet synchronous motors have important applications in many fields due to their superior performance characteristics, especially in scenarios with high requirements for efficiency, power density and precision control. However, in modern permanent magnet synchronous motor servo systems, a transmission mechanism is usually used to connect the motor and the load, and this connection method often leads to mechanical resonance. Existing solutions include active and passive methods: the active method is to eliminate the influence of resonance by changing the structure of the controller or optimizing the parameters, but this method is limited by factors such as complex parameter adjustment and complex algorithms; the passive method is to insert a notch filter in series in the forward channel of the control loop, which is widely used and promoted due to its strong targeting and simple parameter setting.

[0003] Compared with the notch filter, the adaptive notch filter can automatically adjust the notch frequency to the resonant frequency to suppress the mechanical resonance of the servo system, but the effectiveness of the adaptive notch filter depends on the accuracy of frequency detection. In high-bandwidth digital servo systems, due to factors such as delay, digital control, and speed controller output saturation, the system oscillation frequency has the problem of deviating from the resonant frequency to a higher frequency, making the frequency of the notch filter mismatched with the true value, resulting in the failure of the notch filter and even causing the system to produce stronger oscillations after the notch filter is added. Therefore, the problem of oscillation frequency correction is urgently needed to be overcome in modern high-bandwidth digital servo systems. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a digital servo system mechanical resonance suppression method based on automatic frequency deviation correction, which is used for oscillation frequency deviation correction and dynamic suppression of mechanical resonance.

[0005] The technical solution adopted by the present invention to solve the above technical problems is: a method for suppressing mechanical resonance of a digital servo system based on automatic correction of frequency deviation, comprising the following steps:

[0006] S1: Improve the mechanical resonance model of the digital servo system based on the classic dual-inertia servo system model and analyze the oscillation frequency offset mechanism;

[0007] S2: A notch filter is used instead of a low-pass filter; the frequency deviation signal is detected by a single-parameter frequency self-tracking algorithm, which is used to quickly and accurately detect the oscillation frequency without considering the initial condition damping ratio and amplitude, and self-tune the notch filter to achieve dynamic mechanical resonance suppression.

[0008] According to the above scheme, the following steps are also included:

[0009] S0: Let ω m ,ω l , J m , J l , T e , T l , T w , K s 、b w ,θ m and θ l They represent motor speed, load speed, motor moment of inertia, load moment of inertia, electromagnetic torque, load torque, wheelbase, stiffness coefficient, connecting shaft damping coefficient, motor position and load position respectively; the classic dual-inertia servo system model is established as:

[0010]

[0011] According to formula (1), the transfer function from electromagnetic torque and load torque to motor speed is written as shown in formula (2), where the first term of the transfer function is the same as that of the single inertia system, which is an ideal rigid load, and the second term is a separated resonance model:

[0012]

[0013] Get the system resonant frequency ω NFT and the anti-resonance frequency ω ARF The expression is:

[0014]

[0015] Furthermore, in the step S1, the specific steps are:

[0016] Let A be the amplitude limit, X be the width of the linear region, and use the describing function Analyzing the saturated nonlinear link, the saturated nonlinear negative inverse function crosses from the unstable region to the stable region. The system generates self-sustaining oscillation at the intersection of the curves, i.e., point α. Point α corresponds to the open-loop phase crossing angular frequency, as shown in formula (4):

[0017]

[0018] Let τ be the time constant of the hysteresis loop, T asr is the speed loop sampling period; write the open-loop transfer function G of the digital control system based on the mechanical resonance model of the digital servo system L (s) is:

[0019]

[0020] According to the above scheme, in step S2, the specific steps of automatically correcting the frequency deviation are:

[0021] S211: In the first processing cycle, detect whether the system oscillates, detect the oscillation frequency through the single-parameter frequency self-tracking algorithm for frequency deviation signal detection, automatically tune the notch filter, observe whether the resonance is suppressed, and determine the oscillation frequency at this time:

[0022] If the resonance is suppressed, the frequency deviation correction is terminated;

[0023] If the resonance is not suppressed and a larger oscillation is generated, the oscillation frequency is determined to be offset and the oscillation frequency is determined to be ω1 at this time; the notch filter generates a phase so that the oscillation frequency is moved forward from ω1 to ω2;

[0024] S212: Using the oscillation frequency ω2 as the update frequency of the notch filter;

[0025] S213: Repeat step S212 until the detection results of two consecutive rounds of oscillation frequencies are equal, that is, the oscillation frequency has reached the resonance frequency, and tune the notch filter according to the last detected oscillation frequency to determine whether the resonance is suppressed.

[0026] According to the above scheme, in step S2, the specific steps of the single parameter frequency self-tracking algorithm and the optimization of the notch filter are:

[0027] Assume ω0 is the unknown angular frequency, η(t) is white noise, and k is the amplitude; the frequency estimator input signal u(t) is in the form of:

[0028] u(t)=ksin(ω0t)+η(t)(6),

[0029] Assume x1 and x2 are the filter states, y is the output, and θ is the estimated frequency. The optimized state space model expression is as follows:

[0030]

[0031] Assume that γ>0 is the adaptive gain. By adaptively adjusting the error signal in real time, when the input-output error is zero e=uy→0, the dynamic state vector [x1,x2] T The steady-state oscillation is performed at a frequency of θ = ω0; the adaptive law is:

[0032]

[0033] Furthermore, in step S2, when t→∞, the state space models (9) and (10) are global to (6); and as long as θ(t)→ω0, regardless of the initialization of θ(0)>0, when the gain γ>0, the condition for the frequency estimation error δ(t)=ω0-θ(t) to converge uniformly in exponential form is:

[0034]

[0035] where ν>0 and ρ is a constant and has nothing to do with γ, k, and ω0.

[0036] Furthermore, in step S2, the specific steps of designing a single parameter to ensure frequency convergence are:

[0037] The system matrix and the input and output coupling vectors in the state-space models (9) and (10) are represented by A, B and C respectively, I is the 2×2 identity matrix, and is the ω angular frequency variable; the state-space models (9) and (10) are described by the input-to-output transfer function G(jω) as follows:

[0038]

[0039] When the output error is not zero, the expression of the error e(jω) is written as follows according to equation (13):

[0040] e(jω)=y(jω)-u(jω)=(1-G)G -1 y(jω)(14),

[0041] By enforcing the dynamics (9) and (10), |e(jω)|>0 for all ω except ω=ω0; this motivates the adaptive law (11), ensuring The exception is always at θ = ω0; the above transfer function from the filter output to the error is expressed as:

[0042]

[0043] Presenting the above equation in the form of a Bode plot, the error transfer function amplitude |E| has a global minimum at ω=θ, so when θ(t)→ω0, |e(jω)|→0, and is independent of the initial value θ(0) set by the estimator.

[0044] A dual-inertia transmission device, the system comprises a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, when the computer instructions are executed by the processor, the system implements the steps of a digital servo system mechanical resonance suppression method based on automatic frequency deviation correction.

[0045] Furthermore, based on the classic dual-inertia servo system model, the mechanical resonance model of the digital servo system is improved, the current loop is equivalent to a hysteresis loop, PI is used as a proportional-integral controller for the speed loop control of the servo system, and a zero-order holder is added to simulate the limiting link in the saturated nonlinear link of digital control.

[0046] A computer memory stores a computer program that can be executed by a computer processor. The computer program executes a mechanical resonance suppression method of a digital servo system based on automatic frequency deviation correction.

[0047] The beneficial effects of the present invention are:

[0048] 1. The digital servo system mechanical resonance suppression method based on automatic frequency deviation correction of the present invention makes improvements in the correction strategy and the frequency detection method in order to better suppress mechanical resonance in the case of vibration frequency deviation. By constructing a complete digital system mechanical resonance model, the problems of complex traditional suppression strategies and slow correction are solved, the number of required oscillation frequency detection times is reduced and the time consumption of resonance suppression is shortened, thus realizing the functions of oscillation frequency correction and dynamic mechanical resonance suppression.

[0049] 2. Based on the vibration frequency offset mechanism of the digital servo system, the present invention analyzes the problems of long suppression time and failure of the traditional adaptive notch filter suppression strategy. According to the high similarity between the notch filter and the second-order low-pass filter in phase characteristics, the low-pass filter added in the traditional scheme is replaced by a notch filter to solve the shortcomings of the traditional strategy that is complex and has slow correction.

[0050] 3. The present invention designs a single-parameter frequency self-tracking algorithm for frequency deviation signal detection, which ensures that the oscillation frequency is quickly detected without considering the initial conditions, damping ratio and amplitude, and improves the suppression speed from the detection method aspect to adjust the notch filter online. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a model block diagram of a dual-inertia transmission device according to an embodiment of the present invention.

[0052] Figure 2 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0054] Figure 1 It is a model block diagram of a dual-inertia transmission device for realizing a method for suppressing mechanical resonance of a digital servo system based on automatic correction of frequency deviation provided by an embodiment of the present invention, which includes: improving the digital mechanical resonance model based on the classical model, that is, considering factors such as current loop delay, digital control and output saturation of the speed controller. The current loop is equivalent to a hysteresis loop rather than a torque coefficient, PI is used as a proportional integral controller for the speed loop control of the servo system, the addition of a zero-order holder simulates the influence of digital control, and the limiting link in the saturated nonlinear link avoids the occurrence of more serious oscillations.

[0055] The shortcomings of traditional correction schemes are improved from two aspects: correction strategy and frequency detection method, which reduces the number of required oscillation frequency detection and shortens the time of resonance suppression, so as to achieve dynamic mechanical resonance suppression. Figure 2 , an embodiment of the present invention comprises the following steps:

[0056] S0: Establish a classic dual-inertia servo system model; the specific steps are:

[0057] Based on the classic dual-inertia servo system model, its dynamic equation is written as follows:

[0058]

[0059] Among them, ω m ,ω l , J m , J l , T e , T l , T w , K s , b w ,θ m and θ l They represent motor speed, load speed, motor moment of inertia, load moment of inertia, electromagnetic torque, load torque, wheelbase, stiffness coefficient, connecting shaft damping coefficient, motor position and load position respectively;

[0060] According to formula (1), the transfer function from electromagnetic torque and load torque to motor speed is written as shown in formula (2), where the first term of the transfer function is the same as that of the single inertia system, which is an ideal rigid load, and the second term is a separated resonance model:

[0061]

[0062] Get the system resonant frequency ω NFT and the anti-resonance frequency ω ARF The expression is:

[0063]

[0064] S1: Improve the mechanical resonance model of the digital servo system based on the classic dual-inertia servo system model and analyze the oscillation frequency offset mechanism;

[0065] The perfect digital mechanical resonance model and oscillation frequency shift mechanism are as follows:

[0066] The digital mechanical resonance model established takes into account factors such as current loop delay, digital control and speed controller output saturation. The saturation nonlinear link is analyzed as shown in equation (4), where A is the amplitude limit and X is the width of the linear region. The saturation nonlinear negative inverse function crosses from the unstable region to the stable region, and the system will generate self-sustaining oscillation at the intersection of the curves (point α), which corresponds to the open-loop phase crossing angular frequency.

[0067]

[0068] According to the mechanical resonance model of the digital servo system, the open-loop transfer function G of the digital control system is written L (s) are as follows:

[0069]

[0070] Where τ is the hysteresis loop time constant, T asr is the speed loop sampling period; the zero-order holder is equivalent to Output. It can be seen that the phase of the linear system is determined by G1, G i and G s The oscillation frequency deviation is caused by the phase lag of the three, which makes the phase crossing frequency greater than the resonant frequency.

[0071] S2: Improve the traditional deviation correction scheme from two aspects: deviation correction strategy and frequency detection method. According to the high similarity between the notch filter and the second-order low-pass filter in phase characteristics, the low-pass filter added in the traditional scheme is replaced by a notch filter to overcome the problem that the traditional notch filter suppression strategy takes too long and has failure. Design a single-parameter frequency self-tracking algorithm for frequency deviation signal detection, which is used to quickly and accurately detect the oscillation frequency without considering the initial condition damping ratio and amplitude, and self-tune the notch filter to achieve dynamic mechanical resonance suppression.

[0072] The specific method of automatic frequency deviation correction is:

[0073] Since the phase characteristics of the notch filter and the second-order low-pass filter are highly similar, the low-pass filter added in the traditional solution is replaced by the notch filter to overcome the problem that the traditional notch filter suppression strategy has a long suppression time and is ineffective. The steps are as follows:

[0074] The first processing cycle detects whether the system is oscillating. The oscillation frequency is detected and the notch filter is automatically tuned through the single-parameter frequency self-tracking algorithm for frequency deviation signal detection to observe whether the resonance is suppressed and determine the oscillation frequency at this time. If it is suppressed, the process ends. If the resonance is not suppressed and a larger oscillation is generated, it can be seen that the oscillation frequency is offset and the oscillation frequency at this time is determined to be ω1. The phase generated by the notch filter can shift the oscillation frequency, that is, the crossover frequency, from ω1 to ω2.

[0075] The oscillation frequency ω2 is used as the update frequency of the notch filter;

[0076] Repeat the frequency update until the results of two consecutive rounds of oscillation frequency detection are basically equal, that is, the oscillation frequency has reached the resonance frequency. Then tune the notch filter according to the last detected oscillation frequency to determine whether the resonance is suppressed.

[0077] The design of the single-parameter frequency self-tracking algorithm is:

[0078] The frequency estimator input signal u(t) is in the form of

[0079] u(t)=ksin(ω0t)+η(t)(6)

[0080] Where ω0 is the unknown angular frequency, η(t) can be regarded as white noise, and k is the amplitude.

[0081] Let x be the filter state, ξ be the damping ratio, and θ be the estimated frequency; the expression of the traditional adaptive notch filter is

[0082]

[0083] Assume γ>0 as adaptive gain, the frequency estimator equation is

[0084]

[0085] Compared with the conventional adaptive notch filter, this embodiment adjusts the scale factor of the frequency estimator, uses the symbol sign instead of the x state in (7), and cancels the damping ratio ξ design parameter.

[0086] Assume x1, x2 are filter states, and y is output; the specific expression of algorithm optimization design is as follows:

[0087]

[0088] By adaptively adjusting the error signal in real time, when the input-output error is zero, that is, e=uy→0, the dynamic state vector [x1,x2] T The steady-state oscillation is performed at a frequency of θ = ω0; the adaptive law is:

[0089]

[0090] The design of frequency estimation error convergence is:

[0091] For t→∞, the frequency estimators (9) and (10) are global to (6) and as long as θ(t)→ω0, regardless of the initialization of θ(0)>0, as long as there is a suitable gain γ>0, the frequency estimation error δ(t)=ω0-θ(t) converges uniformly in an exponential form, provided that:

[0092]

[0093] where ν>0 and ρ is a constant and has nothing to do with γ, k, and ω0.

[0094] The specific method of designing a single parameter to ensure frequency convergence is:

[0095] The system matrix and the input and output coupling vectors in (9) and (10) are represented by A, B and C respectively, I is the 2×2 identity matrix, and is the ω angular frequency variable; the state space model of (9) and (10) is described by the input-to-output transfer function G(jω):

[0096]

[0097] From the above equation, it can be seen that the 2θ input coupling factor designed in (9) gives the input-output ratio of (9) and (10) which is independent of k in the steady state, making the frequency estimation amplitude insensitive, and no further proportional factor is assigned to the signal u(t). This simplification eliminates the design of the damping ratio parameter. When the output error is not zero, the above equation can be rewritten as

[0098] e(jω)=y(jω)-u(jω)=(1-G)G -1 y(jω) (14)

[0099] By enforcing the dynamics (9) and (10), |e(jω)|>0 for all ω except ω=ω0. This motivates the adaptive law (11), which ensures The exception is always at θ = ω0. The above transfer function, that is, from the filter output to the error, is expressed as:

[0100]

[0101] Presenting the above equation in the form of a Bode plot, the error transfer function amplitude |E| has a global minimum at ω=θ, so when θ(t)→ω0, |e(jω)|→0, and this is independent of the initial value θ(0) set by the estimator.

[0102] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0103] The above embodiments are only used to illustrate the design ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A method for suppressing mechanical resonance of a digital servo system based on automatic correction of frequency deviation, characterized in that: The following steps are involved: S1: Improve the mechanical resonance model of the digital servo system based on the classic dual-inertia servo system model and analyze the oscillation frequency offset mechanism; S2: Use notch filter instead of low-pass filter; The frequency deviation signal is detected by a single-parameter frequency self-tracking algorithm, which is used to quickly and accurately detect the oscillation frequency without considering the initial condition damping ratio and amplitude, and self-tune the notch filter to achieve dynamic mechanical resonance suppression.

2. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 1, characterized in that: The following steps are also included: S0: Let ω m ,ω l , J m , J l , T e , T l , T w , K s , b w ,θ m and θ l They represent motor speed, load speed, motor moment of inertia, load moment of inertia, electromagnetic torque, load torque, wheelbase, stiffness coefficient, connecting shaft damping coefficient, motor position and load position respectively; the classic dual-inertia servo system model is established as: According to formula (1), the transfer function from electromagnetic torque and load torque to motor speed is written as shown in formula (2), where the first term of the transfer function is the same as that of the single inertia system, which is an ideal rigid load, and the second term is a separated resonance model: Get the system resonant frequency ω NFT and the anti-resonance frequency ω ARF The expression is:

3. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 2, characterized in that: In the step S1, the specific steps are: Let A be the amplitude limit, X be the width of the linear region, and use the describing function Analyzing the saturated nonlinear link, the saturated nonlinear negative inverse function crosses from the unstable region to the stable region. The system generates self-sustaining oscillation at the intersection of the curves, i.e., point α. Point α corresponds to the open-loop phase crossing angular frequency, as shown in formula (4): Let τ be the time constant of the hysteresis loop, T asr is the speed loop sampling period; write the open-loop transfer function G of the digital control system based on the mechanical resonance model of the digital servo system L (s) is:

4. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 1, characterized in that: In step S2, the specific steps of automatically correcting the frequency deviation are: S211: In the first processing cycle, detect whether the system oscillates, detect the oscillation frequency through the single-parameter frequency self-tracking algorithm for frequency deviation signal detection, automatically tune the notch filter, observe whether the resonance is suppressed, and determine the oscillation frequency at this time: If the resonance is suppressed, the frequency deviation correction is terminated; If the resonance is not suppressed and a larger oscillation is generated, the oscillation frequency is determined to be offset and the oscillation frequency is determined to be ω1 at this time; the notch filter generates a phase so that the oscillation frequency is moved forward from ω1 to ω2; S212: Using the oscillation frequency ω2 as the update frequency of the notch filter; S213: Repeat step S212 until the detection results of two consecutive rounds of oscillation frequencies are equal, that is, the oscillation frequency has reached the resonance frequency, and tune the notch filter according to the last detected oscillation frequency to determine whether the resonance is suppressed.

5. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 1, characterized in that: In step S2, the specific steps of the single parameter frequency self-tracking algorithm and the notch filter optimization are: Assume ω0 is the unknown angular frequency, η(t) is white noise, and k is the amplitude; the frequency estimator input signal u(t) is in the form of: u(t)=ksin(ω0t)+η(t) (6), Assume x1 and x2 are the filter states, y is the output, and θ is the estimated frequency. The optimized state space model expression is as follows: Assume that γ>0 is the adaptive gain. By adaptively adjusting the error signal in real time, when the input-output error is zero e=uy→0, the dynamic state vector [x1,x2] T The steady-state oscillation is performed at a frequency of θ = ω0; the adaptive law is:

6. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 5, characterized in that: In step S2, when t→∞, the state space models (9) and (10) are global to (6); and as long as θ(t)→ω0, regardless of the initialization of θ(0)>0, when the gain γ>0, the condition for the frequency estimation error δ(t)=ω0-θ(t) to converge uniformly in exponential form is: where ν>0 and ρ is a constant and has nothing to do with γ, k, and ω0.

7. The method for suppressing mechanical resonance of a digital servo system based on automatic frequency deviation correction according to claim 5, characterized in that: In step S2, the specific steps of designing a single parameter to ensure frequency convergence are: The system matrix and the input and output coupling vectors in the state-space models (9) and (10) are represented by A, B and C respectively, I is the 2×2 identity matrix, and is the ω angular frequency variable; the state-space models (9) and (10) are described by the input-to-output transfer function G(jω) as follows: When the output error is not zero, the expression of the error e(jω) is written as follows according to equation (13): e(jω)=y(jω)-u(jω)=(1-G)G -1 y(jω) (14), By enforcing the dynamics (9) and (10), |e(jω)|>0 for all ω except ω=ω0; this motivates the adaptive law (11), ensuring The exception is always at θ = ω0; the above transfer function from the filter output to the error is expressed as: Presenting the above equation in the form of a Bode plot, the error transfer function amplitude |E| has a global minimum at ω=θ, so when θ(t)→ω0, |e(jω)|→0, and is independent of the initial value θ(0) set by the estimator.

8. A dual-inertia transmission device, the system comprising a processor and a memory, characterized in that: The memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method for mechanical resonance suppression of a digital servo system based on automatic frequency deviation correction as described in any one of claims 1 to 7.

9. A dual-inertia transmission device according to claim 8, characterized in that: Based on the classic dual-inertia servo system model, the mechanical resonance model of the digital servo system is improved, the current loop is equivalent to a hysteresis loop, PI is used as a proportional-integral controller for the speed loop control of the servo system, and a zero-order holder is added to simulate the limiting link in the saturation nonlinear link of digital control.

10. A computer memory, characterized in that: A computer program executable by a computer processor is stored therein, and the computer program executes the method for suppressing mechanical resonance of a digital servo system based on automatic correction of frequency deviation as claimed in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Alternating current servo system and resonance control method thereof

    CN109412493A

  • Method for suppressing mechanical resonance of cutting machine servo system

    CN109713968A

  • Alternating current servo system mechanical resonance suppression method based on improved BFGS adaptive wave trap

    CN114400948A

  • Time-frequency cooperative disturbance suppression method and system for rope-driven flexible mechanical arm

    CN116038674A

Cited By

  • Multivariable self-adaptive belt deviation rectification control method and system based on intelligent visual perception

    CN119370519A